5w. mamas Immense mmc SG-iL mm mm {mm a? swam TB’EF’GRASSES Thesis 50? Hm Begum: of 29%;. D. MECHIGAN STATE MFFERSE'E‘Y Robert 13% Cax’t‘ow E972 LIBRARY Michigan State University This is to certify that the thesis entitled SOIL FACTORS INFLUENCING ARSENIC SOIL TESTS AND GROWTH OF SELECTED TURFGRASSES presented by Robert N. Carrow has been accepted towards fulfillment of the requirements for Ph.D deg-66in 8011 Science Majgr professor Date Vii/7% 0-7639 REMOTE STORAGE PLACE IN RETURN BOX to remove thus checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE 2/17 20:: Blue FORIV S/DateDueForms_20l7mdd - 99.5 ABSTRACT SOIL FACTORS INFLUENCING ARSENIC SOIL TESTS AND GROWTH OF SELECTED TURFGRASSES BY Robert N. Carrow Inorganic arsenicals, especially arsenates, are commonly used on turfgrass soils as herbicides, especially for selective control of Poa annua L. The soil factors which influence phosphorus chemistry are often believed to have similar effects on arsenic. However, research on the soil properties influencing soil arsenic activity and their relative importance is limited. Knowledge in this area would allow for more effective utilization of arsenic as a control for Poa annua L. One purpose of this study was to investigate the relationships between arsenic activity in soils and certain soil factors, particularly phOSphorus, soil pH, texture, and moisture. The influences of the soil prop- erties on arsenic activity was determined by arsenic soil test results and by the growth of selected turfgrasses. Robert N. Carrow A second objective was to determine the feasibility of developing a routine arsenic soil test. Arsenic,applied as calcium arsenate, reduced the growth of all turfgrasses used in these studies. However, Poa annua L. exhibited the least tolerance to arsenate compared to Penncross bentgrass, Merion Kentucky blue- grass, and Cohansey bentgrass. Soil arsenic levels (Bray P extractable) of 50-70 ppm reduced the growth of l Poa annua L. by approximately 50 percent. Very high levels of phosphorus (> 180 lb Bray Pl extractable P/acre) decreased arsenic toxicity on these grasses, but the reduction was limited. Arsenic toxicity on Poa annua L. was less affected by increasing phos- phorus levels than the other grasses. Phosphorus had no significant effect on the arsenic soil test when extracted by the Bray P procedure. In contrast, arsenic signifi- l cantly increased the molybdic acid phosphorus soil test. Soil reaction markedly influenced arsenic activity as indicated by turfgrass growth and arsenic soil test data. In general, increasing pH from the range of pH 4.28-5.15 to pH 7.14-7.82 decreased the degree of con- trol a given rate of arsenic exhibited on Poa annua L., and decreased Bray Pl extractable arsenic levels. The 2 Robert N. Carrow magnitude of the soil reaction-arsenic interaction was greater with respect to growth and arsenic soil test than for the phosphorus-arsenic relationship. As clay content of the soil increased arsenic activity generally decreased. However, higher levels of extractable aluminum appeared more closely correlated with reduced arsenic activity than clay content on the soils studied. The Bray Pl extractable arsenic fraction was well correlated with the degree of control achieved at a cer- tain rate of arsenic on a given soil. The correlation was lower when several soils were included due to the influence of other soil properties, such as pH and tex- ture, on arsenic activity. The NaHCO extraction did 3 not appear to be as reliable as the Bray Pl extraction for soil arsenic. Because of the many factors which can affect the arsenic-Poa annua L. relationship, the devel- Opment of a routine arsenic soil test is not yet con- sidered practical. SOIL FACTORS INFLUENCING ARSENIC SOIL TESTS AND GROWTH OF SELECTED TURFGRASSES BY Robert N. Carrow A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Science 1972 e_ ‘ a D ACKNOWLEDGMENTS The author wishes to express sincere appreciation to his major professor, Dr. Paul E. Rieke, for his gui- dance, interest, and encouragement during the course of these investigations. Special thanks are due Dr. B. G. Ellis for his helpful suggestions and advice. The author is grateful for the assistance of Dr. James Beard, Dr. R. J. Kunze, and Dr. C. J. Pollard. To his wife, Rose, he is thankful for her patience, encouragement, and aid during the course of this study. ii TABLE OF LIST OF TABLES. . . . . . LIST OF FIGURES . . . . . INTRODUCTION. . . . . . LITERATURE REVIEW . . . . Poa annua L. . . . . Arsenic . . . . . . . Chemistry of Arsenic. Arsenic and Plant Growth. . METHODS AND MATERIALS . . CONTENTS Field Survey-Arsenic Contaminated Greenhouse Experiment 1 . . Greenhouse Experiment Greenhouse Experiment Greenhouse Experiment Greenhouse Experiment Greenhouse Experiment Greenhouse Experiment 2 3 iii Page vi 14 20 42 53 53 54 57 59 61 62 64 68 TABLE OF CONTENTS (cont.) Growth Chamber Experiment 1 . . . Statistical Methods . . . . . . . Arsenic Determination . . . . . . Arsenic Extractants . . . . . . . Phosphorus Determination... . . . Iron Determination. . . . . . . . Aluminum Determination. . . . . . Organic Matter Determination. . . pH Determination. . . . . . . . . Texture Determination . . . . . . Units of Measurements . . . . . . RESULTS AND DISCUSSION. . . . . . . . Arsenic Determination . . . . . . Phosphorus Determination. . . . . Field Survey-Arsenic Contaminated Greenhouse Experiment 1 . . . . . Greenhouse Experiment 2 . . . . . Greenhouse Experiment 3 . . . . . Greenhouse Experiment 4 . . . . . Greenhouse Experiment 5 . . . . . Greenhouse Experiment 6 . . . . . iv Page 69 71 71 74 75 76 77 77 78 78 78 80 80 85 88 91 97 107 113 118 122 TABLE OF CONTENTS (cont.) Page Greenhouse Experiment 7 . . . . . . . . . . . . 147 Growth Chamber Experiment 1 . . . . . . . . . . 150 GENERAL DISCUSSION. . . . . . . . . . . . . . . . . 157 CONCLUSIONS 0 O O O O O O O O O O O O O O I O O O O 178 APPENDIX. . . . . . . . . . . . . . . . . . . . . . 183 BIBLIOGRAPHY. O O O O O O O O O O I O I O O O O O O 213 Table LIST OF TABLES Maximum and minimum arsenic levels of several surface soils . . . . . . . . . . Chemical and physical properties of soils used in Experiments 1, 2, 3, 4, 5, and 7. Chemical and physical properties of soils used in Experiment 6. . . . . . . . . . . Standard deviation of arsenic standard solutions, using Small and McCants procedure . . . . . . ... . . . . . . . . Standard deviation of arsenic in soil extracts, using Small and McCants procedure I O O O O O O O O O O O O O O 0 Comparison of arsenic analyses by distil- lation and atomic absorption methods. . . Arsenic and phosphorus levels on selected Michigan golf courses . . . . . . . . . . Arsenic, phosphorus and pH levels on Mid- western golf courses. Samples collected by Freeborg . . . . . . . . . . . . . . . Effects of arsenic and phosphorus on the germination and growth of Penncross bent- grass and on the arsenic soil test, phos- phorus soil test and soil reaction. Soil was not incubated after treatment appli- cation. Greenhouse Experiment 1 (Arsenic Rate-Phosphorus Rate) . . . . . . . . . . vi Page 17 55 65 81 82 86 89 90 92 LIST OF TABLES (cont.) Table 10. 11. 12. 13. 14. 15. 16. 17. 18. Poa annua L. clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate-Phosphorus Rate-Grass) . . . . . . . . . . . . . . . Penncross bentgrass. Clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate- Phosphorus Rate-Grass). . . . . . . . . . Cohansey bentgrass. Clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate- PhosPhorus Rate-Grass). . . . . . . . . Merion Kentucky bluegrass. Clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate- Phosphorus Rate-Grass). . . . . . . . . Arsenic soil test, phosphorus soil test, and pH data for Poa annua L. in Greenhouse Experiment 2 (Arsenic Rate—Phosphorus Rate-Grass) . . . . . . . . . . . . . . . . Clipping weights of Poa annua L. for Greenhouse Experiment 3 (High Phosphorus- Arsenic Rate—Grass o o o o o o o o o o o o o Clipping weights of Penncross bentgrass for Greenhouse Experiment 3 (High Phosphorus- Arsenic Rate-GraSS) o o o o o o o o o o o o Clipping weights of Merion Kentucky blue- grass for Greenhouse Experiment 3 (High _Ph03phorus-Arsenic Rate-Grass). . . . . . . Arsenic and phosphorus soil test data for Greenhouse Experiment 3 (High Phosphorus- Arsenic Rate-Grass) . . . . . . . . . . . . vii Page 99 100 101 102 103 108 109 110 111 LIST OF TABLES (cont.) Table Page 19. Clipping weights, arsenic soil tests, and phosphorus soil tests for Greenhouse Experiment 4 (Arsenic Rate-Arsenic Form- Phosphorus Rate). . . . . . . . . . . . . . 115 20. Clipping weights, germination counts, arsenic soil tests, phosphorus soil tests, and soil pH levels for Greenhouse Experiment 5 (Arsenic Rate-Phosphorus Rate-pH) . . . . . 120 21. Loamy sand. Clipping weight and arsenic soil test data for Greenhouse Experiment 6 (Texture-pH-Arsenic Rate) . . . . . . . . . 124 22. Sandy loam. Clipping weights and arsenic soil test data for Greenhouse Experiment 6 (TEXture-pH-Arsenic Rate) 0 o o o o o o o o 125 23. Loam. Clipping weight and arsenic soil test data for Greenhouse Experiment 6 (Texture- pH-Arsenic Rate) 0 o o o o o o o o o o o o o 126 24. Silty Clay Loam. Clipping weight and arsenic soil test data for Greenhouse Experiment 6 (Texture—pH-Arsenic Rate) . . . . . . . . . 127 25. Peat. Clipping weight and arsenic soil test data for Greenhouse Experiment 6 (Texture- pH-Arsenic Rate) 0 O O O O O O O O I O I O 0 128 26. Clipping weight and arsenic soil test data for Greenhouse Experiment 7 (Arsenic Rate- Arsenic Form-Time of Application) . . . . . 148 27. Clipping weights for Growth Chamber Experiment 1 (Arsenic Rate-H20 Rate). . . . 152 28. water consumption and arsenic soil tests for Growth Chamber Experiment 1 (Arsenic Rate-H20 Rate) 0 o o o o o o o o o o o o o o 153 viii LIST OF TABLES (cont.) Appendix Page 29. Analyses of variance for data in Table 9, Greenhouse Experiment 1 . . . . . . . . . . 183 30. Analyses of variance for data in Tables 10, ll, 12, 13, and 14 of Greenhouse Exper- iment 2 . . . . . . . . . . . . . . . . . . 185 31. Analyses of variance for data in Tables 15, 16, 17, and 18 of Greenhouse Experiment 3 . 188 32. Analyses of variance for Table 19 of Greenhouse Experiment 4 . . . . . . . . . . 191 33. Analyses of variance for data in Table 20, Greenhouse Experiment 5 . . . . . . . . . . 194 34. Analyses of variance for data in Tables 21, 22, 23, and 24 of Greenhouse Experiment 6 . 196 35. Correlation coefficients (r) between Bray P1 extractable arsenic, NaHCO3 extractable arsenic, soils, and clipping yields by date. Greenhouse Experiment 6 (Texture- pH-Arsenic Rate). . . . . . . . . . . . . . 200 36. Analyses of variance for data in Table 26. Greenhouse Experiment 7 . . . . . . . . . . 201 37. Analyses of variance for data in Growth Chamber Experiment 1. . . . . . . . . . . . 205 ix Figure LIST OF FIGURES Page Arsenic standard curve for Small and McCants procedure . . . . . . . . . . . . . 208 Total clipping weights versus applied TCA on Poa annua L. (Greenhouse Experiment 2) . 209 Total clipping weights versus applied TCA on Penncross bentgrass (Greenhouse Experiment 2) . . . . . . . . . . . . . . . 210 Total clipping weights versus applied TCA on Cohansey bentgrass (Greenhouse Experiment 2) . . . . . . . . . . . . . . . 211 Total clipping weights versus applied TCA on Merion Kentucky bluegrass (Greenhouse Experiment 2) . . . . . . . . . . . . . . . 212 INTRODUCTION Rapid advancement in science in the past decades has resulted in more leisure time for the average worker than ever before. Much of this leisure time is spent in sports activities. Golf is one such sport which has en- joyed a surge of popularity. Not only are more golf courses being built, but existing courses are experiencing increased play. As golf becomes a more significant ac— tivity in their lives, many golfers expect a higher degree of sophistication in the courses they play. This combination of increased play and increased quality demanded by the average golfer has presented prob- lems to the golf course superintendent. Several pests have become predominant in response to the attempt to grow a turfgrass monoculture under highly unnatural conditions. One of the major pests on golf courses today is Poa annua L., also called annual bluegrass. Poa annua L. will persist and thrive under low cut, irrigated, and high maintenance conditions and can form a very satisfactory turf. However, under many stress conditions, particularly 1 high temperature, annual bluegrass is apt to be severely thinned. Also,Poa annua L. can produce seed heads even at a very low cutting height which are unsightly and provide a less desirable putting surface on greens. Approaches to the problem of annual bluegrass infestations are (1) eradication or control, and (2) main- tenance of Poa annua L. as a turfgrass. The latter has proved unsatisfactory at our present level of knowledge. Controls have been attempted by several means: 1) mechan- ical; 2) competition; 3) biological; and 4) chemical. None of these procedures, except chemical control, has provided any degree of reasonable success. Both preemer- gence and postemergence chemical treatments have been developed. Arsenates, especially calcium arsenate, are common preemergence chemical controls and have been widely used. The success of arsenates has been variable, with superin- tendents following similar programs experiencing widely different results. Much of this variability can be attri- buted to a lack of understanding of the reactions of arsenic in soils. The objectives of the research reported in this thesis were: 1) to study the effects of calcium arsenate on the germination and growth of selected turfgrasses, particularly Pga_annga L., 2) to study the relationships between the effectiveness of arsenic and certain soil properties, namely, phOSphorus, soil reaction, and soil texture, and 3) to determine the feasibility of developing a routine arsenic soil test, which would allow for more efficient utilization of arsenates. LITERATURE REVIEW P03 annua L. Characteristics Poa annua L. is often called annual bluegrass in the United States. In Australia and New Zealand it is termed wintergrass, while in England the term annual meadowgrass applies. Other regional names are speargrass, walkgrass, Junegrass, and Suffolkgrass (40). The scien- tific name is Poa annua L., which was designated by Carl Linnaeus in 1753. Annual bluegrass originated in Europe but is now found in all parts of the world under irrigated conditions (14, 40). Beard (14) gives the following description of Poa annua L.: Vernation folded; sheaths distinctly compressed, glabrous, whitish at base, keeled, split with overlapping, hyaline margins; ligule membranous, 1-3mm long, thin, white, acute, entire; collar conspicuous, medium broad, glabrous, divided by the midrib; auricles absent; blades V-shaped, 2-3mm wide, usually light green, glabrous, soft, boat-shaped apex, transparent lines on each side of midrib, parallel sided, flexuous transversely, margins slightly scabrous and 4 hyaline; stems flat, erect to spreading some- times rooting at the nodes and forming stolons; inflorescence a small, pyramidal, open panicle, branches few and spreading. While Poa annua L. is often believed to be an annual, recent evidence demonstrates considerable varia— tion due to a differential response of the species, within its genetic capability, to the environment. The products of such responses are termed ecotypes, strains, ecological races, ecospecies, or biotypes (13, 40, 118). Beard (13) reported many strains of Poa annua L. ranging from annual to perennial, prolific seed producers to mainly vegetatively-propagated ecotypes, and creeping to bunch strains. The bunch ecotypes tended to be strong seed producers and behave as annuals. The creeping strains were more prostrate, had little seed production, and gen- erally behaved as perennials. Youngner (118) noted ecotype selection due to the influence of environment on Poa annua L. Under turf condi- tions, he distinguished three ecotypes. The field type was annual, bunchy, upright, and a prolific seed producer. The putting green type was prostrate, perennial, spread vegetatively, and produced few seeds. The intermediate type possessed pr0perties between the above extremes. Gibeault (40) studied the characteristics of both annual and perennial forms of Poa annua L. with regard to blade length and width, ligule length, culm length, in- florescence characteristics, primary tiller number, shoot and root dry weight, and seminal root number. The annual strains had lower leaf and node numbers, lower adventitious root numbers, lower secondary tiller numbers, a greater percentage of flowering tillers, and an enhanced repro- ductive maturity rate. A post harvest dormancy was ex- hibited by annual strains but not by perennial strains. Gibeault (40) also studied the distribution of annual and perennial strains in the Oregon and Western Washington area. Samples were collected from three dif- ferent locations at 32 sites: golf course greens which received ample irrigation during the dry periods of summer; golf course fairways or lawn areas that were moderately irrigated; and golf course roughs that received no irri- gation. Of 65 samples, over 50 percent exhibited peren- nial characteristics. Both types were evenly distributed over the sampling area. The perennial types dominated the irrigated areas, while annual types were dominant in the unirrigated areas. Adaptation and Competitiveness Poa annua L. tends to invade areas where conditions are unfavorable for other more desirable species. Many factors affect the ability of Poa annua L. to persist and compete in a turf stand. Annual bluegrass grows best under fertile condi- tions. Beard (l3, 14) stated that high nitrogen levels of 0.5 to 1.0 lbs N per 1000 ft2 per month during the growing season were necessary for Optimum growth. High phosphorus favored Poa annua L. Juska and Hanson (51, 53) found Poa annua L. grew best when N, P, and K were added together. Under these conditions growth of tOps and roots were optimal, while N and P toqether resulted in increased topgrowth and re- duced root growth. Sprague and Burton (100) conducted several fer- tility studies using different ratios and carriers on Poa annua L. Low fertility levels allowed Poa annua L. invasion of most bentgrass turfs, while high nitrogen pre- vented encroachment. Only very aggressive creeping bent- grasses, such as Agrostis palustris Huds. MetrOpolitan and Agrostis palustris Huds. Washington, resisted invasion. Acid forming fertilizers tended to prevent Poa annua L. invasion, while slow release organic nitrogen sources gave the opposite response. Looking at the effects of N, P, and K on the growth of Poa annua L. the above investigators found nitrogen produced the largest increase in clipping yields. Little response to phosphorus was observed on a soil high in natural phosphorus, but significant growth response occurred on a soil of medium phosphorus availability. P and K together stimulated seedhead production. Soil reaction has been reported to affect the growth of Poa annua L. with excessively acid conditions causing a reduction in plant Vigor. Beard (l3, 14) noted that growth is best in soils having a pH range of 5.5 to 6.5. Juska and Hanson (53) observed significant pH effects. On a loamy sand at pH 4.5 and limed to pH 6.5, oven dry weights of clippings, crowns, and roots more than doubled at the pH 6.5 level. Seedhead counts were much higher at the higher pH. In the same study using a silt loam soil, no significant differences were observed due to pH. The authors contributed the lack of response to a higher nutrient availability in the silt loam. Sprague and Burton (100) found that liming a heavy loam soil from pH 5.3 to 6.3 produced greater yields in a greenhouse experiment. In a sand culture study Poa annua L. was kept at pH levels of 5.0 and 6.5. The highest growth occurred at pH 6.5. In a Colonial bentgrass (Agrostis tenuis)-Poa annua L. turf, acid forming fertilizers reduced Poa annua L. encroachment. Another factor affecting the growth and competi- tiveness of annual bluegrass is soil compaction. Bartlett and Troll (7) reported that Pga annua L. is often found in compacted areas. Beard (13, 14) noted the same trend and found annual bluegrass to have good tolerance to compac- tion relative to other turfgrasses. Youngner (ll8) stated that soil compaction restricted oxygen exchange in soil. Poa annua L. can survive under low oxygen diffusion rates and is more competitive relative to other turfgrasses under compacted conditions. Moisture influences Poa annua L. growth (7, 58, 100, 118). Beard (13, 14) reported that annual bluegrass had a high relative wilting tendency, poor relative drought resistance, and fair relative submersion tolerance compared to other common turfgrasses. Prolonged periods of water- logged soil conditions cannot be tolerated by Poa annua L. 10 Frequent irrigation without producing waterlogged soil conditions promotes Poa annua L. growth. Temperature is another factor which influences the growth and competitive ability of annual bluegrass. Beard (9, 10, 11, 12, 13, 14, 15) discussed the effects of both high and low temperature stress. Temperatures of 40 to 41.1 C were sufficient to kill Poa annua L. even under moist conditions. With moisture stress, kill can occur as low as 37.8 C. Bentgrasses, Kentucky bluegrasses, and bermudagrasses were much more heat tolerant. Low temper- ature tolerance of annual bluegrass was 2.8 to 5.6 C higher than for bentgrasses or Kentucky bluegrasses. Poa annua L. was found to have reduced tolerance to ice cover compared to other common turfgrasses. Optimum shoot growth occurred at 15.6 to 21.1 C, while optimum root growth was at 10 to 15.6 C. Bartlett and Troll (7) re- ported that cool night temperatures allowed Poa annua L. to withstand periods of warm weather better than warm nights and warm days. Sprague and Burton (100) stated that death of Poa annua L. was hastened by hot, humid weather, especially in conjunction with unfavorable soil conditions. Germination of annual bluegrass is also re- ported to be affected by temperature (13, 24, 40). 11 Light is another important environmental factor which can affect Poa annua L. Sprague and Burton (100) found that full sunlight in July and August was less favorable to annual bluegrass compared to continuous light shade. After mid—August full sunlight produced the best growth. Seedhead production occurred much quicker on plants exposed to full sunlight compared to shaded plants or plants receiving only S-hours of sunlight. Beard (13) found Poa annua L. to exhibit good shade tolerance. Youngner (118) reported that light necessary for seedling development may be a limiting factor for Poa annua L. under some circumstances. Poa annua L. demon- strated superior growth at low cutting heights compared to higher cutting heights in a bermudagrass (Cynodon dactylon) turf. Gibeault (40) noted that light was essen- tial for optimum germination and high light intensity after germination produced maximum growth. Closely associated with the response of Poa annua L. to light is its response to cutting height (7, 118). Beard (13, 14) reported that annual bluegrass was most competitive at cutting heights of 0.7 inch or less. It can persist and produce seedheads down to 0.25 inch. 12 Poa annua L. is susceptible to several diseases. Beard (l3, l4) and Bartlett and Troll (7) reported sus- ceptibility to Sclerotinia homeocarpa, Fusarium nivale, Typhula spp., Rhizoctonia solani, Pythium spp., Fusarium blight, and Helminthosporium vagans. Other factors can also influence the competitive- ness of Poa annua L. Beard (13) reported annual bluegrass to have poor relative smog tolerance and poor relative wear tolerance. The insect Hyperodes weevil can cause serious damage to annual bluegrass in some areas (14). Control The inherent lack of stress tolerance by Pga 22222.L° has caused it to be considered the major weed problem on golf courses and other intensively managed ir- rigated turfs in many regions of the United States. In various attempts to control 223.35233 L., several different approaches have been tried. Mechanical removal with a cup cutter or similar device has been used (40). This method is successful but is not feasible for extensive areas. 13 Maintaining the turf so that the more desirable grasses have a competitive advantage is another approach (7, 13, 93, 100). Errors in cultural practices, diseases, insects, nematodes, injury due to man, or environmental stress can result in the failure of this approach. Even though this method by itself may not give adequate con- trol, it is an essential part of any control program. Biological control with the Hyperodes weevil may work in some areas on a limited basis but it is not yet applicable to widespread utilization (14). Chemical control methods are the most widely used (7, 26, 36, 39, 50, 51, 52, 100). Of the various herbi- cides used on Poa annua L., the arsenicals, especially calcium arsenate and lead arsenate, have been most often employed. The success of arsenicals has been variable. The influences of soil texture, phosphorus, iron, aluminum, soil reaction, moisture, organic matter, time of arsenic application, maintenance practices, and environmental stresses on arsenic toxicity to Poa annua L. have been reported as possible reasons for erratic results. 14 Arsenic Occurrence Minerals of arsenic are widely distributed in nature. Arsenopyrite, FeAsS, is the most important min- eral of arsenic. Other arsenic minerals containing sulfur are orpiment, As S 2 3, and realgar, As S Arsenate forms 2 2° of arsenic often contain the (A304) anion and form com- pounds similar to phosphates which contain the (P04) anion. Isomorphic substitution of (A504) for (P04) can occur in nature, for example, the minerals minetite, Pb5((Cl) (A504)3) and apatite, Ca5((F,Cl,OH) (AsO ). 4)3 Arsenic occurs in nature in the elemental form and as arsenic trioxide deposits. Uncontaminated Soils Arsenic is a natural constituent of soils and is universally present in soils. Williams and Whetstone (114) examined a number of divergent soil types in the United States and Mexico for arsenic levels. They determined the arsenic content of the various soil profiles and stnniiEd . . . ' l the relationships between arsenic content, parent materla ' 15 soil prOperties, and climatic conditions. All soils examined contained arsenic ranging from 0.3 to 40 ppm. Fifty percent of the soils contained 5 to 10 ppm, while 30 percent had lower levels and 20 percent were higher. There was no apparent trend in arsenic distribution within the profiles. Parent material and climatic conditions were not clearly related to arsenic content. Jones and Hatch (48) sampled 10 soils in the fruit growing regions of Oregon for arsenic content and distri- bution within the profiles. The soils contained 2.8 to 14.1 ppm arsenic and the levels tended to be uniform with depth. Olson, et a1. (78) conducted a similar study on 10 South Dakota soils and found arsenic levels of 7.1 to 18.4 ppm and uniform contents to a depth of 36 inches. Vandecaveye et al. (108) examined two Washington soils for total arsenic and found 3.0 to 8.0 ppm. Twenty-four Ontario soils investigated by Miles (70) were found to contain 1.1 to 26.6 ppm arsenic with the majority less than 8.0 ppm. Vinogradov (111) conducted a survey of 53 Russian soils for arsenic. The arsenic contents averaged 0.1 to 1.0 ppm. 16 Contaminated Soils Arsenic levels in soils which have received arsenic can be much higher than for similar soils which have not received arsenic. Table 1 gives the maximum and minimum total arsenic levels as determined by several investigators on a wide variety of surface soils, which have received arsenic. While uncontaminated soils generally contain less than 10 ppm total arsenic, contaminated soils can have much higher levels. Immobility of arsenic is demonstrated by the lack of substantial movement of arsenic down the profile even in soils which have received high arsenic rates. Jones and Hatch (48) determined total arsenic at different depths in the soil profile on both contaminated and uncontaminated soils. The soils which had received no arsenic had uniform arsenic levels throughout the profiles with 14.1 ppm arsenic the maximum. On similar contaminated soils, the surface 8 inches had arsenic levels of 17.8 to 440.8 ppm, the 8-16 inch depth had levels of 5.2 to 21.8 ppm, and the 16-24 inch depth had levels of 2.1 to 15.6 ppm. Williams and Whetstone (114) noted little move- ment of arsenic down the profile. On a Sassafras sandy l7 .. o.H v o.mmmm mm Amaav cowaooz mmmuo 23mg o.omH 0.0mm 0H Aeaav meoumumns .. o.oo o.oem o cam memHHHHz eumnouo o.me o.me m Amoav scheme can Hmcnom .m>m>mowpcm> manommm o.mm o.meH e Incas umnumm one commeoee .. H.H o.H~H mm Lone mafia: eumeouo «.mH m.oss Hm Away nouns can mmcoe enmeouo o.~ o.moa om lass mm>mmuo eumeouo m.m e.ema mm Away eHoEmfleo ecm monmfim 30A nmwm mmHmEmm mmD pcmq umnEsz condom Afidmv owcmmud .mHHOm convusm Hmum>mm mo mam>ma vacmmum ESEAGHE cam Eseflxmzlr.a mqmda 18 loam the surface 3 inches contained 270 ppm arsenic, while the 5-8 inch zone contained only 4 ppm arsenic. Vandecaveye et al. (108) observed similar trends on orchard soils both with total and 0.1 N ammonium acetate extractable arsenic. Freeborg (39) did not determine total arsenic but analyzed for the Bray Pl extractable fraction with depth on several golf courses. Little movement of arsenic was observed. Use of Arsenicals Organic and inorganic arsenicals have been employed in agriculture for many years. Inorganic arsenicals in- clude calcium arsenate, sodium arsenite, arsenic trioxide, lead arsenate and Paris green. These compounds have been used as insecticides, soil sterilants, desiccants, silvi- cides, and herbicides. Organic arsenicals include mono- ammonium methanearsonate, methanearsonic acid, disodium methanearsonate, dimethyl calcium propyl arsonate, calcium methyl arsonate, and arsinic acid. Current usage of these substances includes cotton, orchards, silviculture, and turf. 19 The predominant use of arsenicals in turf has been for control of crabgrass (Digitaria sanguinalis) and Poa annua L. Monteith and Bengtson (73) noted that prior to 1933 arsenicals had been used on fairways, putting greens, and lawn turfs for selective control of clover (Trifolium repans), pennywort (Hydrocotyle sibthorpioides), ground ivy (Glechoma hederacea), Galium spp., knotweed (Polygonum spp.), chickweeds (Cerastium vulgatum and Stellaria media), and heal-all (Prunella vulgaris). They reported on re- search conducted at the Arlington Turf Garden between 1934 and 1938, which showed arsenicals effectively controlled crabgrass and that Poa annua L. was sensitive to arsenic. Welton and Carroll (112) conducted extensive research on crabgrass control with arsenicals during the 1930-1936 period. Sprague and Burton (100) found lead arsenate con- trolled Poa annua L. and crabgrass in a study conducted between 1930 and 1934. Daniel (26, 27) reported that prior to 1944 golf superintendents used lead arsenate for insect control. After 1944 many superintendents began using other insec- ticides in place of lead arsenate. By 1950 it was noted that courses which continued to use arsenate had less infestation of Poa annua L. than courses applying other 20 insecticides. This observation lead to a renewed interest in the arsenicals for use on Poa annua L. In the past two decades several researchers have investigated various aspects of the arsenic-Poa annua L. relationship (26, 27, 36, 39, 50, 51, 52). Soil properties, environmental factors, and maintenance practices have been reported to alter this relationship. As a result the efficient use of arsenicals for Poa annua L. control presents a complex problem involving many possible interactions. Chemistry of Arsenic Arsenic has often been assumed to undergo reac- tions similar to phosphorus. While many of the soil factors affecting arsenic and phosphorus are recognized, and are similar,litt1e is known about the actual reactions of arsenic in soils. The relationships between arsenic and various soil components are discussed in the following sections. 21 Soil Texture and Clay Heavier-textured soils have generally been observed to fix or inactivate greater quantities of arsenic than lighter, sandy soils. Johnson and Hiltbold (47) found that 85% or more of applied arsenic resided in the clay fraction of a sandy loam soil. Dickens and Hiltbold (32) reported that 10 times more DSMA was absorbed by the clay fraction than the sand or silt fractions of an Augusta silt loam. Similar observations have been reported by numerous investigators (19, 22, 25, 30, 32, 33, 45, 56, 65, 84, 89, 95, 98). The type of clay affects the degree of arsenic retention. Dickens and Hiltbold (32) found kaolinite to be a much more effective sorbent of DSMA arsenic than vermiculite or montmorillonite. Kaolinite has many exposed hydroxyl groups which may be involved in arsenic absorb- tion. Sieling (95) and Swenson et al. (105) reported that ballmilled kaolin was effective in the sorption of ar- senate. Evidence presented in these papers suggested that hydrous alumina in kaolin was the active constituent. Boischot and Hebert (19) investigated the influ- ence of the type of cation present on the clay on the 22 degree of arsenic adsorption. Calcium—saturated clays absorbed much more arsenate than potassium-saturated clays. Iron Iron-arsenic interactions in soils have been demonstrated by many investigators. Albert (1, 2, 3), COOper et a1. (22, 23), and Dorman et al. (33) reported that plants grown on soils high in iron tended to be affected little by additions of arsenate. Dorman et a1. (33) determined the percent Fe of several soils and 203 found soils high in Fe203 exhibited little toxic effect due to arsenic. Application of iron to soils can reduce arsenic toxicity. COOper et a1. (22) found that iron sulphate greatly increased the yield of COWpeaS (yigna Spp.) on an arsenic toxic soil high in phosphorus. Vandecaveye et al. (108) applied FeSO to two soils containing toxic levels 4 of arsenic. Yields of barley (Hordeum vulgare) were greatly enhanced. Vandecaveye et al. (109) in one exper- iment found no response due to FeSO application, but in 4 another reported that FeSO4 and Fe2(SO4)3 substantially increased yields of barley and alfalfa (Medicago sativa) 23 on two soils. Kardos et a1. (55) applied FeSO4 and ob- served a marked reduction of water soluble arsenic. Misra and Tiwari (71) added Fe203 to soil and reduced the toxi- city of arsenic to several plants. Keaton and Kardos (56) investigated arsenic fixa— tion by iron compounds. They observed that 1 g of Fe203 fixed 2.5 mg arsenic. Oxidized iron forms were found to fix arsenic best as has been suggested by the data of Vandecaveye et al. (109). Swenson et al. (105) demonstrated arsenic fixation by hydrous iron oxide. Margulis and Bourniquel (65) also investigated arsenic fixation by iron, and the amount fixed was independent of soil pH. In studies on anion exchange reactions of soils, Dean and Rubins (30) noted a marked reduction of arsenate retention of several soils after re- moval of free iron oxides. However, the reduction in arsenate retention was not preportional to the quantities or iron removed. Several investigators have found iron associated with arsenic in soils. Johnson and Hiltbold (47) looked at arsenic in various fractions of a sandy loam soil which had received organic arsenicals. They found 8% of the arsenic associated with dithionite and citrate extractable 24 iron oxides. The 0.1 N NaOH fraction yielded 8% of the arsenic and was assumed to remove iron arsenates (Fe-As). Jacobs et a1. (45) conducted a similar analysis on three soils. They found arsenic sorption to increase with in- creasing free iron oxide content. Removal of oxalate extractable iron and citrate-dithionite-bicarbonate (CDB) extractable iron both reduced arsenic fixation. Most of the arsenic was associated with oxalate extractable iron which is amorphous iron. Between 8% and 27% of the ar- senic was extractable with 0.1 N NaOH. Woolson et al. (117) fractionated 34 arsenic contaminated and uncontam- inated soils. In most of the soils the Fe-As fraction was predominant. In a few soils low in reactive iron, the aluminum and calcium arsenate fractions dominated. Aluminum Aluminum has been demonstrated to react with arsenic to form relatively insoluble aluminum arsenate (Al-As) compounds. Martin et al. (68) studied the AsZOS-AlZOS-HZO system at 20 C and 60 C. They concluded that A1A804.2H20, A1(H2ASO4)3.5H were formed. Albert (1) observed that aluminum hydroxide O, and A1(H2AsO4).2H O 2 2 25 would adsorb arsenic, while Keaton and Kardos (56) noted that A1203 fixed arsenic. Aluminum fixation of arsenic in soils was observed by Margulis and Bourniquel (65) and re- ported to be independent of pH. Dorman et a1. (33) determined the %A1203 of sev— eral soils high in arsenic. Soils with the highest %A1203 exhibited the least toxic effects of arsenic. Sieling (95) reported arsenic sorption by hydrated alumina, ballmilled kaolin, and dehydrated A120 The 3. sorption by the ballmilled kaolin was attributed to par- tially hydrated alumina in the kaolin. Swenson et al. (105) also attributed arsenic sorption to alumina asso- ciated with kaolin and noted that arsenate can be precipi- tated by aluminum directly. Johnson and Hiltbold (47) fractionated the forms of arsenic in a sandy loam soil, which had received ar- senic. The neutral 0.5 N NH4F extraction removed 14% of the total soil arsenic and was assumed to represent Al-As. The final extraction was also neutral 0.5 N NH4F and re- moved an additional 18% of the total arsenic, believed to be occluded Al-As. Jacobs et a1. (45) investigated three soils. The ability of the soils to sorb arsenic was proportional to 26 their A1203 and Fe203 contents as determined by oxalate extraction and CDB extraction. The amorphous aluminum oxides and possibly amorphous aluminosilicate components were believed to be involved in the arsenic fixation. Between 60-70% of the total arsenic was found in the NH4F extractable fraction and was presumed to be Al—As. Woolson et al. (117) fractionated 34 soils which had received arsenic previously. The Fe-As fraction pre- dominated in most of the soils but Al-As was dominant in soils high in aluminum and low in oxalate extractable iron. Soils high in NaOH extractable aluminum were found to be less phytotoxic after application of arsenic than soils low in aluminum. Calcium Arsenic can react with calcium or calcium com— pounds resulting in a decrease in the activity of arsenic. Several investigators have looked at the CaO-Aszos-HZO system under different conditions. Smith (99) concen- trated on the acid region at 35 C. Dicalcium orthoarsenate monohydrate (CaHAsO4.H20) and monocalcium orthoarsenate (CaH4(AsO4)2) were found to be stable under these 27 conditions. Pearce and Norton (82) studied the region of this system more basic than dicalcium orthoarsenate mono— hydrate and at 90 C. Four compounds were isolated: dicalcium arsenate (CaHAsO4), pentacalcium arsenate (Ca5H2(ASO4)4), tricalcium arsenate (Ca3(AsO4)2), and basic calcium arsenate ((Ca3(AsO4)2)3.Ca(OH)2). In a similar study at 35 C, Pearce and Avens (81) identified dicalcium arsenate, pentacalcium arsenate, and tricalcium arsenate. Margulis and Gane (66) demonstrated the forma- tion of Ca2H2(AsO3)2, followed by transformation into Ca3(AsO3)2. The ability of CaCO or Ca(OH)2 to fix arsenic has 3 been demonstrated by several researchers. COOper et a1. (22) and COOper et a1. (23) reported a reduction in the toxic effects of calcium arsenate after application of limestone to soils which were acid. Vandecaveye et al. (108) applied CaCO3 and CaSO4 to two arsenic toxic soils. Substantial beneficial effects were noted on one soil but no consistent effects on the other. Welton and Carroll (112) noted reduced effectiveness of lead arsenate on grass after liming. Misra and Tiwari (71) reported similar results on other plants. The %Ca0 present in several arsenic toxic soils was correlated to the effectiveness 28 of the arsenic by Dorman et a1. (33). The soils with a high %CaO exhibited reduced toxicity. Boishot and Hebert (l9) conducted an extensive study of the AsZOS-CaCO3 system in solutions. The fixa- tion of arsenic was directly proportional to the amount of arsenic present. Increasing CaCO fixed increasing 3 quantities of As up to 94% of the total present. Most 205 of the fixation occurred within 1 hour. The nature of the reaction was adsorption, not combination. They also observed that a calcium saturated clay fixed more arsenic than a potassium saturated clay. Margulis and Bourniquel (65) found CaCO to be quite effective in fixing arsenic 3 in both solutions and soils. Several investigators have observed the presence of calcium arsenates (Ca-As) upon arsenic fractionation of soils. Johnson and Hiltbold (47) found substantial arsenic extracted by l N NH Cl which is assumed to remove 4 exchangeable calcium. Also the 0.5 N H2504 extract con- tained about 8% of the total arsenic in the soil investi- gated, and was believed to be Ca-As. Jacobs et a1. (45) found from 3.5 to 29% of the total arsenic extracted by 1 N NH4C1 and 0.9 to 5.0% in the 0.5 N H2804 fraction in three soils. Woolson et al. (117) fractionated 34 soils. 29 Soils low in oxalate extractable iron had large amounts of Ca-As. Soils with high exchangeable calcium and high oxalate extractable iron did not necessarily contain large quantities of Ca-As. Soil Reaction Soil reaction has been found to influence arsenic activity in soils. COOper et a1. (23) reported an increase in the toxicity of calcium arsenate at high soil acidity, but no specific crOps were mentioned. Kerr (58) recom— mended a pH of 6.0 as best for control of Poa annua L. with arsenicals. He stated that at low pH or pH above 7.8 arsenicals are less effective. Calcium arsenate reduced plant counts and yields of Poa annua L. more at pH 4.8 than at pH 6.5 according to Juska and Hanson (52). Everett (38) also found increasing pH to reduce arsenic toxicity on bluegrasses. However, Freeborg (39) noted no significant correlation between soil reaction and degree of Poa annua L. control in a survey of several golf course soils. Boischot and Hebert (19) found that arsenate was fixed by clays much slower at acid pH than at neutral pH. 30 Maximum fixation in a potassium-saturated clay occurred at pH 6.8 to 7.2, while for a calcium-saturated clay re- tention was greatest between pH 6.5 to 8.7. They also demonstrated that arsenate fixation by humus increased with increasing pH between pH 7.0 and 8.9. Sieling (95) looked at arsenate sorption on ball- milled kaolin in solutions at different pH's. Acid pH resulted in much greater sorption than neutral pH. He attributed this response to the effect of pH on the hydrous alumina associated with kaolin. Soil reaction can affect the percent base satura- tion of a clay, reactive iron and aluminum contents, calcium content, and pH dependent exchange sites of clays, sesquioxides, and organic matter. Changes in these soil factors may result in altering the form and activity of arsenic in soils. Phosphorus Phosphorus and arsenic belong to Group V of the periodic table, each having five valance electrons (szp3) in their outer most energy level. Their radii and elec- tronegativities are similar and both elements commonly 31 exhibit valance states of +5, +3, or -3. The analogous prOperties of these elements have led to the assumption that they undergo similar reactions in soils. Very little research has been done to substantiate this belief. Sieling (95) found that in dilute solutions the sorption of arsenate by ballmilled kaolin and freshly pre- cipitated hydrous alumina was similar to phosphate, but in concentrated solutions the sorption patterns differed due to secondary effects. Arsenate was able to displace only a small percentage of the sorbed phosphate from ballmilled kaolin, while phosphate replaced most of the arsenate. Dean and Rubins (30), Deb and Datta (31), and Bastisse (8) observed that phOSphorus retention of soils was greater than arsenate retention. Generally, twice as much phosphorus was adsorbed as arsenate. Swenson et al. (105),Dean and Rubins (30), and Deb and Datta (31) noted that phosphate replaced most or all of the arsenate from soils as well as precipitated aluminum. Arsenate replaced only a small portion of the phosphate. Woolson (115) removed 77% of the arsenate from a soil by continuous leaching with phosphate. However, Misra and Tiwari (71) removed less than half of the ar- senate by phosphate leaching. 32 Reduction of phosphorus and arsenic retention on soils after removal of free iron oxides was demonstrated by Dean and Rubins (30). Jacobs et al. (45) made similar observations for arsenate. The importance of the As:P ratio has been demon- strated by several investigators. In nutrient solution studies, Hurd-Karrer (43) found that arsenic toxicity was a function of phosphate concentration. With a 1:1 ratio, regardless of the absolute concentration, a high degree of arsenic toxicity was observed on wheat (Triticum spp.). Ratios greater than 1:4 substantially reduced arsenic toxicity. Clements and Munson (21) observed that increas- ing phosphorus reduced arsenate absorption of bean (Phaseolus vulgaris L.), Sudan grass (Sorghum vulgare var. sudanese), and tomato plants (Lyc0persicon spp.). Phos- phorus had little effect on arsenite absorption. Kardos et a1. (55) also noted little effect of phosphorus on arsenite toxicity, but a marked reduction in arsenate toxicity. Several other investigators have reported a similar influence of phosphorus on arsenate in nutrient culture studies (38, 87, 91). Evidence from soil studies also point out the influence of the As:P ratio on plant growth. Measuring 33 Truog (0.002 N H 804) reagent soluble arsenic and phos- 2 phorus, Kardos et a1. (55) found As:P ratios greater than 1:1.3 reduced toxicity on several soils. Decreased arsenic toxicity on soils after addition of phosphorus has been reported by a number of researchers (16, 26, 27, 39, 43, 51). However, substantial data demonstrates that adding phosphorus may caused enhanced arsenic toxicity in some cases (3, 22, 23, 94, 98, 115). Woolson (115) suggested that when phosphorus was added to soils low in reactive iron the arsenate may be replaced from iron arsenates, and result in higher soluble arsenate levels. Organic Matter The relationship between arsenic and organic matter in soil is unclear. Freeborg (39) studied arsenic fixation of four soils: Houghton muck, KoKomo silty clay loam, Crosby silt loam, and a greenhouse organic sand mix. Bray P extractable arsenic was least for the Houghton 1 muck and greatest for the greenhouse mixture. The Houghton muck contained 42.0% clay and 28.5 % organic matter, while the greenhouse mix was 3.1% clay and 8.3% organic matter. 34 Even though fixation was highest on the muck, the organic matter may not have been the active agent. Freeborg (39) also noted that percent organic matter was not signifi- cantly correlated to Poa annua L. inhibition by arsenate in a survey of several golf course soils. Boischot and Hebert (19) investigated arsenic fixa- tion by humus. Increasing pH resulted in enhanced arsenic fixation. More arsenic was fixed after 3 days than after 1/2 hour, but only about twice as much. Increasing the quantity of humus increased the percent arsenic fixed. Under no conditions was the percent of arsenic fixed greater than 18.0% of that added. The authors concluded that arsenic retention by humus was minor compared to clay or limestone. Jacobs et a1. (45) stated that organic matter was not significantly related to arsenate retention, as mea- sured by Bray P1 or NH4OAc extraction, on two soils vary- ing from 0.2 to 21.1% organic carbon. Johnson and Hilt- bold (47) fractionated arsenic from a soil after additions of various organic arsenicals. Organic phosphorus and arsenic were determined by H202 oxidation and NH4F extrac- tion. While 68% of the phosphorus was in the organic fraction, no arsenic was detected in the extract. 35 Moisture Soil moisture content has been reported to influ- ence the degree of arsenic toxicity exhibited by a soil. Kerr (58) recommends draining low areas when applying arsenicals for Poa annua L. control because wet soils enhance arsenic toxicity. In studies conducted on rice soils, Epps and Sturgis (37) found that the solubility of commercial and natural mineral arsenicals increased upon flooding. Release of arsine gas from flooded soils was shown, which indicates reduction of arsenates occurred upon flooding. Arsenite has been demonstrated to be much more toxic to plants than arsenate (4, 21, 55). Leaf and Smith (59) observed the effectiveness of arsenic trioxide under a wide variety of conditions. They found toxicity to be greater on wet soils compared to well- drained soils. Arnott and Leaf (6) grew Monterey pine (Pings radiata D. Don.) at several A5203 levels and under normal and excessively wet conditions. Excessive moisture re- duced the amount of As 0 required to obtain a given level 2 3 of phytotoxicity by about 25% compared to the normal 36 moisture level. However, water and alcohol extractable arsenic were less on the excessively watered treatments. Rosenfels and Crafts (89) determined the quantity of A3203 required to obtain a given degree of toxicity on oats under field capacity and when watered only after signs of wilting appeared. No differences due to water treat- ments were observed. Other Factors Affecting Arsenic In addition to clay content, clay type, iron, aluminum, calcium, soil reaction, phosphorus, organic matter, and moisture, other factors have been reported to alter arsenic reactions in soils. Temperature can alter the activity of arsenic. Albert (2) reported sol- uble arsenic from collodion bag diffusates to be 3 to 5 times higher from bags maintained at 40 C and 60 C than at 27 C. No difference in water soluble arsenic extracted from soils dried at 60 C, 105 C, and undried soils was found. However, Freeborg (39) found that as temperature increased from 15 C to 30 C Bray P1 extractable arsenic decreased as much as 44%. He contributed this decrease to enhanced arsenic fixation at higher temperatures. 37 Reduction of arsenic by microorganisms in soils was reported by Alexander (5). Trimethyl arsine, As(CH3)3, not arsine, AsH gas was indicated to be the reaction 3: product. Alexander also noted that biological oxidation of arsenite to arsenate in soils has been observed. This transformation followed logarithmic rates typical of bac- terial culture growth and was inhibited by enzyme poisons. Dickens and Hiltbold (32) presented evidence of adaptive microbial oxidation of disodium methanearsonate (DSMA) on a Norfolk loamy sand with organic matter added. Greaves (41) attempted to correlate water soluble arsenic to several soil factors. No correlation appeared between water soluble arsenic and total soluble salts, sodium chloride, sulfates, calcium carbonate, and organic nitrogen in the soil. Sodium carbonate and nitrate were slightly correlated. The above are factors which directly influence arsenic in the soil. There are additional factors which may not alter the chemistry of soil arsenic but are still important in any arsenical program used for control of £2§.EBBEE.L° The response of Pga_angua L. to arsenic has been reported to be affected by photOperiod, temperature, topdressing, and nutrient level (7, 26, 27, 39, 100). 38 Essentially anything which alters the physiological condi- tion or competitive ability of annual bluegrass can poten- tially change the Poa annua L.-arsenic relationship. Fixation and Movement Arsenic could be potentially harmful if present in free form in nature. Fortunately, soils generally possess the ability to retain large quantities of arsenic in rela- tively inactive forms. Leaching studies and field moni- toring of arsenic distribution in soils have demonstrated the fixation of arsenic by soils. Laboratory experiments with various soil components, such as iron, aluminum, cal- cium, and clay, have also indicated arsenic retention by various mechanisms. Arsenic movement studies using leaching columns have been conducted by several investigators (6, 25, 32, 101). Such studies have utilized a variety of soils, leaching rates, arsenic rates, arsenic forms, indicator plants, and/or arsenic detection procedures. Except in sandy soils at high arsenic and leaching rates, little movement was reported. Even under-the above conditions, movement beyond 20 inches was not indicated (25). 39 Welton and Carroll (112) applied a total of 7.8 lb arsenic per 1000 ft2 to a lawn area. Arsenic was found to a depth of 10 inches but with greatest concentrations in the 0-4 inch zone. Williams and Whetstone (114) found little movement of arsenic beyond the 0-3 inch layer in a Sassafras sandy loam and a Hagerstown silt loam after arsenic applications. Jones and Hatch (48) investigated arsenic movement in several orchard soils. Little move- ment beyond 16 inches was observed, while most of the arsenic was found in the 0-8 inch depth. In contrast, Miles (70) reported high, uniform, arsenic levels on an orchard soil to a depth of 6 inches. He did not indicate if cultivation had been practiced prior to sampling. Small and McCants (98) sampled three North Carolina soils with depth for arsenic after applying up to 48 lb arsenic per acre. No accumulation was found below 12 inches. Freeborg (39) determined the Bray Pl extractable arsenic in 22 golf course green, fairway, and rough areas of var- ious textures. In most cases arsenic was predominantly in the 0-2 inch depth for heavier-textured soils. Lighter- textured soils had a more uniform distribution throughout the upper 0-6 inch layer, but with the 0-2 inch depth containing the most. 40 Several investigators have demonstrated increased fixation with time (21, 25, 95). Rosenfels and Crafts (89) studied the fixation of arsenic on 9 soils after application of 340 ppm As Fixation varied from 19.5% 203. to 91.0% after 18 hours to 72.4% to 99.9% after 7 weeks. Boischot and Hebert (19) found arsenic retention to in- crease with an increase in time of contact on both clay and CaCO3. On the clay, fixation was 30% complete after 1/2 hour, 42% after 5 hours, and 90% after 19 days. With the CaCO fixation was 96% complete at 1 hour. Bray P 3' l and NH4OAc extractable arsenic measured on six soils were shown to decrease with time by Jacobs et a1. (45). Equil- ibrium was reached after 1 to 6 months depending on the soil texture and arsenic added. Coarse-textured soils with low arsenic applications reached equilibrium first. Woolson (115) noted the changes in water soluble, aluminum, iron, and calcium arsenates with time after arsenic application on two soils. Percent water soluble arsenic decreased over time, with the greatest decrease at the low arsenic rates. Aluminum arsenate was\initially high but decreased with time while iron arsenate increased. Calcium arsenate remained constant after a very slight 41 increase. Juo and Ellis (49) reported a similar pathway for phosphorus. Arsenic fractionation studies have indicated some of the soil components involved in arsenic retention (45, 115, 116, 117). These include iron, aluminum, and calcium. The form in which arsenic is retained can influence its activity and toxicity potential. Woolson (115) and Woolson et al. (116) investigated the phytotoxicities of Na-, Fe—, A1—, and Ca-As compounds. Using corn (Egg mays) as an indicator crOp, they found the order of phytotoxi- city to be Na(H2AsO4)>Al(HZAsO4)3>Ca(H2AsO4)2>Fe(H2AsO4)3. Woolson reported Fe-As to be less soluble than Al-As or Ca-As. The solubility product constants and concentra- tions of arsenate at equilibrium were: -19 -5 Ca3(AsO4)2 — 6.8 x 10 (A304) - 9.1 x 10 -15 -8 A1(AsO4) = 1.6 x 10 (A504) = 1.3 x 10 Fe(AsO4) = 5.7 x 10‘21 (A304) = 7.6 x 10'11 This indicates that Fe-As is much less soluble than Ca-As or Al-As. The above experimental data in conjunction with previously mentioned reports of arsenic fixation by CaCO3, clays, and organic matter indicates that arsenic is 42 readily fixed in soils by soil components, and over time the arsenic becomes less toxic and less soluble. The limited data on arsenic fixation indicates that in acid soils hydrous iron and aluminum oxides are probably the main soil component responsible for arsenic fixation (30, 31, 45, 47, 65, 95, 105). The arsenate anion is probably exchanged for hydroxyls of the hydrous oxides. Fixation by hydroxyl exchange on the edges of 1:1 clay lattices is also probably involved in fixation (32, 65). In alkaline soils which contain CaCO retention 3 by absorption on the CaCO3 surface would be indicated to be important (19). Chemical combination of active calcium in the soil solution with arsenic is also possible (19, 66, 81, 82, 99). The formation of a arsenate-Ca-clay linkage may be possible but is likely: to be of minor importance (19). Arsenic and Plant Growth Arsenic is not recognized as an essential element for plant growth, but is known to have a toxic effect on plants. Enzymes which contain thiol groups or lipoic acid 43 are especially sensitive to arsenite (42). The converr sions of pyruvate to acetyl CoA and d-keto—glutarate to succinate in the Krebs cycle are important reactions which are inhibited by arsenite (42). Arsenate can uncouple substrate-linked phosphorylation associated with the oxidation of glyceraldhyde 3-phosphate to 1, 3-diphosphoglyceric acid in glycolysis. The overall oxidoreduction reaction occurs but heat instead of ATP is formed (42, 60). Arsenate can also uncouple oxidative phosphorylation from electron transfer in the electron transport-oxidative phosphorylation chain (42). Arsenic Concentration and Distribution Within Plants Arsenic concentrations of 0.02 to 0.20 ppm were found in pears (Pygus spp.) and apples (Malug spp.) grown on soil which had never received arsenic (48). Williams and Whetstone (114) analyzed the arsenic content of var- ious plants grown on uncontaminated soils. Concentrations varied from 0 to 10 ppm arsenic with the majority contain- ing less than 1.0 ppm. Arsenic contents of plants grown on contaminated soils can be considerably higher. Jones and Hatch (48) 44 found from 0.05 to 0.82 ppm arsenic in peeled pears and apples grown on soils which received arsenic. Williams and Whetstone (114) determined the arsenic content of .vegetation grown on contaminated soils and reported values of <1 to 83 ppm. Many reports of arsenic concentrations in many different plants are available in the literature (34, 46, 47, 48, 61, 63, 64, 69, 74, 78, 84, 90, 98, 104, 106, 107, 109, 114, 115). Liebig (61) has written an excellent review of this tOpic. He reported plant tissue values ranging from 0 to 1245 ppm arsenic. 'Several investigators have looked at the arsenic levels in grasses. Olson et a1. (78) collected tissue samples from grasses grown on uncontaminated soils. Western wheat grass (Agropyron smithii), feather grass (Stipa viridula), blue grama grass (Bouteloua gracilis), and needle grass (Stipa spartea) were analyzed for arsenic. Values ranged from 1.0 to 4.3 ppm arsenic. Machlis (63) grew sudan grass in solution cultures with arsenic rates of 0 to 10 ppm. Concentrations of arsenic from the 10 ppm arsenic solution based on dry weight and fresh weight (in parenthesis) of different plant parts were: leaves 87.4 (12.5) ppm, nodes 126.8 ppm, internodes 31.2 ppm, inflor- escences 4.5 ppm, tillers 52.8 (8.9) ppm, primary shoots 45 67.2 (14.6) ppm, plant tOps 56.9 (10.3) ppm arsenic. In the 10 ppm arsenic solution, yield was reduced by 78%. McBee et a1. (69) foliarly applied MSMA to Coastal bermuda- grass (Cynodon dactylon) at the rates of 0 to 5 lb arsenic per acre. The foliage was removed 7 days after treatment. Analysis of the regrowth was made 36 days after treatment and arsenic contents were 0.10 to 17.24 ppm arsenic on a dry weight basis. Duble et a1. (34) applied DSMA at var- ious rates to Coastal bermudagrass as a foliar application (4.5 kg DSMA/ha), soil application (17.9 kg DSMA/ha), and nutrient solution (12 ppm DSMA). The concentrations of arsenic found in the leaves and roots (in parenthesis) for each application method as determined 7 days after treatment were: foliage 68 (45), solution 38 (88), and soil 1.6 (15) ppm. As indicated by the above studies, distribution of arsenic within the plant can vary considerably. Liebig (61) surveyed the literature with respect to ar- senic concentrations of plants and its distribution within the plant and concluded that the tOpS of plants accumulate little arsenic, but roots can accumulate large amounts. Jacobs et a1. (46) reported little increase in arsenic content of potato leaves (Solanum tuberosum L., var 46 'Burbank') with increasing arsenic, while tuber contents increased markedly. Small and McCants (98) found a sim- ilar trend on tobacco (Nicotiana tabacum). Several investigators have looked at arsenic dis- tribution in grasses and grains. Machlis (63) grew Sudan grass at solution concentrations of 0 to 18 ppm arsenic. At the 8 ppm level in solution, the leaves contained 40 ppm (dry weight basis) and the roots 906 ppm arsenic. At the 18 ppm level in solution, the roots contained 1461 ppm, while leaf values varied widely due to non-functional roots. Vandecaveye et al. (108) measured the arsenic contents of tOps and roots of barley grown on two arsenic toxic soils. Concentrations in the teps were 7.6 and f 13.3 ppm arsenic, while root values were 599 and 1246 ppm. Rumburg et a1. (90) determined the activity of DMA-As76 and labeled sodium arsenite after application to the second leaf below the terminal leaf of crabgrass. Some basipetal movement was noted. Duble et a1. (34) looked at arsenic distribution within Coastal bermudagrass after foliar application, soil uptake, and solution uptake of DSMA. The foliar treatment demonstrated basipetal move- ment, with leaf and root concentrations of 45 and 68 ppm arsenic, respectively. Soil and solution treatments 47 showed acropetal movement, with leaf and root levels of 1.6 and 15 ppm arsenic, respectively, for soil and 38 and 88 ppm, respectively, for solution studies. Arsenic Effects on Seed Germination. f Seed germination of cotton (Gossypium Spp.), corn, and soybeans (Glycine £213) after calcium arsenate treat- ment on several soils was investigated by Dorman et a1. (33). No effects on germination was observed on any of the creps, regardless of the soil type until the 400 1b calcium arsenate per acre rate. Boischot and Hebert (19) found no inhibition of grain germination until the 148 mg A8205 level in solution culture. Stapledon (102) reported that 14 lb per 1000 ft2 of lead arsenate did not inhibit germination of seeded fescues (Festuca Spp.), bentgrasses, Poas, perennial rye- grass (Lolium perenna L.), or crested dogtail (Cynosurus cristatus). Naylor (77) investigated the response of Kentucky bluegrass (Poa pratensis L.) seeds to various arsenious acid rates in sand cultures. At rates of 0.6 and 1.5 lb arsenious acid per 1000 ftz, seed germination 48 was superior to the control pots, while at the 3 to 6 1b rates it was below the control. Juska (50) studied the effects of 85% tricalcium arsenate and 98% lead arsenate on germination of several grasses over a period of 5 months after application. In a greenhouse experiment, 350 and 520 1b CaAs per acre and 440, 870, 1740 lb PbAs per acre reduced the germination of Merion and annual bluegrasses but not seriously. In a field study, 520 1b of CaAs per acre reduced Pga_annga L. germination 100%, tall fescue 98%, bentgrass 87%, red fescue 34%, and Merion Kentucky bluegrass 20%. Lead arsenate at 1090 1b/A prevented 95% of the Egg annua L. from germinating but did not effect the other grasses. Crabgrass was also inhibited by CaAs and PbAs. Welton and Carroll (112) noted substantial reduc- tion in crabgrass plant counts due to CaAs and PbAs treat- ments. Stadtherr (101) reported very effective reduction of crabgrass germination after treatment with the arsen- ical PAX. Germination of Kentucky bluegrass and Highland bentgrass seeds was delayed 12 weeks at high rates of PAX but then germinated. Freeborg (39) investigated the effect of arsenic on Poa annua L. seed germination. Even 49 high rates of 48% tricalcium arsenate did not prevent germination in petri dishes or soil. The conflicting data on the effects of arsenicals on seed germination may be due to counting procedures. Some investigators counted only live plants on the count- ing date, while others counted all seeds that germinated, even if they did not live. 'Arsenic Effects on Plant Growth Reduction of crop growth or yield by arsenic has been reported by many investigators (6, 16, 18, 22, 23, 25, 33, 37, 46, 47, 55, 59, 64, 84, 89, 94, 96, 98, 104, 109, 110, 115, 117). Data from solution culture studies give insight into the levels of arsenic in a solution which are toxic to plants. Albert and Arndt (4) found 0.6 ppm arsenic to greatly reduce the growth of cotton in nutrient solution. Hurd-Karrer (43) studied arsenic toxicity on wheat in nutrient solution at various phos- phorus levels. Significant stunting occurred at 10, 60, and 120 ppm'arsenio with phorphorus levels of 10, 60, and 120 ppm, respectively. Machlis (63) found lethal concen- trations of arsenic were 18 and 2 ppm arsenic for Sudan 50 grass and bush bean (Phaseolus vulgaris var. humilis), respectively, while levels of 2.5 and 0.7, respectively, significantly reduced growth. In solution cultures, Clements and Munson (21) investigated the effects of arsenate and arsenite on tomato, Sudan grass, and beans at various phosphorus levels. At the 10 ppm phosphorus level, 5, l, and 0.25 ppm arsenate reduced growth of tomato, Sudan grass and bean, respectively, while at 120 ppm phosphorus, arsenate concentrations of 60, 12, and 3 ppm were required. Arsenite was found to be 10 times more toxic to Sudan grass and tomato and 4 times as toxic to bean plants as the arsenate form. Phosphorus had little effect on arsenite toxicity. Numerous studies have shown reduced growth of grasses after arsenical application (21, 34, 38, 54, 63, 69, 73, 77, 88, 96, 101, 112). In addition to these re- ports, several investigators have looked at arsenic toxicity and selectivity on £93 annua L. Freeborg (39) and Juska and Hanson (51) reported increasing arsenic toxicity at increasing arsenic rates. Roberts and Mark- land (87) noted decreased yields of foliage and roots at higher arsenic levels. Daniel (26, 27, 28) reported similar results and found Merion Kentucky bluegrass, 51 ryegrass, and creeping bentgrass to be more tolerant of arsenic than Poa annua L. Merion Kentucky bluegrass, red fescue, Colonial bentgrass, tall fescue, Zoysia-grass (Zoysia japgnia), creeping bentgrass, and bermudagrass were all found to be less susceptible to arsenic injury than annual bluegrass by Juska (50). He also indicated that increasing arsenic reduced Poa annua L. stands. Juska and Hanson (52) found calcium arsenate to control Poa annua L., but also noted some injury to Old Orchard, Seaside, Arlingtion, and Pennlu bentgrasses. Seven other bentgrass varieties were not injured. Sprague and Burton (100) and Engel et al. (36) indicated control of Poa annua L. in a bentgrass turf, using arsenicals. While the normal reSponse of plants to arsenic application is a reduction in growth, several reports of enhanced growth after arsenical treatment have been pub- lished (21, 22, 23, 63, 104). Freeborg (39) observed the response of Poa annua L.'to 48% tricalcium arsentate rates of 6, 12, 18, and 24 lb per 1000 ft2 on four soils in the greenhouse. Growth on the Kokomo silty clay loam was better at all arsenic rates than the control. The best growth was at the high arsenic rate and was 73% greater than the control. The Houghton muck and Crosby silt loam 52 also showed enhanced dry weight yields but only at the highest arsenic rate. In another greenhouse experiment, similar data was obtained, using the same soils. Freeborg indicated that differences could be due to extractable arsenic levels, varying nutrient levels between soils, or possible release of phosphorus by arsenic in the soil. MATERIALS AND METHODS The following experiments were conducted. Field Survey-Arsenic Contaminated Soils Several golf course soils in Michigan, which had received arsenic in the past, were sampled. All soils were sampled to a depth of 2 inches below the soil sur— face. Any thatch, which was present, was discarded. Between 15 and 20 cores were obtained per sample. Samples were analyzed for arsenic and phosphorus. In addition to the above field samples, Dr. Ray Freeborg supplied samples from various golf course areas in the Midwest. These samples were obtained from sites which had previously received arsenic applications. Cores were taken at 6 to 8 feet intervals across fairways. Greens were sampled 3 to 5 feet inside the perimeter (39). The first 2 inches below the soil surface was sampled, after discarding the thatch layer. Soil pH, available 53 54 phosphorus, and Bray P extractable arsenic were analyzed 1 on these samples. Greenhouse Experiment 1 Arsenic Rate-PhOSphorus Rate-68: The effects of various arsenic and phosphorus rates on the germination and growth of Penncross creeping bentgrass (Agrostis palustris Huds.) and on the arsenic and phosphorus soil test results were investigated. A Colwood sandy loam topsoil very low in natural phOSphorus was used (Table 2). The soil was obtained from Harsen's Island, Michigan. Waxed, 16 oz cottage cheese cartons, 4.5 inches in diameter, were used as growing con- tainers with 500 g air dry soil placed in each container. In the completely randomized 5X6 factorial exper- iment, all combinations of 0, 4.9, 9.8, 24.5, and 49.0 kg of 48% tricalcium arsenate(powder)/100 m2 and 0, 0.22, 0.43, 0.86, 1.72, 3.44 P(applied as 0-20-0)/100 m2 were investigated. All treatments were replicated 3 times. The arsenic source was 48% tricalcium arsenate (Ca3(AsO4)2), 18% actual arsenic, produced by Rhodia Inc., 55 muoumnonmq mowumwe AmuHEHH maomuomump 30Hmnv maomuomump uocr|.p.c HHom muflmum>fico mumum cmmflnoflz an owns mmuspmooum so owmmmses manmuomupxw mpHGOHnuflplmumuuHUITDMQOQHMOHmes U¢Ovmz 2 H .H< managomuuxme om»m mam «vow oma om HH.m III: m.mH o.mm ~.mm .p.c n mmm mama Hma mma vh.m mm.a mv.o m.m m.mm m.~m «v.0 He uonnm mm mmm mna mHH ma.m mm.a mm.o m.m m.mm m.mm mm.o mm uonrm mma mafia Hma mam mH.m mm.a me.o m.m m.mm m.mm mm.o we ponIm mm mmm Hma nma mn.v mm.H mm.o m.m m.mm m.mm vv.o mm uoamrm pm mourns mm course «mm mmmm mom mha mm.n om.m III: m.ma o.mm «.mm .p.s pm mem mmvm «ma mHH Hm.n om.m Inna m.mH o.mm N.mm .p.c om mmm mmwm Ana moa mm.> om.o III: m.ma o.mm ~.mm .o.c om mmv mmmm sea om mm.n om.w av.o m.ma o.m~ ~.wm .p.: om Hem mmvm oma Ha mm.n mm.m mm.o m.w m.am m.mn .p.c N .H . . m m a: mu m m z o «s 0 mm smao seam Seem embasz m ooa\me mm ucmEHummxm sssTHUM\Qq w «H< wmsoscwmuw .A can .m .e .m .m .H musmEHummxm CH poms mafiom mo mmfluummoum HMOfimhnm was HMOAETflUII.N mqmde 56 Chipman Division. The notation TCA will be used for this chemical throughout this thesis. Both the arsenic and phosphorus were mixed thoroughly with the soil prior to seeding. The pots were seeded on 7-26-68, immediately after applying the treatments. Seeding was at the rate of 2 1b/1000 ft2 (0.09 g/pot) of Penncross bentgrass seed. Germination counts were taken on 8-24-68 and 9-19-68. Clippings were collected periodically as the growth of the grass indicated. All 90 pots received supplemental nitrogen (.15 g urea/pot) and potassium (.22 g 0-0-60/pot) prior to seeding. Additional nitrogen was applied approximately monthly during the course of the experiment, depending on the growth and color of the turf. Supplemental light was used to provide a 14 hour photoperiod. Greenhouse temperatures were kept at 18 C or above if the air temperatures were greater than 18 C. Pots were rotated periodically to reduce border effects. The containers were watered daily, and twice a week brought to field capacity by weighing. The experiment was termi- nated on 11-7-68. All clippings were oven dried at 35 C before weighing. At the termination of the experiment, the soils 57 were air dried and stored at 22 C until analyzed for arsenic, pH, and phosphorus. Greenhouse Experiment 2 Arsenic Rate-Phosphorus Rate-Grass-69: The germi— nation and growth of Poa annua L., Poa pratensis L. Merion, Agrostis palustris Huds. Penncross, and Agrostis palustris Huds. Cohansey as affected by various rates of arsenic and phosphorus were investigated. The influence of arsenic and phosphorus on the arsenic and phosphorus soil tests were also explored. The Colwood sandy loam soil described in the pre- vious greenhouse experiment was also utilized in this investigation. Waxed, 32 oz cottage cheese cartons, 4.5 inches in diameter, were used. Each container received 1000 g air dry soil. The experimental design was a completely random- ized 5X4X4 factorial with 3 replications per treatment. All combinations of 0, 2.4, 4.9, 9.8, and 19.5 kg of 48% TCA(powder)/100 m2 and 0, l, 2, and 4 kg P(mono-calcium phosphate)/100 m2 were applied. The amount of soil 58 required for one treatment for all four grasses was placed in a polyethylene bag, and the apprOpriate treatment rates applied. The treated soil was then brought up to field capacity and allowed to dry. Each soil was incubated for 7 weeks through 7 wetting and drying cycles. At the end of 7 weeks, each bag was divided into 12 pots of 1000 g apiece. The appropriate grass variety was then planted on 7-17-69. Seeding rates for Poa annua L., Merion Kentucky bluegrass, and Penncross bentgrass, respectively, were 0.50, 0.34, and 0.11 g seed/pot. Cohansey was applied approximately at the rate of 20 bushels of stolons/lOOO ftz. Germination counts were obtained on 7-24-69 for annual bluegrass, Penncross bentgrass, and Merion Kentucky blue- grass. Each grass was clipped to a height of 1.5 inch when required. Supplemental nitrogen and potassium were applied prior to planting. Additional nitrogen was added as re- quired by the growth of each grass at the rate used in greenhouse experiment 1. Mildex was applied on 8-15-69 to retard powdery mildew activity. A photo-period of 14 hours was maintained with the aid.of an artificial light source. Temperature was main- tained at air temperature, but no lower than 18 C. 59 Individual pots were rotated weekly to avoid border ef- fects. Water was applied every other day, and all pots were brought to field capacity twice a week. Termination of the experiment was on 9-12-69. Clippings and soil samples were handled in the same manner as in Greenhouse Experiment 1. Greenhouse Experiment 3 High Phosphorus-Arsenic Rate-Grass-70: The growth of Poa annua L., Penncross bentgrass, and Merion Kentucky bluegrass as influenced by various arsenic rates in con- junction with various phosphorus levels was studied. Effects of arsenic rates and soil phosphorus level on the arsenic and phosphorus soil tests were also investigated. A Wisner sandy loam soil obtained from a long-term phosphorus level study at the Monitor Sugar Company Plant in Bay City, Michigan, was used. Soil was collected from phOSphorus plots, which ranged from high (80 lb P/A Bray P1 extracted) to extremely high (178 lb P/A) in phosphorus levels. This area was chosen in order to obtain a soil which had a range of high phosphorus levels, but had not 60 received phosphorus within 5 years. Golf course turf- grasses are often grown under similar high phosphorus level conditions. Each 16 oz cottage cheese container received 600 g air dry soil from the apprOpriate phosphorus plot area. The pots were 4.5 inches in diameter and 2.5 inches deep. A completely randomized 4X4X3 factorial design with 3 replications was used. Arsenic rates were 0, 4.9, 9.8, and 19.5 kg 48% TCA (powder)/100 m2. Phosphorus levels, based on the phosphorus soil tests, were 80, 109, 113, and 178 lbs P/acre. Prior to seeding, each treatment combination was incubated in the manner described in Greenhouse Experiment 2. £23,3n233 L., Penncross bent- grass, and Merion Kentucky bluegrass were seeded at the rates of 0.10, 0.10, and 0.15 g seed/pot, respectively, on 3-23-71. Each grass was clipped when required at a height of 1.5 inch. Watering, light, temperature, and other practices were similar to Greenhouse Experiment 2. The experiment was terminated on 7-8-71. 61 Greenhouse Experiment 4 Arsenic Rate-Arsenic Form-Phosphorus Rate-71: The previous experiments were conducted using powdered TCA, which was thoroughly mixed in the soil. This experiment was initiated to investigate the response of Poa annua L. to various arsenic rates, arsenic forms, and phosphorus rates. Both powdered and granular 48% TCA were used. The granular form is most commonly used by golf course super- intendents. The soil was the same as used in Greenhouse Exper- iment 3 (Table 2). Soil obtained from plots having phos- phorus soil tests of 80 and 178 lbs P/acre, respectively, provided the 2 phosphorus levels. Arsenic rates were applied at 0, 4.9, and 9.8 kg 48% TCA/100 m2 for both arsenic forms. The experimental design was a completely randomized 4X2x2 factorial with 3 replications per treat- ment. To a 16 oz cottage cheese carton 600 g of soil was added and seeded to Pga’annga L. on 2-24-71 at the rate of 0.10 g seed/pot. An additional seeding on 4-1-71 at the rate of 0.05 g seed/pot was necessary in order to obtain the desired stand. The grass was clipped at 1.5 inches 62 when needed. On 5-12-71 the arsenic rates were applied to the soil surface. Nitrogen was applied as needed during the course of the study. Light conditions, temperature conditions, and other practices were similar to Greenhouse Experi- ment 2. The experiment was concluded on 7-27-71. Greenhouse Experiment 5 Arsenic Rate-Phosphorus Rate-pH-70: The influence of soil reaction and phosphorus level on the effectiveness of 48% TCA as a control for annual bluegrass was investi- gated. The effects of soil reaction and phosphorus level on the arsenic and phosphorus soil tests were also studied. The experimental design was a completely randomized 4X2X4 factorial with 3 replications per treatment. Soils with pH levels of 4.78, 5.16, 6.19, and 6.74 were used. Arsenic rates were 0, 2.4, 4.9, and 9.8 kg 48% TCA(powder)/100 m2, while phosphorus rates were 0 and 0.43 kg P(mono-calcium phosphate)/100m2. The soil was a Hillsdale sandy loam collected from the Nitroqen Rate and Carrier Plots on the Michigan State 63 University Soils Farm at East Lansing (Table 2). In order to obtain an appropriate pH range, soil was collected from specific plots. Soil from plot 36 (no lime, no N) was used for the pH 4.78 set, plot 42 (no lime, 160 lbs N/A/yr as (NH4)ZSO4) for the pH 5.16 set, plot 25 (lime, 160 lbs N/A/yr as Ca(NO3)2) for the pH 6.19 set, and plot 41 (lime, 160 lbs N/A/yr as Ca(N03)2) for the 6.74 set. Soil from plot 41 received supplemental CaCO3 at the rate of 2 tons CaCO3/acre (2.10 g/pot) in order to raise the pH level. Arsenic and phosphorus treatments were mixed in the soil on 7-21-70. Each treatment combination was incu- bated until 9-17-70 in the manner described in Greenhouse Experiment 2. On 9-17-70, 600 g of treated soil was added to 16 oz cottage cheese containers and each pot seeded with 0.10 g Poa annua L. seed/pot. Environmental conditions and cultural practices were similar to those used in Greenhouse Experiment 2. Germination counts on all pots were obtained on 9-24-70. The experiment was terminated on 11-13-70. 64 Greenhouse Experiment 6 Texture-pH-Arsenic Rate-72: The effectiveness of arsenic for control of Poa annua L. as related to soil texture and soil reaction was investigated. The effects of soil texture and soil reaction on the arsenic soil test results were also explored. Five soils, varying in texture and other properties but similar in pH and phosphorus level, were used (Table 3). The experimental design was a completely randomized 5X3X3 factorial with 4 replicates per treatment. All soils were initially acid with pH values of 4.28 to 5.15. A portion of each soil was limed to approx— imately pH 6.5, while another portion was limed to about pH 7.5. The amounts of lime required to obtain these pH values were estimated by the following procedure (35); the initial pH of each soil was determined. To a mixture of 100 9 soil and 100 ml distilled water increasing rates of Ca(OH)2 were added. The mixture was shaken at room tem- perature for 3 days using a wrist-action shaker. At the end of this incubation period, pH was again determined. If there was a difference between the initial pH of the soil and the 100 9 soil + 100 ml distilled water (check), 65 huoumuonmq mcflumme HHom AmuHEHH manmuooump onva manmuomump uos||.p.s muflmum>flca mumum cmmflnoflz an poms monsomooum so memmsse manmuomnuxm muHQOASDHUIwummuflormmeOQHMOHmss necemz z H .H« manmuomuuxms -- -- -- ma «H.e om.am em.o -- -- -- .e.c d m-m -- -- -- mm oo.m om.em «v.0 -- -- -- .e.e m m-m mes move ems mm ma.s om.em mm.o -- -- -- .e.e a H-m -- -- -- ma em.n om.m 05.0 o.mm m.~m m.HH mm.o Hum m-o -- -- -- NH ma.m om.m em.o 0.0m m.mm m.aa mo.H Hum m-o mm mama pom NH Hm.e om.m em.o G.Gm m.~m m.HH mm.H Hum H-o -- -- -- me mm.e mm.m as.o H.m m.oe H.Hm om.o H m-o -- -- -- em mm.m mm.e mm.o H.m m.oe H.Hm no.0 H m-o ANN Home Ham me ma.m mm.e mm.o H.m m.oe H.Hm mm.o a H-o -- -- -- am we.» me.H mm.o m.oa m.em m.Hm oo.H Hm m-m -- -- -- ow ee.m me.H em.o m.oa m.e~ m.Hm mm.o Hm N-m am new mm om mm.e me.H Gm.o m.oa m.em m.Hm mm.~ Hm H-m -- -- -- mm mm.e am.H mm.o m.m H.m 4.4m mm.H ma m-a -- -- -- mm om.o em.a as.o m.o H.m 4.4m mm.H ma N-a as Hoa am am sm.v AG.H mm.o m.m H.m v.4m ~m.m ma H-a as mu m .m .z.o .«mOmmm swan seam. seem nonesz a . m ooaxme m HAom «sstUM\Qq m «Hm - mmsonsmouw .w usmeHmmxm ca poms mHHom mo mmfluummoum HMOflmmsm one HMUHEanII.m mamma 66 then all pH determinations were corrected with this value. The above data were then plotted and the lime requirement of each soil for the desired pH was calculated. Incubation of all soils was initiated on 11-23-71. Lime was applied at the appropriate rates as a 1:1 ratio of Ca(OH)2 and Mg(OH)2. Supplemental nutrients were added to Specific soils as indicated by soil test results. Spe- cific treatments applied at the start of the incubation period are given below for each soil. Soil A (Plainfield loamy sand); On an air dry basis, 575 9 soil was used for each pot. To obtain approximate pH values of 6.5 and 7.5, 0.47 and 0.80 g/pot each of Ca(OH)2 and Mg(OH)2 were applied, respectively. Potassium (.10 g KCl/pot) and magnesium (.50 g MgSO4/pot) were added. Soil B (sandy loam); Each pot received 650 g air dry soil. Ca(OH)2 and Mg(OH)2 rates were 0.85 and 1.30 g/pot of each to obtain pH levels of about 6.5 and 7.5, respec- tively. Phosphorus (.16 g Ca(H2PO4)-H20/pot), potassium (.10 g KCl/pot), and magnesium (.50 g MgSO4/pot) were added as supplements. Soil C (Conover loam); The weight of air dry soil used per pot was 500 9. Values of pH 6.5 and 7.5 were achieved by addition of 0.55 and 1.50 g/pot, respectively, A _- 67 each of Ca(OH)2 and Mg(OH)2. PhOSphorus (.16 g Ca(H2PO4) HZO/pot) was also added. Soil D (Kent silty clay loam); Each pot received 500 9 soil on an air dry basis. The quantities of Ca(OH)2 and Mg(OH)2 required to obtain approximate pH levels of 6.5 and 7.5 were 0.60 and 1.25 g each/pot, respectively. Magnesium (.50 g MgSO4/pot) was also added. Soil E (Rifle peat); The weight of air dry soil used per pot was 200 g. Ca(OH)2 and Mg(OH)2 rates of 1.00 and 2.00 g each/pot was necessary to achieve approximate pH levels of 6.5 and 7.5, respectively. No additional nutrients were necessary according to soil tests. Each treatment combination was incubated from 11-23-71 until 1-24-72. The soils were mixed periodically during this period and allowed to undergo several wetting and drying cycles. All pots were seeded on 1-24-72 at the rate of 0.10 g Poa annua L. seed/pot. The sandy loam pots required overseeding on 2-16-72, due to soil crusting resulting in non-uniform stands. On 3-17-72, arsenic was applied at rates of 0, 4.9, and 9.8 kg 48% TCA(granular)/100 m2 to the soil surface. 68 Greenhouse temperatures were maintained at 18 C or above depending on the air temperature during the course of the study. Each pot was brought up to field capacity every day. Nitrogen (.13 g NH NO3/pot) was applied uni- 4 formly to all treatments as required by growth of the annual bluegrass. Several foliar applications of phos- 205/pot) were required on the phorus as KH2P04 (.0011 g P loamy sand soil. Manganese as MnC12.4H20 (0.5 ppm solu- tion) was applied periodically as a foliar treatment to the peat. Foliar applications were to correct apparent deficiencies. Other environmental conditions and cultural prac- tices are the same as reported in Greenhouse Experiment 2. The study was ended on 5-25-72. Greenhouse Experiment 7 Arsenic Rate-Arsenic Form-Time of Application-71; The effects of granular and powdered calcium arsenate forms on the growth of Poa annua L. were investigated. Several rates applied both prior to seeding and to the established turf were included. The influence of the 69 above factors on the arsenic soil test results were also studied. A completely randomized 3X2X2 factorial with 3 replications was the experimental design. The soil collected from the same site as the soils for Greenhouse Experiment 3. It was a sandy loam with a phOSphorus soil test of 90 lbs P/acre (Table 2). Cottage cheese cartons, 16 oz, holding 600 9 air dry soil were used. Arsenic rates were 0, 4.9, and 9.8 kg 48% TCA/100 m2. A11 arsenic was surface applied. The pots were seeded on 2—24-71 with 0.10 g seed/pot. Due to poor germination, an overseeding was necessary on 4-1-71 at the rate of 0.05 g/pot. Arsenic application times were on 2-24-71, prior to seed germination, and 5-12-71, to a relatively mature turf (three leaf stage or older). All other prac- tices were as stated in Greenhouse Experiment 2. Growth Chamber Experiment 1 Arsenic Rate-H20 Rate-71: The influence of ar- senic rate and water rate on the growth of Poa annua L. and on the arsenic soil test results were investigated. The study was conducted in a controlled environment 70 chamber using a completely randomized 3X3 factorial design with 3 replications. The soil was a sandy loam, the same as used in Greenhouse Experiment 7 (Table 2). Each thoz cottage cheese container received 600 g air dry soil. Arsenic rates were 0, 4.9, and 9.8 kg 48% TCA(granular)/100 m2. The field capacity of the soil was determined by using a pressure chamber set at 1/3 atm. The water rates were 100%, 85%, and 70% of field capacity, which were equal to 100, 85, and 70 ml HZO/pot, respectively. Seeding was on 2-24-71 at the rate of 0.10 g P23 annua L. seed/pot and again on 4-7—71 at 0.059/pot. The second seeding was necessary in order to achieve the de- sired density. Arsenic was applied on 5-12-71 to the surface of the pots. Prior to this date all pots were watered at 100% field capacity. After this date the watering rates were applied daily. Nitrogen was applied to all pots when needed. Growth chamber temperatures were maintained at 24 C day and 15.6 C night. The daylength was 12 hours. .Pots were rotated daily. Other practices were as conducted in Greenhouse Experiment 2. 71 Statistical Methods An analysis of variance was compiled on each de- pendent variable in each experiment. The FACREP 3600 Computer Routine was used in all Greenhouse and Growth Chamber Experiments. Treatment means with significant F tests were examined at the 5% level, using the least significant difference (103). If there was no signifi- cant interaction between factors, then comparison of the average of one level of a factor against that of another level was allowed. Arsenic Determination (97) Reagents Reagent grade chemicals and distilled water were used in preparing all reagents. 1. Standard arsenic solution, 240 ppm As-O.100 g disodium hydrogen arsenate (heptahydrate) was dissolved in water and diluted to 100 m1. Work- ing standards were prepared from this stock solution. 72 Iodine solution, 0.02 N-In 25 ml water, 2.54 g of iodine and 8 g of KI were dissolved. The solution was brought to volume in a 1 liter volu— metric flask and stored in a dark bottle. Iodine solution, 0.001 N--This solution was pre- pared as needed from the 0.02 N iodine stock. Ammonium molybdate solution--To 300 ml water con- taining 70 ml concentrated H 80 5 g of ammonium 2 4' molybdate was added. The solution was diluted to 500 ml. Hydrazine sulfate solution, 0.15%. Potassium iodide solution, 15%--Prepared fresh as needed. Stannous chloride solution, 40%--Prepared fresh as needed. Zinc, mossy, coarse particles. Lead acetate, saturated solution--Prepared fresh as needed. 73 The distillation apparatus was constructed by the Glassblowing Shop at Michigan State University, following the plans of Small and McCants (97). Procedure An appropriate aliquot (usually 2 ml) of arsenic standard solution or soil extract solution was added to a 125 ml digestion flask. Enough distilled water was added to bring the total volume to 20 ml, followed by 10 ml concentrated HCl, 2 ml 15% KI solution, and 0.5 ml 40% SnClZ. The solution was mixed and allowed to stand. After 15 minutes, 5 g of mossy zinc was added and the flask was quickly connected to the funnel trap containing lead acetate saturated cotton. The dispersion tube was emerged in 30 ml of 0.001 N iodine in an ice bath. After one hour, the dispersion tube was removed from the col- lecting solution. To the test tube containing the col- lecting solution, 2 m1 of ammonium molybdate solution and 0.8 ml 0.15% hydrazine sulfate were added. After mixing, the test tubes were immersed in a boiling water bath for 20 minutes for color development. Color was read at 840 mu, using a Bausch and Lomb Spectronic 20 74 spectrophotometer. Color was stable for at least 24 hours but all samples were determined within 1 hour after color development. A distilled water blank was used for the 100% transmission setting. Arsenic Extractants Bray Pl Extractant (44) F + 0.025 N HCl Bray P extraction The 0.03 N NH 1 4 solution was used as a measure of "available" arsenic. To 2.50 g of soil in a 200 ml extraction bottle, 20 m1 of Bray P1 solution was added. The mixture was shaken for 1 minute and filtered using No. 42 Whatman filter paper. In the Results and Discussion section extractable arsenic will be referred to as the Bray Pl extractable arsenic fraction in all cases, unless otherwise stated. Sodium Bicarbonate Extractant (44) Another measure of "available" arsenic was deter- mined using 0.5 M NaHCO adjusted to pH 8.5 with NaOH. 3’ To 5 g of soil in a 250 ml extraction bottle, 100 m1 of 75 0.5 M NaHCO was added. The mixture was shaken for 30 3 minutes and filtered through a Whatman No. 40 filter paper. No carbon black was used. Phosphorus Determination (35) Reagents 1. Ammonium molybdate--In 850 ml water, 100 g ammon- ium molybdate was dissolved. A second solution of 1700 ml of concentrated HCl mixed with 160 m1 of water was made. The first solution was slowly added to the second while stirring. After mixing, 110 g of H3BO3 was added. 2. 1,2,4-amino reducing reagent--The following chem- icals were mixed together and ground to a fine powder; 2.5 g of 1-amino-2-naphthol-4-sulfonic acid, 5.0 g of Na SO , and 146.25 g Na When 2 3 needed 8.0 g of this powder was dissolved in 50 ml 25205' of warm water and filtered immediately before use. 76 3. Phosphorus standard, 218 ppm P40.100 g disodium hydrogen phOSphate (anhydrous) was dissolved in water and diluted to 100 ml. Working standards were made from this stock when required. Procedure An appropriate aliquot (usually 5 ml) of phos- phorus standard or soil extract was added to a 50 m1 flask. Distilled water was added to make a total volume of 12.5 ml, followed by 0.5 ml ammonium molybdate, and 0.5 ml 1,2,4-amino reducing reagent. The.solution was mixed. After exactly 15 minutes color was determined at 660 mu, using a Bausch and Lomb Spectronic 20 spectro- photometer. Iron Determination Free iron oxides were removed from the experi- mental soils using the bicarbonate-dithionite-citrate extraction given by Jackson (44). Iron was determined in the extract by the KSCN method reported by Jackson (44). 77 WOrking standards were prepared from aJJMM)ppm iron stock solution. A Bausch and Lomb Spectronic 20 was used to determine color. Aluminum Determination (86) Extractable aluminum was extracted by N NH4OAc adjusted to a pH of 4.8. The extraction procedure and the determination of aluminum in the extract by the alum- inum method was the same as reported by Rieke (86). Aluminum standards were prepared from a 1000 ppm aluminum stock solution. Color was determined on a Bausch and Lomb Spectronic 20 spectrOphotometer. Organic Matter Determination Organic matter was determined by dry combustion. Analyses were done by the Michigan State University Soil Testing Laboratory using a Leco Induction Furnace. 78 pH Determination Soil pH was determined using a 1:1 soil to dis- tilled water ratio. The suspension was stirred inter- mittently with a glass rod and pH determined after 1 hour. The above procedure was used for all mineral soils. With organic soils, a 1:2 ratio was used. All determinations were made with a Sargent-Welch Model DR pH meter. Texture Determination (29) Particle-size analysis was conducted using the hydrometer method reported by Day (29). All determina- tions were made in duplicate. Units of Measurements l. The following are standard terms used throughout 'the thesis. Definitions are: ppm As = ppm elemental arsenic in the soil ppm P = ppm elemental phosphorus in the soil 79 ppm As in extract = ppm elemental arsenic in the soil extract solution ppm P in extract = ppm elemental phosphorus in the soil extract solution 2. Conversions for the Bray Pl extract are: ppm As = (ppm As in extract)(8) ppm P = (ppm P in extract)(8) 3. Conversion for the NaHCO3 extract is: ppm As = (ppm As in extract)(20) 4. Conversion of units are: 1 lb 48% TCA/1000 ft2 = 0.49 kg 48% TCA/100 m2 1 lb P205/1000 ft2 = 0.22 kg P/100 m2 0.43 lb P/1000 ft2 RESULTS AND DISCUSS ION Arsenic Determination Distillation and Colorimetric Method The colorimetric arseno—molybdate method of Small and McCants (97) was used for all arsenic determinations. This method is sensitive, simple, reproducible, and rela- tively insensitive to interfering substances. Woolson (115) compared arsenic determinations by colorimetric and coulometric methods. Both procedures exhibited good recovery of added arsenic in standard solutions. The precision and accuracy of the two proce- dures were similar. Woolson (115) also investigated the influence of added potassium phosphate and ferrous sulfate on the Small and McCants (97) distillation method. No response to these substances was detected. The reproducibility of the Small and McCants (97) procedure was determined using standard arsenic solutions (Table 4). The highest relative standard deviation was for the 1.2 ppm arsenic standard, which occurs on a less 80 81 TABLE 4.--Standard deviation of arsenic standard solutions, using Small and McCants (97) procedure. PPM-As Percent Standard in solution Transmission Deviation (PPM-As)* 0.0 100.0 0.00 1.2 95.3 0.12 3.6 83.5 0.14 4.8 78.7 0.30 9.6 59.9 0.15 14.4 45.2 0.05 19.2 35.3 0.20 28.8 20.7 1.10 *Based on averages for 3 determinations. accurate portion of the arsenic standard curve (Figure l). The least deviation occurred between 9.6 and 28.8 ppm arsenic, which coincides with the most accurate portion of the curve between 20 and 60 percent transmission. Several soils, used in Greenhouse Experiments and 'which received arsenic, were sampled 3 times and arsenic determined each time by Small and McCants (97) procedure. Standard deviations were then calculated (Table 5). The standard deviation reflects not only errors encountered 82 TABLE 5.--Standard deviation of arsenic in soil extracts, using Small and McCants (97) procedure. Soil pH PPM-As in Standard Extract* Dev1ation Loam 5.15 0.43 0.15 " 4.83 6.97 0.67 " 4.81 14.80 0.69 " 6.52 0.60 0.00 " 6.25 3.97 0.23 " 6.16 11.37 1.75 " 7.82 0.67 0.46 " 7.75 3.57 0.29 " 7.69 7.67 0.42 Clay Loam 4.46 0.87 0.50 " 4.36 4.50 0.26 " 4.43 11.30 1.11 " 6.00 0.43 0.25 " 6.02 3.70 0.10 " 6.01 9.13 0.55 " 7.14 ' 0.43 0.35 " 7.17 3.03 1.04 " 7.18 6.53 1.02 Peat 4.43 29.13 3.18 " 7.18 11.40 0.75 Sandy Loam 6.93 4.97 0.32 " 6.92 5.47 0.50 " 6.87 51.80 3.27 *Based on triplicate extractions. 83 in the arsenic determination procedure but also sampling errors. The greatest relative standard deviation occurred on the samples very low in arsenic content. No differ- ences in standard deviation due to soil or soil reaction were apparent. Standard deviation was generally small in the 3 to 15 ppm arsenic (in extract) range, where most arsenic determinations are made. Selective Solvent Extraction and Colorimetric Method Paul (80) reported a procedure to simultaneously determine arsenic and phosphorus based on selective solvent extraction using isobutyl acetate as the solvent. After separation, arsenic and phosphorus were determined colori- metrically as arseno-molybdic and phosphomolybdic com- plexes. This procedure was investigated to determine if it was applicable to arsenic and phosphorus analyses in soil extracts. Several problems were encountered. When analyz- ing standard arsenic solutions, the results sometimes deviated markedly, even though a very good standard curve could be obtained. Another problem was the occurrence of a colloidal suspension, which interfered with color 84 determination, upon addition of ammonium molybdate to the test solution. This problem was overcome by cooling the ammonium molybdate and test solution to 4.5 C, and adding the ammonium molybdate slowly. A third problem was the inconsistency among isobutyl acetate batches, even though they were obtained from the same chemical company. Color development varied among batches. The above difficulties precluded the use of the Paul (80) procedure for routine arsenic analyses. Atomic Absorption Method Freeborg investigated the use of atomic absorption as a quick, efficient method for analysis of arsenic in soil extracts (39). In his procedure, arsenic was deter- mined directly from the Bray Pl extract. In order to compare the atomic absorption method with the Small and McCants procedure, Freeborg suggested joint analyses of extracts obtained from soils differing widely in properties. The soils were collected and ex- tracted by Freeborg, using the Bray Pl extract at a 1:10 ratio of soil:extract. Subsamples were forwarded by Freeborg for arsenic and phosphorus analyses, using the 85 distillation procedure for arsenic. Freeborg analyzed for arsenic by the atomic absorption method. Results are presented in Table 6. All determinations were made in duplicate. Freeborg (39) found the coefficient of determi- nation to be r2 = 0.63. In every case, the arsenic con- centration as determined by atomic absorption was lower than that found by distillation. However, the values were not uniformly lower as indicated by the r2 value. The greatest variability occurred at low arsenic levels. Interference from other substances present in the soil extract may result in low values in the atomic absorption procedure. Phosphorus Determination Phosphorus was determined by a colorimetric pro- cedure, utilizing the blue colored, heteropoly, complex of molybdo-phosphoric acid. Arsenate (As+5) can also produce a blue color upon formation of molybdoarsenic acid, resulting in erroneous phosphorus tests. Reduction of arsenate to arsenious acid prior to the addition of ammonium molybdate usually prevents interference (44). 86 TABLE 6.--Comparison of arsenic analyses by distillation and atomic absorption methods. PPM—As in Extract* Soil Percent _ . . Atomic B Of A Distillation (A)** Absorption (B)*** l 0.1 0.0 -- 2 0.1 0.0 -- 3 1.0 0.3 30.0 4 1.4 1.3 92.3 5 2.9 2.3 79.3 6 3.1 1.3 41.9 7 3.2 2.5 78.1 8 3,4 2.5 73.5 9 3.7 1.5 40.5 10 3.9 3.0 76.9 11 4.0 3.5 87.5 12 4.8 3.3 68.8 13 5.0 4.0 80.0 14 5.3 3.8 71.7 15 5.8 3.8 65.5 16 5.9 3.5 59.3 17 5.9 3.5 59.3 18 6.7 4.0 59.7 19 7.1 4.5 63.4 20 7.1 4.8 67.6 *Based on duplicate determinations. **Small and McCants (97). ***Freeborg (39). 87 Table 6.--Cont- PPM-As in Extract 5011 Percent .. ' Atomic B Of A Distillation (A) Absorption (B) 21 7.2 3.3 45.8 22 7.3 5.0 68.5 23 7.5 5.5 73.3 24 7.9 5.5 69.6 25 8.3 5 0 60.2 26 8.5 6.0 70.6 27 8.5 6.0 70.6 28 9.2 6.0 65.2 29 9.4 5.5 58.5 30 9.6 6.3 65.6 31 9.7 7.0 72.2 32 9.8 6.3 64.3 33 10.4 6.8 65.4 34 12.2 8.5 70.0 35 13.4 9.5 70.9 36 14.6 9.8 67.1 37 15.2 9_3 61.2 38 20.0 13.5 67.5 39 21.9 15.3 70.0 40 24.5 14.0 57.1 88 Jackson (44) stated that sodium bisulfite is often utilized to selectively reduce arsenate. Even though sodium bisulfite was used in all phOSphorus determina— tions in this study, appreciable positive interference due to arsenic occurred. This suggests that phosphorus soil tests from soils which have received arsenic may be incorrect. Field Survey-~Arsenic Contaminated Soils The arsenic and phosphorus soil test levels for green and fairway areas of several Michigan golf courses are given in Table 7. Arsenic soil levels ranged from 4.0 to 141.6 ppm arsenic, while phosphorus levels varied from 22.2 to 300 ppm phosphorus. The high phosphorus levels probably reflect arsenic interference as well as high phosphorus‘applications. Greens appeared to have higher arsenic levels than fairways, but the sample number was too small to indicate definite trends. Soil samples were collected on 28 golf courses in 6 Midwest states by Freeborg (39). All courses were using arsenicals for control of Poa annua L. Analysis for 89 TABLE 7.--Arsenic and phosphorus leVels on selected Michigan golf courses. Sample Site Years As Total Arsenic 2 PPM PPM Applied Applied/1000 ft As P Detroit GC, Detroit‘ #8 Green (N) 3 45 lb PbAs 60.4 112.0 #15 Green (N) 3 , .45 lb PbAs 78.0 252.0 Nursery 3 22 lb CaAs 106.4 300.0 Point O'WOOds GC, Benton Harbor #7 Fairway 6 45 lb CaAs 34.0 204.0 #15 Fairway 6 45 1b CaAs 8.8 76.0 #16 Fairway 6 45 1b CaAs 4.0 66.0 #7 Green 6 33 1b CaAs 51.2 119.2 #16 Green 6 33 lb CaAs 45.6 110.0 Riverside GC, Battle Creek #8 Green (F) 3_ 16 lb CaAs 93.6 154.0 #10 Green (B) 6 32 lb CaAs - 77.2 126.0 Maple Lanes GC, Warren #14 Green (W) 20 110 lb PbAs 83.2 202.0 #18 Green (W) 20 75 lb PbAs 88.0 162.0 Nursery 20 100 lb PbAs + 7 1b CaAs 141.6 206.0 Kensington GC, Brighton #3 Fairway 3 22 1b CaAs 52.0 22.0 #10 Fairway 3 24 1b CaAs 38.4 62.0 Average = 62.7 144.9 90 arsenic, phosphorus, and pH were determined by this author and the general results are given in Table 8. TABLE 8.--Arsenic, phosphorus and pH levels on Midwestern golf courses. Samples collected by Freeborg (39). Green ‘Fairway Nursery PPM-As, Range 29-219 1-245 -- PPM-As, Average 91 67 74 PPM-P, Range 88-210 12—252 -- PPM-P, Average 121 145 112 pH, Range 5.78-7.13 5.25-6.88 —- Number of Samples 16 25 l The average arsenic levels for both greens and fairways are high. The very high arsenic levels may be due in some cases to sampling within recent arsenic appli- cations. Greens tended to have higher arsenic soil tests than fairways. This could reflect higher application rates or less fixation due to lighter-textured soils. The very high phosphorus soil tests indicate ex- cessive phosphorus applications. Also arsenic interfer— ence in the phosPhorus soil tests could result in higher apparent levels. 91 Greenhouse Experiment 1 Phosphorus has been reported to reduce the effec- tiveness of arsenate as a control of annual bluegrass (26, 27, 39, 51, 58, 87). While it is generally recommended that phosphorus not be used by golf course superintendents on an arsenate program, there is little information on the magnitude of this antagonistic relationship. The purpose of this investigation was to gain insight into the rela- tive importance of the arsenate-phosphorus relationship. A soil very low in natural phosphorus was used. All combinations of 6 phosphorus rates and 5 arsenic rates were mixed with soil on which Penncross bentgrass was grown. High phosphorus rates were included because many golf course soils are high in phosphorus, and to study the interactions between high phosphorus and arsenic. Table 9 demonstrates the effects of arsenic and phosphorus on germination and growth of Penncross bentgrass and on arsenic soil test, phOSphorus soil test, and soil reac— tion. The analyses of variance for each dependent var- iable are in Table 29, Appendix. Increasing phosphorus, applied as ordinary super- phosphate, immediately prior to seeding, caused a highly TABLE 9.--Effects of arsenic and phosphorus on the germination and growth of Penncross bentgrass and on the arsenic soil test, phosphorus soil test and soil reaction. bated after treatment application. (Arsenic Rate-Phosphorus Rate). 92 Soil was not incu- Greenhouse Experiment 1 Clipping Weights Treatment Kg P Kg TCA (g/pot) Germ' PPM PPM per per Counts pH Number 100 m 100 m2 Per Pot As P 9-11 9-20 11-7 1 0 0 .03 .01 .28 407 0 5.6 7.38 2 0 4.9 .00 .00 .02 291 206 34.4 7.17 3 0 9.8 .00 .00 .00 251 462 58.4 7.11 4 0 24.5 .00 .00 .00 162 1368 115.2 7.16 5 0 49.0 .00 .00 .00 13 2584 153.6 7.20 6 .22 0 .19 .23 .20 305 0 17.6 7.12 7 .22 4.9 .16 .04 .30 239 203 57.6 7.12 8 .22 9.8 .01 .00 .04 178 452 66.4 7.14 9 .22 24.5 .00 .00 .00 75 1416 113.6 7.15 10 .22 49.0 .00 .00 .00 8 2728 150.0 7.16 11 .43 0 .25 .18 .18 271 1 38.4 6.99 12 .43 4.9 .17 .07 .14 259 201 61.6 6.99 13 .43 9.8 .00 .00 .10 114 506 89.6 7.12 14 .43 24.5 .00 .00 .00 66 1360 140.0 7.12 15 .43 49.0 .00 .00 .00 17 2424 173.6 7.10 16 .86 0 .16 .14 .35 107 1 89.6 6.91 17 .86 4.9 .09 .06 .21 98 214 100.0 6.93 18 .88 9.8 .01 .00 .06 66 458 156.0 6.92 19 .86 24.5 .00 .00 .00 10 1176 148.0 6.77 20 .86 49.0 .00 .00 .00 3 2344 177.6 6.83 21 1.72 0 .24 .20 .26 235 2 164.0 6.47 22 1.72 4.9 .11 .07 .22 130 186 191.2 6.47 23 1.72 9.8 .01 .01 .12 67 402 210.4 6.56 24 1.72 24.5 .00 .00 .01 7 1152 246.4 6.67 25 1.72 49.0 .00 .00 .00 1 2392 233.6 6.89 26 3.44 0 .14 .23 .46 139 1 401.6 6.10 27 3.44 4.9 .08 .02 .44 97 195 442.4 6.15 28 3.44 9.8 .03 .01 .26 25 402 442.4 6.10 29 3.44 24.5 .00 .00 .01 2 1064 450.4 6.22 30 3.44 49.0 .00 .00 .00 1 2088 437.6 6.40 LSD .08 .05 .09 103 151 37.6 0.11 .053 93 significant decline in germination, as shown by Treatment Numbers 1, 6, ll, 16, 21, and 26. Germination counts were obtained 30 days after treatment application and seeding. Any seeds which germinated were counted regardless whether the plant was still alive. A reduction in germi- nation due to phosphorus would not normally be expected. However, Kinra et a1. (58a) reported decreased oat and wheat germination after application of superphosphate. They attributed the response to water soluble fluorine present in superphosphate. The differences in plant density between phos- phorus levels precludes any definite conclusions on the effects of phosphorus on Penncross bentgrass growth. How- ever, two trends are indicated. The first is a positive growth response to phosphorus between Treatments 1 and 6. Little response is apparent after the initial phosphorus application of 0.2 kg P/100 m2. This pattern is especially evident on the first two clipping dates 9-11 and 9-20. The second is a tendency for reduced arsenic toxicity with increasing phosphorus. This trend is indicated by clip- ping weights on 11-7, especially. Also, visual observa- tions pointed out that high phosphorus levels increased 94 seedling survival at high arsenic levels, even though growth was slow. Phosphorus soil tests increased significantly after each added increment of phosphorus, except between the 0 and 0.22 kg P/100 m2 treatments. Treatment Numbers 1, 6, ll, 16, 21, and 26 reflect the true phosphorus soil tests, while the remaining values are erroneous due to a positive arsenic interference on the phosphorus soil test. The arsenic soil test values at the two highest arsenic levels were significantly reduced by increasing phosphorus. The presence of calcium in superphOSphate may reduce the solubility of tricalcium arsenate due to the common ion effect. This could result in lower arsenic soil tests. The above trend was observed only with the unusually high arsenic treatments and would be of little or no importance at normal arsenic levels. Soil reaction was significantly decreased by in- creasing phosphorus levels. The initial reaction of mono- calcium phosphate in soils is formation of CaH PO and 2 4 CaHPO4 in the soil solution. The saturated solution around the fertilizer granule is very acid and could decrease the soil pH. 95 Increasing arsenic significantly reduced germina- tion at all phosphorus levels. Without incubation of the arsenic prior to seeding, arsenic levels could be very high initially, as noted by several investigators (19, 21, 25, 45, 89). Such levels may result in direct kill of the seeds. Even though substantial differences in plant den- sity among phosphorus levels does not allow for definite comparisons of clipping yields among phosphorus levels to be made, a significant reduction in clipping yields due to arsenic is indicated. Increasing arsenic reduced yield at all phosphorus levels. High phosphorus levels, espe- cially on 11—7, reduced arsenic toxicity. The arsenic soil test results were significantly increased with increasing arsenic rates at all phosphorus levels. The correlation coefficient (r) between arsenic applied and the arsenic soil tests results was 0.93. The phosphorus soil test results were markedly affected by arsenic rates. With increasing phosphorus levels the influence of arsenic becomes less apparent. However, at lower phosphorus levels very serious errors in the phosphorus soil test occurred. Arsenic and phos- phorus were not separated prior to analyzing for phosphorus. 96 Since arsenic can form a blue-colored complex with ammon- ium molybdate, similar to the blue complex formed by phos- phorus, any arsenic present can cause an enhanced phos- phorus test. Arsenic exhibited no consistent influence on the soil pH regardless of the phosphorus level. Substantial differences in germination counts, due to phosphorus and arsenic levels, did not allow for meaningful comparisons of yields between various treat- ments. However, the preliminary information gained in the above experiment was helpful in planning later exper- iments. The importance of incubation of arsenic and phosphorus prior to seeding was indicated. The normal mode of action of arsenate is not inhibition of seed germination. Thus, to obtain a true measure of arsenic toxicity on turfgrasses the influence of arsenic on germi— nation needed to be reduced. Incubation allows for an equilibrium position to be approached where arsenic levels are sufficiently low enough so as not to influence germi- nation. The need for a phosphorus source other than super- phosphate was also indicated. Use of superphosphate introduced the possible extraneous influence of fluoride. 97 With use of superphOSphate, any effects due to fluoride could appear to be a phosphorus response. Greenhouse Experiment 2 The objectives of Greenhouse Experiment 2 were: 1) to investigate the influence of various arsenic and phosphorus levels on the germination of annual bluegrass, Merion Kentucky bluegrass, and Penncross bentgrass, 2) to investigate the influence of arsenic and phosphorus levels on the growth of the above grasses plus Cohansey bentgrass, and 3) to investigate the effects of arsenic and phos- phorus levels on the arsenic and phosphorus soil tests. The basis for the use of arsenic as a control for annual bluegrass is a selectivity for annual bluegrass over more desirable species. Thus, knowledge of the degree of selectivity of arsenic among the grasses is useful in estimating the safety margins in an arsenate program. All arsenic and phosphorus treatments were mixed into the soil 7 weeks before establishment of the respec- tive grasses. This period allowed for an equilibrium position to be approached, thus, eliminating the effects 98 of initially high arsenic and phosphorus soil levels. The phosphorus form was reagent grade monocalcium phosphate, which removed the possibility of fluoride effects. The clipping weights, germination counts, and water use data for annual bluegrass, Penncross bentgrass, Cohansey bentgrass, and Merion Kentucky bluegrass are given in Tables 10, ll, 12, and 13, respectively. Table 14 lists the arsenic soil test, phosphorus soil test, and pH determinations for all treatment combina- tions. The analyses of variance for Greenhouse Experi- ment 2 are in Table 30, Appendix. Germination counts were made on 7-24-69, 8 days after seeding. All seeds which germinated were counted, whether the plant survived or not. Phosphorus had a sig- nificant effect on germination. However, no particular pattern was apparent. The germination counts of Merion Kentucky bluegrass, which germinates slowly, varied the most and may have had an undue influence on the degree of significance. Clipping yields for annual bluegrass, Penncross bentgrass, and Merion Kentucky bluegrass were signifi- cantly affected by phosphorus on all dates. The greatest phosphorus response occurred between the 0 and 1 kg P/100m2 99 TABLE 10.--Poa annua L. clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate-Phosphorus Rate-Grass). Clipping Weights Trt. Kg P Kg TCA (g/pot) Germ. H 0 Use per per Counts on 8-25 Number 100 m 100 :11 Per Pot ml 8-1 8-6 8-18 8-28 9-12 1 0 0 .04 .12 .09 .18 .20 937 95 2 0 2.4 .04 .ll .09 .12 .10 965 85 3 0 4.9 .03 .11 .05 .07 .02 942 73 4 0 9.8 .02 .02 .00 .00 .00 904 43 5 0 19.5 .00 .00 .00 .00 .00 912 60 6 1 0 .19 .26 .26 .35 .71 876 115 7 1 2.4 .14 .23 .24 .33 .54 952 115 8 l 4.9 .10 .23 .22 .26 .33 933 100 9 l 9.8 .03 .13 .15 .18 .05 884 92 10 1 l9 5 .01 .02 .03 .02 .00 940 75 ll 2 0 .20 .22 .18 .37 .62 845 130 12 2 2.4 .16 .18 .18 .40 .53 920 152 13 2 4.9 .09 .16 .15 .30 .32 928 97 14 2 9.8 .03 .09 .10 .12 .11 781 77 15 2 19.5 .01 .03 .02 .02 .01 824 78 16 4 0 .19 .22 .18 .35 .68 985 135 17 4 2.4 .22 .25 .18 .44 .55 834 137 18 4 4.9 .12 .19 .13 .20 .29 913 137 19 4 9.8 .06 .19 .15 .23 .26 940 97 20 4 ‘19.5 .03 .08 .06 .06 .04 856 90 LSD a .03 .04 .04 .09 .13 107 24 100 TABLE ll.--Penncross bentgrass. Clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate- Phosphorus Rate-Grass). Clipping Weights Kg P Kg TCA Germ. H 0 Use Trt. (g/pot) N er per per 2 , Counts on 8-25 “a: 100 00 P P t 1 m 1 m 8-1 8-6 8-18 8-28 9-12 er 0 m 1 0 o .00 .04 .12 .21 .32 427 125 2 0 2.4 .00 .02 .12 .19 .20 345 108 3 0 4.9 .00 .02 .07 .13 .13 340 90 4 0 9.8 .00 .01 .04 .08 .03 357 82 5 0 19.5 .00 .00 .01 .08 .00 421 68 6 1 0 .03 .14 .26 .22 .62 328 125 7 1 2.4 .03 .13 .28 .16 .38 329 138 8 1 4.9 .02 .10 .24 .32 .34 313 127 9 1 9.8 .02 .09 .23 .23 .44 392 115 10 1 19.5 .01 .05 .18 .18 .18 399 108 11 2 0 .06 .18 .12 .21 .58 331 167 12 2 2.4 .04 .14 .19 .23 .58 381 157 13 2 4.9 .04 .13 .15 .31 .34 332 147 14 2 9.8 .03 .10 .16 .22 .28 389 122 15 2 19.5 .02 .06 .14 .13 .20 383 118 16 4 ' 0 .07 .18 .13 .21 .69 439 148 17 4 2.4 .06 .17 .17 .23 .77 421 153 18 4 4.9 .08 .17 .14 .24 .63 509 170 19 4 9 8 .04 .10 .11 .20 .45 471 137 20 4 19 5 .01 .05 .11 .24 .31 440 132 LSD = .03 .04 .04 .09 .13 107 24 TABLE 12.-~Cohansey bentgrass. and water use data. Phosphorus Rate-Grass). 101 Clipping weights, germination counts, Greenhouse Experiment 2 (Arsenic Rate- Clipping Weights Trt. Kg P Kg TCA. (g/pot) H 0 Use per per on 8-25 Number 100 m 100 m m1 8-1 8-6 8-18 8-28 9-12 1 0 0 .06 .09 .13 .16 .20 100 2 0 2.4 .05 .08 .11 .16 .18 92 3 0 4.9 .03 .07 .10 .13 .14 98 4 0 9.8 .07 .09 .08 .11 .12 90 5 0 19.5 .07 .08 .06 .08 .08 85 6 1 0 .08 .11 .16 .27 .30 117 7 1 2.4 .06 .11 .15 .25 .26 105 8 1 4.9 .07 .10 .18 .24 .30 100 9 1 9.8 .05 .11 .15 .21 .23 88 10 1 19.5 .07 .10 .12 .18 .19 100 11 2 0 .12 .13 .10 .25 .31 112 12 2 2.4 .06 .09 .12 .23 .27 108 13 2 4.9 .08 .12 .14 .24 .31 120 14 2 9.8 .05 .08 .13 .19 .22 100 15 2 19.5 .04 .06 .07 .13 .17 93 16 4 0 .06 .10 .14 .25 .33 132 17 4 2.4 .05 .08 .14 .22 .34 112 18 4 4.9 .06 .08 .13 .22 .27 127 19 4 9.8 .04 .08 .11 .21 .27 137 20 4 19.5 .03 .05 .11 .19 .21 115 LSD .03 .04 .04 .09 .13 24 102 TABLE l3.--Merion Kentucky bluegrass. Clipping weights, germination counts, and water use data. Greenhouse Experiment 2 (Arsenic Rate-Phosphorus Rate-Grass). Clipping Weights Kg P Kg TCA Germ. H 0 Use Trt. per per (g/pot) Counts on 8-25 Number 100 m 100 m2 Per Pot m1 8-1 8-6 8-18 8-28 9-12 1 0 0 .00 .00 .00 .09 .07 122 87 2 0 2.4 .00 .00 .00 .06 .041 125 82 3 0 4.9 .00 .00 .00 .04 .02 127 77 4 0 9.8 .00 .00 .00 .02 .00 125 68 5 0 19.5 .00 .00 .00 .00 .00 113 52 6 l 0 .00 .00 .07 .50 .39 169 117 7 l 2.4 .00 .00 .02 .25 .21 120 100 8 l 4.9 .00 .00 .02 .29 .33 59 138 9 l 9.8 .00 .00 .03 .26 .17 130 113 10 1 19.5 .00 .00 .03 .18 .07 121 105 11 2 0 .00 .00 .06 .41 .32 149 112 12 2 2.4 .00 .00 .03 .33 .24 107 115 13 2 4.9 .00 .00 .02 .17 .16 62 95 14 2 9.8 .00 .00 .02 .19 .14 105 95 15 2 19.5 .00 .00 .02 .19 .17 100 90 16 4 0 .00 .00 .05 .49 .42 126 132 17 4 2.4 .00 .00 .05 .42 .37 135 122 18 4 4.9 .00 .00 .07 .34 .26 183 112 19 4 9.8 .00 .00 .06 .24 .28 161 117 20 4 19.5 .00 .00 .02 .14 .15 106 118 LSD = .03 .04 .04 .09 .13 107 24 .05 103 TABLE l4.--Arsenic soil test, phosphorus soil test, and pH data for Poa annua L. in Greenhouse Experi- ment 2 (Arsenic Rate—Phosphorus Rate-Grass). Treatment Earp ngECA PPM PPM pH Number 100 m2 100 m2 As P 1 0 0 2.4 5.6 6.92 2 0 2.4 44.0 12.0 6.92 3 0 4.9 84.8 20.0 6.94 4 0 9.8 195.2 25.6 6.92 5 0 19.5 390.4 39.2 6.87 6 1 0 1.6 64.8 6.89 7 1 2.4 41.6 54.4 6.93 8 1 4.9 85.6 60.0 6.96 9 1 9.8 219.2 80.0 6.92 10 1 19.5 407.2 112.0 6.95 11 2 0 2.4 121.6 6.68 12 2 2.4 48.8 128.0 6.73 13 2 4.9 98.4 134.4 6.77 14 2 9.8 189.6 142.4 6.79 15 2 19.5 346.4 136.0 6.88 16 4 0 2.4 278.4 6.40 17 4 2.4 48.8 268.0 6.48 18 4 4.9 97.6 264.0 6.56 19 4 9.8 211.2 264.0 6.53 20 4 19.5 391.2 292.0 6.45 LSD 17.8 16.1 .03 104 rates. Little significant increase in yield due to phos- phorus was apparent beyond the 1 kg P/lOO 102 rate. Cohansey was less responsive to phosphorus, but the general trend was still present. Increasing phosphorus tended to reduce arsenic toxicity on all grasses, with allclipping dates except 8-28 causing a significant arsenic-phosphorus interaction. The greatest reduction in arsenic toxicity by phosphorus application occured between the 0 and 1 kg P/lOO m2 levels for all grasses. Increasing phosphorus beyond 1 kg P/lOO m2 had less effect on arsenic toxicity. The degree of the arsenic-phosphorus interaction is shown in Figures 2, 3, 4, and 5 for the 4 grasses. Water use data was collected dyer a 2-day pefldod starting on 8-25—69. Phosphorus significantly affected water use of all 4 grasses. Increasing phosphorus en- hanced the water consumption rate of all grasses. The increased water usage due to phosphorus was probably a result of more growth at higher phosphorus rates. Penn- cross consumed the most water, while the other grasses were lower. Differences could be due to plant density or inherent differences between grasses. 105 All chemical analyses were conducted on the Egg annua L. soils. It was assumed that no differences in phosphorus soil test, arsenic soil test, or pH occurred between grasses. Applied phosphorus produced significant increases in the phosphorus soil test at all levels as demonstrated by Treatment Numbers 1, 6, ll, and 16 (Table 14). Soil reaction was significantly reduced by applied phOSphorus at all levels. While phosphorus exhibited a significant influ- ence on arsenic soil test results at high arsenic levels, there was no consistent pattern due to phosphorus level. Arsenic levels had no significant effect on the germination of any of the grasses. Incubation of the arsenic prior to establishment could have reduced arsenic soil levels below concentrations inhibitory to the turf- grass seeds. Clipping weights of all grasses were significantly reduced by arsenic. This reSponse occurred on almost every clipping date for every grass. Poa annua L. was most affected by increasing arsenic, followed by Merion Kentucky bluegrass, Penncross bentgrass, and Cohansey bent- grass. The selectivity of arsenic for Poa annua L. has been noted by other investigators (26, 27, 28, 36, 39, 106 50, 52, 100). Generally, arsenic soil levels between 40 and 98 ppm As resulted in significant reduction of annual bluegrass growth. Figures 2, 3, 4, and 5 demonstrate arsenic toxicity and selectivity on the grasses. Increasing arsenic significantly reduced water consumption of all Species, but Poa annua L. was most af- fected. Less growth with increasing arsenic could decrease water usage. Species responses could be due to differences in plant density, plant growth rate, or inherent water consumption rates. Water consumption included transpira- tion and evaporation. Phosphorus soil test results were significantly enhanced by arsenic. The 2 and 4 kg P/lOO m2 levels were much less affected by arsenic than the 0 and 1 kg P/lOO m2 rates. At the levels of phosphorus encountered in most soils, the presence of arsenic could cause serious errors in the phosphorus soil test when determined by these pro- cedures. Applied arsenic was significantly reflected in the arsenic soil test. The correlation coefficient (r) was 0.95 between arsenic soil test and arsenic rate. While arsenic was indicated to have a significant influence on pH, no consistent trends were obvious. Any 107 change in soil pH due to arsenic is apparently of minor importance. Greenhouse Experiment 3 In Greenhouse Experiment 3, the influence of var- ious arsenic and phosphorus rates on the growth of several grasses was investigated. Interactions between phosphorus and arsenic were also studied. In previous experiments a soil very low in phos— phorus was used and phOSphorus was applied to obtain dif- ferent soil phOSphorus levels. The amounts of phosphorus required to achieve the desired levels were in excess of those normally applied on turfgrass soils at any one time. This may have resulted in a higher degree of arsenic- phosphorus interaction than would be generally encountered. To eliminate this possibility, soil was collected from long term phosphorus study plots which had not received phosphorus for 4 years. Soil was obtained from selected plots which provided a range of high phosphorus levels. The clipping weights for Poa annua L., Penncross bentgrass, and Merion Kentucky blugrass are in Tables 15, 16, and 17, respectively. The soil test data for all 108 .Umuomamm maaom mzu mo mummu aaOm mduonmmosm« ooo. moo. oaa. ooa. moa. amo. mmo. a mo ama ooo. ooo. ooo. ooo. voo. omo. hao. ona m.oa ma mmo. vma. moa. ova. aoa. mma. ooo. ooa m.m ma oma. omm. mom. omm. oma. mma. moa. ova o.v va omm. vvv. mvm. 5mm. mam. avm. aam. ooa o ma ooo. moo. moo. vao. ooo. moo. moo. maa m.ma Na moo. mom. moa. oma. mma. Nma. vma. maa m.o aa moo. mmm. mam. omm. mma. boa. voa. maa o.v oa mom. mmv. vom. mom. mvm. mmm. mom. maa o o aoo. ooo. ooo. mao. ooo. nao. ooo. moa m.oa m omo. mma. ooo. oma. vva. aaa. vmo. ooa o.o o moa. ham. omm. amm. mmm. voa. moo. moa m.v m omm. mom. vam. ovm. moa. «mm. ama. ooa o m ooo. ooo. ooo. ooo. ooo. ooo. oao. om m.oa v mmo. mma. mma. moa. ama. ooo. mmo. om o.o m oaa. oma. mom. oma. oma. mma. moo. om m.v m mom. mov. mmm. mom. nma. ona. mma. oo o a min mane mum amum hum mmuv malv ¢\m na NE ooa gonadz «am>ma mom cmeummue Auoo\oo unoawz mcaomaao m 409 mm u .ammmnwnmumm oasmmudlmsuozmmonm smamv m usmEaummxm omsoncmmuo How .a macaw mom mo munmam3 msammaa011.ma mamas 109 .owuowamm maaom mnu mo mummy aaOm m« moo. moo. oaa. ooa. moa. amo. mmo. own awo. vmo. Haa. who. oma. boa. mao. oha m.ma ma mom. mvm. mma. ova. mmm. mmm. amo. mha o.m ma omv. mom. vow. vmm. mmm. hma. hmo. mba m.v va oam. mom. mom. mmm. mom. ama. ooo. mha 0 ma moa. moo. Naa. woo. woo. amo. moo. mad m.ma Na oaa. ooo. hmo. omo. boo. mmo. ooo. maa m.o Ha hmv. mom. mam. mom. mmm. 5mm. vva. maa m.v oa amm. ovm. mmm. mom. amN. mom. mom. maa o m moo. vmo. mma. hmo. who. amo. aoo. moa m.ma m vmm. mom. mmm. vba. maa. boo. mNo. ooa m.o h mov. va. oam. mom. 5mm. amN. omo. ooa m.v o Hem. mom. ham. mvm. hum. mam. mmo. moa o m ooo. ooo. vmo. ovo. ovo. mmo. ooo. om m.ma v moa. who. ooo. mmo. ovo. ovo. ooo. om m.m m. omv. 5mm. Ham. mma. wvm. mmm. moo. om m.v N omm. 0mm. mvm. mma. ama. aaa. Nao. om o a win male mlm amlm him omlv walv ¢\m Qa NE ooa HwQEdz aao>ma mom so some 2666\mv unmawz maamaaao m ma mom coaummua Anom\mo unmams mcammaao a mom on » .Ammmnwnmumm oacmmh¢lmsuonmmonm soamo m uswfia Inwmxm mmsoncmmnw mom mmMHOGSaQ axospcmm soanmz mo magmam3 osammaa011.ha mamda 111 grasses are in Table 18. Analyses were determined on the Poa annua L. soil, but selected treatments were analyzed on the other grasses with no difference in results. Table 31 in the Appendix contains the analyses of variance for the data in the above tables. TABLE 18.--Arsenic and phosphorus soil test data for Greenhouse Experiment 3 (High Phosphorus- Arsenic Rate-Grass). Kg TCA P Treatment per Level* PPM PPM Number 100 m2 Lb P/A As P l 80 1.6 40.0 2 80 63.2 48.6 3 80 108.8 59.4 4 80 243.2 73.4 5 109 0.8 54.6 6 109 68.0 68.0 7 109 134.4 79.4 8 109 256.0 98.6 9 113 0.8 56.6 10 113 65.6 74.0 11 113 137.6 82.6 12 113 260.0 102.0 13 178 1.6 89.0 14 178 72.8 99.4 15 178 133.6 97.4 16 178 237.6 122.0 LSD.05 = 21.0 6.2 *Phosphorus soil tests of the soils selected. 112 Phosphorus had a significant influence on the growth of all grasses. Generally, as soil phOSphorus test increased yield of annual bluegrass and, to a lesser ex- tent, Penncross bentgrass increased. Merion bluegrass tended to respond positively to phosphorus up to the 113 lb P/A level, then decreased in growth at the 178 lb P/A level. Poa annua L. is favored by high phosphorus which may explain its stronger response to phosphorus (l3, 14). There was no apparent explanation for the reduction in growth of Merion at the highest phosphorus level. One possibility could be phosphorus induced iron deficiency. In general, high phosphorus decreased arsenic toxicity, but a significant interaction did not occur on all dates. Reduced toxicity was most conspicuous on Merion and least on £23.32ggg L. The data from this experiment and Greenhouse Experiment 2 indicate that while Pga_angga L. exhibits decreased arsenic toxicity with increasing phosphorus, the magnitude of the interaction may not be great. Arsenic soil test results were significantly in- fluenced by phosphorus. However, significant differences occurred only at the 2 highest arsenic rates, and with no 113 consistent relation to phSOphorus level. Similar observa- tions were made in Greenhouse Experiment 2. Clipping yields on all dates were significantly affected by arsenic. Increasing arsenic decreased yields of all grasses. Poa annua L. was most sensitive to ar- senic. Based on clipping weights and arsenic soil test results arsenic soil levels of 50 to 60 ppm were generally sufficient to significantly reduce Poa annua L. growth. As in Greenhouse Experiments 1 and 2, arsenic significantly affected phosphorus soil test results. Increasing arsenic levels markedly increased the soil phosphorus tests with the most noticeable influence occurring at the low phosphorus levels. The arsenic soil test values significantly re- flected the applied arsenic, regardless of phosphorus level. The correlation coefficient (r) was 0.97. Greenhouse Experiment 4 In Greenhouse Experiments 2 and 3 powdered tri— calcium arsenate had been thoroughly incorporated into the soil and allowed to incubate in order to approach an 114 equilibrium position. The various grasses were then established. Such a procedure allowed for arsenic soil test determinations to be conducted at a period in time when equilibrium conditions were approached. These values could then be related to the growth of each species. While the above situation allows the investigator to determine with some degree of certainty the actual arsenic levels in the soil which are toxic to a particular Species, it does not simulate natural conditions. Golf course superintendents usually apply granular form tri— calcium arsenate to the surface of a mature turf. The current experiment compared powdered and gran- ular tricalcium arsenate forms at several rates on Pea annua L. All arsenate was surface applied to a mature turf stand (older than the 3-leaf growth stage). The soil was the same as used in the previous experiment. Table 19 contains the clipping weight and soil test data for Greenhouse Experiment 4. Table 32, Appendix, contains the analyses of variance for data in Table 19. Phosphorus affected yield on 3 of the 5 clipping dates. On 5-21 a significant increase in clipping weight was apparent at the high phosphorus level. However, on 7-13 and 7-27 the reverse was generally true, regardless 115 <09 Hmascmumulu doe Hmo3omlum« mo. v.n a.mm aoa. moo. moo. mmo. ooo.." oma o.mm v.vm moo. amo. omm. mom. vav. mha m.m O oa o.nm m.mm mva. amo. mom. Nvm. mmm. ona o.v 0 m o.vaa m.aoa mmo. mao. mvm. mom. aom. mha o.o m m m.aoa v.mo amo. mao. Nam. mom. 5mm. mha m.v m h m.om o mom. mma. mom. mvm. amv. oha o I o o.mm o.maa oNN. moo. omm. mom. mom. om m.m U m v.mm m.av mom. moo. mmm. ovm. amm. om m.v w v m.vm m.noa vva. omo. mmm. oma. mam. om o.m m m o.om o.mn voa. amo. mam. noa. amm. om o.v m m o.vm o mov. mma. amm. Nmm. mom. om o I a m mm owns mans maum vim amnm MM mauwma «such quEsz 2mm 2mm Au0m\ov unoamz mcammaau m 409 mm «09 unmaumoue .Amumm monogamonmIEuom oaaomudlmumm oacomufiv v ucmEaummxm mmsoncmmuw How mummy aaOm monogamonm paw .mummu aaOm oacmmum .munmamz mcammaa011.oa mamme 116 of arsenic rate or form. Phosphorus induced iron (or other micronutrient) deficiency as the experiment pro- gressed may have caused this trend. In Greenhouse Exper- iment 3, a similar tendency of growth reduction at higher phOSphorus rates was observed for Merion Kentucky blue- grass grown on the same soil type, but Poa annua L. did not. The differences in reSponse of Poa annua L. between experiments could be due to the time factor. In Greenhouse Experiment 4, annual bluegrass was grown for a longer period of time before yields were collected, which may have reduced the levels of readily available iron. Phosphorus had little influence on arsenic toxi- city or arsenic soil test results. No significant inter- action occurred between phosphorus level and arsenic form. Arsenic significantly reduced growth on all dates, but the trend became more pronounced with time. The simple correlation coefficients (r) between arsenic soil test and clipping yields for the dates 5-21, 6-4, 6-19, 7-13, and 7-27 were -.21, -.62, -.47, -.60, and -.70, respectively. These correlations are averages over phosphorus levels, arsenic levels, and arsenic forms. A soil arsenic level of 50 ppm was capable of significantly reducing growth based on the 7-27 clipping weight data. 117 Arsenic form had a significant influence on growth. Powdered TCA generally decreased yields more than similar rates of granular TCA. This trend was conspicuous at both phosphorus levels, but more pronounced at the lowest. Increasing arsenic significantly increased the arsenic soil test results, irrespective of form. However, significant differences were evident between arsenic forms. Powdered TCA caused substantially higher arsenic soil tests than granular TCA. This is in agreement with the yield data, where powdered TCA produced greater toxicity. Ap- parently, the greater surface area of powdered TCA allows it to become active at a faster rate than granular TCA. Given sufficient time one would expect the 2 forms to pro- duce similar soil test results. As in previous experiments increasing arsenic significantly enhanced the phosphorus soil test results. PhOSphorus soil tests were influenced most at the low phOSphorus level. Powdered TCA produced the greatest positive influence on phosphorus soil tests which would agree with the higher arsenic soil activity for powdered TCA. The arsenic soil tests were well correlated to arsenic applied with an r of 0.89. This value is somewhat 118 lower than the correlations present in previous experi- ments. However, it reflects an average of arsenic rate, arsenic form, and phosphorus level. Greenhouse Experiment 5 In previous experiments arsenic and phosphorus interactions in plants and soils were studied. The soils used in these investigations were all calcareous, and the results may not have been representative of more acid soils. The objective of this experiment was to study arsenic and phosphorus interactions over a range of pH values. Soil was collected from selected plots in the long—term Nitrogen Rate and Carrier study at the Michigan State University Soils Farm, East Lansing. This allowed for a range of pH values to be obtained on the same soil type. In order to obtain a wider range in pH levels, one of the soils (from plot.4l)received supplemental lime in this study. Arsenic was applied as powdered TCA, mixed into the soil, and incubated for several weeks prior to seeding-with Poa annua L. 119 Table 20 contains the clipping weight, germination, arsenic soil test, phOSphorus soil test, and pH data for Greenhouse Experiment 5. The analyses of variance for the data in Table 20 are in Table 33, Appendix. Considerable variation in clipping weights occurred between pH levels. This variation was not due to germina- tion differences (Table 20). Treatment Numbers 17 to 24 exhibited poor growth on all dates. Treatment Numbers 9 to 16 also produced less growth than expected. The grass in these Treatments (9 to 24) differed markedly in appear- ance from grass in Treatments 1 to 8 or 25 to 32. The turf was light green in color and exhibited leaf dieback progressing from the leaf tip downward. The youngest leaves were affected first. The visual symptoms along with the soil test data indicated possible magnesium deficiency. Magnesium was applied to the soil at the rate of 0.33 g MgSO4/pot on 10-16-70 and again on 10-23-70. Some recovery was ob- served but it was minimal, which implied that magnesium was not the cause of the poor growth. In separate pots, MgSO was thoroughly mixed in the soil used for Treat- 4 ments 17 to 24 at the rate of 1.45 g/pot with no apparent effect. Thus, the poor growth was due to another factor. 120 TABLE 20.--Clipping weights, germination counts, arsenic soil tests, phosphorus soil tests, and soil pH levels for Greenhouse Experiment 5 (Arsenic Rate-Phosphorus Rate-pH). Trt. Kg TCA Kg P Clipping weight(g/pot) Germ. PPM PPM Number per per pH* Counts As P 100 m 100 m 10-15 10-23 11-3 11-13 Per Pot 1 0 0 4.78 .248 .116 .158 .057 229 1.6 78.6 2 2.4 O 4.85 .166 .102 .120 .063 271 32.0 100.0 3 4.9 0 5.03 .005 .000 .000 .000 231 84.0 121.4 4 9.8 0 5.27 .000 .000 .000 .000 233 203.2 153.4 5 0 .43 5.19 .301 .124 .192 .091 261 1.6 100.6 6 2.4 .43 5.17 .110 .086 .117 .074 257 30.4 115.0 7 4.9 .43 5.31 .018 .011 .002 .000 243 79.2 132.0 8 9.8 .43 5.56 .001 .000 .000 .000 224 212.0 168.0 9 0 0 5.16 .057 .039 .072 .060 242 4.0 108.6 10 2.4 0 5.43 .074 .043 .043 .031 270 48.8 126.0 11 4.9 0 5.33 .027 .016 .009 .004 285 110.4 151.4 12 9.8 0 5.59 .007 .000 .000 .000 245 247.2 174.0 13 0 .43 5.21 .125 .053 .109 .066 271 1.6 156.0 14 2.4 .43 5.54 .112 .067 .089 .062 284 44.0 168.6 15 4.9 .43 5.53 .026 .015 .017 .011 265 107.2 177.8 16 9.8 .43 5.58 .013 .009 .005 .003 259 230.4 227.4 17 0 0 6.19 .011 .005 .007 .006 259 2.4 59.2 18 2.4 0 6.29 .015 .014 .018 .018 250 44.8 70.0 19 4.9 0 6.25 .020 .013 .011 .012 233 113.6 82.6 20 9.8 0 6.33 .031 .019 .015 .014 263 234.4 99.4 21 0 .43 6.12 .013 .012 .028 .014 248 4.0 94.6 22 2.4 .43 6.24 .020 .021 .059 .028 275 52.0 114.2 23 4.9 .43 6.30 .035 .026 .038 .015 232 110.4 126.6 24 9.8 .43 6.47 .032 .028 .045 .031 264 228.0 142.0 25 O 0 6.74 .169 .105 .155 .053 257 2.4 78.0 26 2.4 0 6.84 .195 .093 .161. .043 250 39.2 90.0 27 4.9 O 6.84 .163 .073 .087 .058 284 82.4 98.0 28 9.8 0 6.87 .177 .091 .117 .037 287 172.0 124.6 29 0 .43 7.06 .311 .134 .150 .054 288 4.0 92.0 30 2.4 .43 7.12 .223 .116 .114 .054 245 40.0 108.8 31 4.9 .43 7.16 .191 .119 .133 .058 247 88.8 122.0 32 9.8 .43 7.18 .216 .109 .097 .055 255 173.6 136.0 LSD 05- .17 -- -- -- -- 44 10.7 8.9 *pH treatment. Treatments 1-8 from plot 36, 9-16 from plot 42, 17-24 from plot 25, and 25-32 from plot 41. 121 The variation in yields of the grass grown on soils from different plot areas prevents formation of any definite conclusions. However, two trends were apparent, which indicated the possible importance of soil reaction. One very striking observation was the degree of arsenic toxicity on the turfgrass grown at the high pH level. While increasing arsenic caused some reduction of growth, the magnitude was not great. In contrast, in- creasing arsenic at the low pH level caused substantial yield decreases. Another important observation was the change in arsenic soil test results with pH. Extractable arsenic levels tended to increase from the pH 4.78 level to the pH 5.16 level and then decrease from the pH 6.19 to 6.74 levels. The intermediate levels were similar. The above results indicated that arsenic action in soil was influenced by soil reaction. While the yield data did not allow fair comparisons to be made, the mag- nitude of this interaction appeared to be substantial compared to, for example, the arsenic-phosphorus inter- action. Variation in the chemistry of the soils used in this experiment precludes any meaningful discussion of the soil test data. 122 Greenhouse Experiment 6 The data in Greenhouse Experiment 5 indicated the possibility of a substantial arsenic-soil reaction rela- tionship. The elucidation of such an interaction could be of major importance in explaining some of the varia- bility in arsenate control of annual bluegrass on golf courses. The literature on the subject is scant and not in agreement. Kerr (58) recommended applying lime to greens or fairways if pH was below 6.0. He stated that arsenicals are less available at low pH or at pH above 7.8. However, other authors reported arsenate to be more toxic at lower pH (23, 38, 52) on other plants. Freeborg (39) surveyed several golf courses and found no relation- ship between pH and control of annual bluegrass with arsenate. Soil texture has been recognized as having an effect on arsenic toxicity (19, 22, 25, 30, 32, 33, 39, 45, 56, 65, 84, 89, 95, 98). The magnitude of this effect with respect to P93 EEBEE.L° control has not been investi- gated in detail. The objectives of the present investigation were: 1) to study the effects of soil reaction on arsenic toxi- city to annual bluegrass and on the arsenic soil test, and 123 2) to study the magnitude of the interaction between tex- ture and arsenic level with respect to Poa annua L. growth and the arsenic soil test. Four mineral soils ranging in texture from a loamy sand to a silty clay loam and one peat soil were chosen. All soils were initially acid in pH (4.28 to 5.15) and low to medium in phosphorus levels. A portion of each soil was limed with a 1:1 mixture of Ca(OH)2 and Mg(OH)2 up to about pH 6.5 and another portion to about pH 7.5. Very fine powdered reagent grade Ca(OH)2 and Mg(OH)2 were used and each soil incubated for 2 months. Grass was then seeded, and arsenic treatments were applied on 3-17, 7 weeks after seeding. Tables 21, 22, 23, 24, and 25 contain the clipping weight and arsenic soil test data for the loamy sand, sandy loam, loam, silty clay loam, and peat soils, re- spectively. The analyses of variance for the data in these tables are in Table 34, Appendix. Table 35, Ap- pendix, lists correlation coefficients between clipping yields by date and arsenic soil tests for the various soils. 124 mo. O.N h.OH hfio. OmO. MOO. th. OmO. HmO. OmO. mmO. OVO. HvO. " DOA 0.0m N.OHH NNO. NOO. ONO. HOH. NNH. OOH. moo. OOO. OmO. llll 0.0 ON.h O o.vN v.0v OmO. OOH. NNH. OOH. hNH. ONH. ONO. mOO. OmO. Illl O.v hN.h O O.VH 0.0 hOH. mON. Ohm. NHN. mvN. HON. MHH. mHH. NFO. HmH. O Om.h h o.vm O.hHH mHO. ONO. who. OmO. HOO. ONH. MOO. hOH. who. III! 0.0 ON.O O O.Nm O.Nm MOO. mOH. OOH. HMH. OOH. NOH. OOO. OVH. ONO. llll O.v mv.® m o.vH O.MH OOH. OON. mwm. Omm. HON. OON. hOH. ONH. ONO. OmH. O Om.w V O.Nv v.OmH OVO. HOO. moo. who. th. HBO. MVO. ONO. hfiO. llll 0.0 ON.m m O.©N o.vm NOO. mmH. HHH. hOO. ONH. HMH. va. VOO. OOO. illl O.¢ ¢H.m N O.NH ¢.O OHN. mON. va. OOH. HhH. mvH. HOO. OOO. ONO. HmH. O ¢O.v H mm ma 1 a u n .. 7... : n, u .. .. .s mN m OH O HH m v m 5N Q ON v HH fi v v ON m NH m N OOH Em HOQEDZ 2mm 2mm Mom aaom uswfiummua moommz am mmum Auom\mv unmaoz mcammaau . €09 ox . m ucofiaummxm mmsoncwmuw Mow .Aoumm vasomndlmmlmuduxmeo mumo ummu aaom oacmmum paw unmam3 ocammaau .Ucmm NEMOAII.aN mamda 125 m.m h.oa ovo. omo. moo: .hvo. omo. amo.r mmo. -mmo. .wvon avo. ama o.ma 0.0m ohm. mmv. mav. wvv. omm. oov. mam. vmm. mom. IIII m.m mm.n m 0.0a v.om mom. mmv. mmv. mmv. mov. mom. ovw. omm. mom. IIII m.v mv.o w 0.6 o.m mom. ooo. mmm. mom. mam. va. omm. mom. mom. mam. o Nv.o o o.mm N.ooa mam. amv. 5mm. vmv. mvm. Nmm. mmm. mvm. mom. IIII m.m mo.o o o.om ©.>m mmm. mov. mom. mav. vmv. mom. mom. aom. mam. IIII m.v mo.o m o.ma o.m mav. ooo. mmm. mom. omm. mov. ovm. mmv. mom. Nmm. o vo.o v 0.0m v.oma maa. vvm. mmm. ohm. wvm. 0mm. maa. omm. avm. IIII m.m mm.v m o.mm v.vm mma. 0mm. nvm. aam. vwm. mam. moa. mom. omm. IIII m.v om.v m o.m N.m mvm. ovo. vmm. mmv. mmv. mom. boa. mov. mom. Nov. 0 mm.v a m m I I I I I I I I I I E d m mm m ma m aa m v m om v om v aa v v v mm m 5a m m ooa mm HonEdz m Saw a 2mm Hmm aaom unweummue Oommz m wmum Auom\mv unmamz mcammaau «09 mm . o ucmfiaummxm mmsoncmmuw Mow .Aoumm oacmmumlmmeHsuxmeo mumo umwu aaOm oasmmum new munmam3 mcammaau .EMOa hocmmII.NN mammfi 126 mo. m.m m.oa ovo. omo. mmo. ovo. omo. amo. omo. mmo. mvo. avo. u omq o.mm o.om vow. ooa. mom. mom. mom. oav. aom. vmm. ooa. IIII m.o om.o o o.va m.~m mmm. amm. mmv. mam. mam. mom. oom. mmm. mvm. IIII o.v mo.o m o.v m.v omv. mom. ovv. mmm. mam. oom. nan. omm. mma. oam. o mm.o o o.ov m.ao omo. omo. omo. omo. moo. mma. moo. oma. vma. IIII m.o ma.m m o.mm m.av amo. ovo. ooo. amo. mma. «ea. omo. mma. voa. IIII o.v mm.m m o.m o.v mov. aam. emv. omo. ooo. mvm. oma. omo. omo. mam. o mm.m v o.mm m.mma oao. mao. mmo. moo. mmo.. moo. mao. omo. mma. IIII m.o am.v m o.m~ m.am omo. mao. omo. amo. vmo. moo. moo. vmo. Nma. IIII o.v mm.v m o.m o.v aov. mom. omv. oam. mmm. mmm. maa. mvm. ama. omo. o ma.m a m m I I I I ..I . .r .-I I I . I E a a mm m ma m aa m v m om v on v aa v v v mm m oa m m ooa on no 2 m 2mm a 2am I umm aaom bumsmmwue oommz o omum Auoo\oo unoams ocaomaao mos ox . ImmImHDuxmBV w ucmfiaummxm mmsoscmmuw sow memo 9mm» aaOm oacmmum .Awumm cacomum 6cm unmamz mcammaao .EmOAII.mN mam<8 127 mo. m.oa ovo. omo. mmo. evo. omo. amo. moo. mmo. mvo. avo. omq m.om vea. mom. mov. ama. oaa. oma. moo. ama. ooa. IIII m.o vm.o o m.om mam. mam. mmv. Ham. aoa. omo. moa. mma. moa. IIII o.v ov.o m m.v mvm. oam. omv. mam. avm. omo. mma. mma. voa. vvm. o om.o o m.oN mmo. ama. vmm. vma. mom. mma. mma. oma. moa. IIII m.o am.m m v.vm boa. amm. mom. ova. oma. mma. moa. mma. moa. IIII o.v ma.m m o.v mom. aom. mom. moa. mma. boa. vma. «ma. oma. ooo. o ma.m v v.vo vmo. voa. ovm. mma. mma. mma. voo. ooa. oma. IIII m.o vm.v m m.mv amo. baa. omo. oma. mma. oma. mmo. oma. moa. IIII o.v am.v N m.m moa. mom. Hem. oea. oma. mma. mma. oma. mom. mam. o ae.v a m I I I I I I I I I I . E saw mm m ma m Ha m v m om v om v aa v v v mm m oa m m ume no amassz H GO NON o ammo luooxoo usoamz ocaomaao mos ox a.om 0 Eu 9 .Awumm oacmmumlmmlmusuxmav m ucmaauwmxm omsoccooHO How mumm ummu aaom oacomum ocm unmams moammaao .amoa hmau huaamII.vm mamma 128 m.oa ovo. omo. omo. ovo. amo. amo. mmo. moo. mvo. avo. oma m.vm mvo. omo. ooo. oma. mma. ova. ooa. vma. omo. IIII m.o ma.o o m.mv amo. omo. aom. moo. mma. ama. oma. oma. mmo. IIII o.v oa.o m m.v moo. vma. mom. vom. oma. moa. mma. mom. mom. mam. o va.o o m.ava ooo. ooo. avo. amo. omo. maa. ovo. moo. vma. IIII m.o ao.m m m.vm ooo. ooo. moo. boo. moo. ova. mmo. moa. mam. IIII o.v oo.m m m.o omo. ooa. ooo. moa. moa. mvm. moa. ooa. mmm. mmm. o oo.m v v.mvo ooo. ooo. ooo. ooo. ooo. ooo. ooo. vao. ooo. IIII m.o mv.v m v.oma ooo. ooo. ooo. ooo. ooo. mao. moo. omo. moo. IIII o.v mm.v m o.v vmo. omo. mom. omo. boa. mam. oaa. ama. omo. mmm. o mv.v a m I I I I I I I I I I E a mm m ma m aa m v m on v on v aa v v v mm m oa m m ooa mm uwnssz 2mm mom ao unmaummua m ovum luoo\oo unoams ocaomaao «o9 ox a. m ImmImHsuxmBO o unmEaummxm mononcmmuw Mom mumm ummu aaom oacmmum .Amumm oacmmud mam unmam3 maammaau . umwmll . mN W342... 129 Loamy Sand Soil pH significantly affected the growth of Pea annua_L. Not all dates exhibited a significant difference between pH levels, but the trend was for increased growth between low and intermediate pH levels with little differ- ence between intermediate and high pH. In general, arsenic had a significant effect on yield, especially on the clipping dates after 4-11. In- creasing arsenic reduced the yields irrespective of pH level. Table 35, Appendix, lists the correlation coeffi- cients for clipping yield by date against the arsenic soil test results. Applied arsenic significantly increased the Bray P arsenic soil test values at all pH levels. The corre- 1 lation coefficient (r) for applied arsenic against Bray Pl arsenic was 0.96. This reflects an average over pH levels. The NaHCO3 soil test results were also signifi- cantly influenced by arsenic. At the low and high pH levels, increasing arsenic significantly increased the soil test values. However, no significant difference occurred between the 4.9 and 9.8 kg TCA/100 m2 rates at 130 the intermediate pH. The r value for NaHCO3 versus ap- plied arsenic over all pH levels was 0.85. Significant differences in growth due to arsenic- pH interactions were apparent on some dates. There was a slight tendency for increasing arsenic to be more toxic at the intermediate pH. Arsenic—pH interactions also significantly affected the Bray P and NaHCO arsenic soil tests. The Bray P l 3 l arsenic soil test results were significantly higher at the low pH level for both applied arsenic rates (except 0 kg TCA/100 m2). No differences between the intermediate and high pH were apparent. While significant differences occurred between pH levels on the NaHCO arsenic soil test, 3 the pattern was not consistent. Sandproam Growth of Poa annua L. was significantly affected by soil pH. The intermediate and high pH levels produced greater yields than the low pH. A decrease in growth be- tween the intermediate and high pH levels occurred on several clipping dates when no arsenate was applied. 131 Arsenic significantly affected yields at all pH levels. Increasing arsenic reduced clipping weights. Significant increases in Bray Pl extractable arsenic with increasing arsenic rate occurred at all pH levels. The r value was 0.94. At the low and intermediate pH levels, the NaHCO3 soil test significantly increased with arsenic rates. No significant difference occurred at the high pH level between the two arsenic rates. Applied arsenic and NaHCO3 soil test results had a r = 0.70. Yield was significantly affected by a pH—arsenic interaction. As pH increased arsenic toxicity at a par- ticular arsenic rate generally decreased relative to the control. Thus, arsenic exhibited the greatest degree of control of annual bluegrass at the low pH level. Though some control occurred at the high pH, the growth of the turf even at high arsenic rates was still very good. Soil test results from both extractants were sig- nificantly affected by pH—arsenic interactions. No sig- nificant difference at the 4.9 kg TCA/100 m2 rate between pH levels occurred with the Bray P soil test. However, 1 at the 9.8 kg TCA/100 m2 rate the Bray Pl arsenic soil test significantly decreased with increasing pH. The 132 NaHCO3 soil test values significantly decreased with in- creasing pH at all arsenic rates. Loam Soil reaction significantly affected the growth of annual bluegrass, but differences on some dates were not significant. In general, the low pH level produced the least growth, while the intermediate and high pH levels had similar clipping yields when no arsenic was applied. Arsenic had a significant effect on clipping weights at all pH levels, with increasing arsenic signif- icantly reducing growth in most cases. The Bray P extractable arsenic fraction was sig- 1 nificantly increased by applied arsenic with a r value of 0.90. Similar results occurred with the NaHCO3 soil tests, with a r value of 0.85. Soil reaction-arsenic interactions with respect to yield were highly significant. Increasing arsenic signif- icantly decreased the growth of annual bluegrass at the low and intermediate pH levels on every date after 3-28. However, the magnitude of the response was considerably greater at the low pH. At the high pH significant control did not occur until the last 2 clipping dates. 133 The arsenic soil test results were affected by the pH-arsenic interaction. As pH increased the Bray P1 ex- tractable arsenic decreased. The trend was most apparent at the high arsenic rate. With NaHCO3 extractable arsenic, no significant difference at the 4.9 kg TCA/100 m2 rate occurred between the low and intermediate pH levels. All other arsenic soil test results significantly decreased with increased pH. Silty Clay Loam Clipping weights were significantly affected by pH. While differences on all dates were not significant, the high pH level generally produced yields significantly greater than the low pH at the zero arsenic rate. The intermediate pH exhibited a growth rate between the high and low pH levels. Arsenic significantly affected the growth of P23 annua L. Increasing arsenic tended to decrease growth but differences depended on date and pH. Increasing arsenic rate significantly reduced Bray Pl extractable arsenic. Applied arsenic and extrac- table arsenic had a correlation coefficient of 0.94. The NaHCO3 extractable arsenic was not determined on this soil. 134 The degree of control due to arsenic rate was sig- nificantly influenced by soil pH. A significant degree of control was achieved at the low pH level by 4-27. No significant decrease in growth with increasing arsenic was apparent at the intermediate and high pH levels until 5-18 and 5—25, respectively. Even on 5-25 at the high pH, 9.8 kg TCA/100 m2 reduced growth by only 30%. Soil reaction had a significant effect on Bray Pl extractable arsenic. With increasing pH, the arsenic soil test results declined, although there was no significant difference on the intermediate and high pH soils at the 4.9 kg TCA/100 m2 rate. At the 9.8 kg TCA/100 102 rate, differences were significant at all pH levels. Peat Soil reaction significantly affected clipping yields. No significant trend occurred until 4-27, after which the growth rate at the low pH was generally less than at the intermediate and high pH levels when no arsenic was added. No significant difference between growth at the intermediate and high pH levels was apparent. Growth declined after 5-18 due to suspected micronutrient deficiencies. 135 Increasing arsenic significantly decreased clip- ping weights. The magnitude of the response was dependent upon date and pH. Bray P extractable arsenic was significantly 1 affected by applied arsenic, irrespective of pH. The r value was 0.79. A significant pH-arsenic interaction was very apparent with respect to growth. At the low pH level ap- plied arsenic reduced growth within 2 days of application. This is demonstrated by the drastic decline in clipping weights on 3-28 only 11 days after receiving arsenic. The arsenic toxicity was also exhibited by wilting of the turf at low pH even at field capacity, which suggests damage to the root system or crown tissue at the soil surface. At the intermediate pH level significant de- creases in growth also occurred by 3-28, but the magni- tude was much less severe. No significant reduction in growth was apparent at the high pH until 5-4. Bray P arsenic soil test results were quite sig- 1 nificantly affected by a pH-arsenic interaction. At both levels of applied arsenic an increase in pH markedly re- duced extractable arsenic. The magnitude of this reduction was 50-60 percent of the original test between pH levels. 136 Comparison of Soils Soil type was a significant factor in many of the responses observed in this experiment. Yield data indi- cated that considerable differences in growth rates of Poa annua L. occurred between soil types. The total clip- ping weights averaged over all three pH levels (Treat- ments 1, 4, and 7) for the loamy sand, sandy loam, loam, silty clay loam, and peat soils were 1.91, 4.71, 2.90, 2.23, and 1.97 g, respectively. Soil type-pH interactions were significant with respect to yield. The loamy sand, loam, and peat soils responded similarly to pH. Each showed an increase in growth between the low and intermediate pH levels with no difference between intermediate and high pH. The total clipping weights for each of these soils at the low, intermediate, and high pH levels (Treatments 1, 4, and 7) were 1.71, 2.05, and 1.98 g; 2.67, 2.97, and 3.05 g; and 1.67, 2.14, and 2.09 g for the loamy sand, loam, and peat, respectively. In contrast, the sandy loam soil exhibited a marked increase in yield between the low pH and the inter- mediate pH with total clipping weights of 4.22 and 5.12 g, 137 respectively. However, between the intermediate and high pH, growth declined to 4.80 g. The silty clay loam soil demonstrated an increase in yield only between the inter- mediate and high pH levels. Total clipping weights were 2.13, 2.15, and 2.42 g, respectively, for low, intermed- iate and high pH at the zero arsenic rate. Arsenic toxicity to annual bluegrass was signifi- cantly affected by soil type. The loam and peat soils produced the greatest response to arsenic, irrespective of pH. The sandy loam and silty clay loam soils exhibited the least toxicity, while the loamy sand was intermediate. Soil type—pH-arsenic interactions significantly affected clipping yields. The sandy loam, loam, silty clay loam, and peat soils demonstrated decreased arsenic toxicity with increasing pH. While the magnitude of the interaction varied with soil type, the trend was still evident regardless of arsenic rate. The loamy sand ex- hibited a somewhat different pattern. Arsenic toxicity was greatest at the intermediate pH level, while the toxicity at the low and high pH levels was less. Bray P extractable arsenic was significantly in— 1 fluenced by soil type. In general, arsenic soil tests on the mineral soils decreased as the sand content decreased. 138 Thus, Bray P extractable arsenic generally decreased, ir- l respective of pH level, on the order of loamy sand > sandy loam > loam > silty clay loam. This trend was most ap- parent at the 9.8 kg TCA/100 m2 arsenic rate. The peat soil had considerably higher Bray Pl arsenic soil tests than the mineral soils. While NaHCO3 extractable arsenic was also signifi- cantly affected by soil type, the pattern was much less apparent. Large differences in arsenic soil test between arsenic rates and pH level prevented any statement on a meaningful trend due to soil type. Soil type-pH-arsenic interactions affected the Bray Pl soil test results. The loam, silty clay loam, and peat soils demonstrated decreased Bray Pl extractable arsenic with increasing soil pH. This was observed at both the 4.9 and 9.8 kg TCA/100 m2 arsenic rates. The sandy loam soil had a similar trend at the high arsenic rate, but little difference in arsenic soil tests occurred with increasing pH at the 4.9 kg TCA/100 m2 rate. On the loamy sand, arsenic soil tests tended to decline with an increase in pH at the 4.9 kg TCA/100 m2 rate. However, at the high arsenic rate a decrease occurred between the 139 low and intermediate pH levels, but not between the upper 2 pH levels. Discussion of Greenhouse Experiment 6 Differences in growth response of Pga_angga L. on the different soil types would be expected. Each soil is unique with reSpect to its chemical, physical, and bio- logical properties and because of this each soil is an unique growth medium. The relatively slow growth rate of annual blue- grass on the loamy sand soil at the earlier clipping dates was apparently due to inadequate soil phosphorus levels. Foliar applications of phosphorus resulted in enhanced growth. The growth of annual bluegrass was also slow on the peat soil, but at the later clipping dates. Manganese applied as a foliar spray produced limited recovery. There was a positive response in the growth of Poa annua L. with increasing pH. On every soil the low pH level produced the poorest growth at the zero arsenic rate. No difference in growth between the intermediate and high pH was apparent on 3 soils (loamy sand, loam, peat). On the sandy loam soil growth was best at the 140 intermediate pH and declined at the high pH, while on the silty clay loam growth was best at the high pH. Poa annua L. is reported to grow best between pH 5.5 and 6.5 (13, 14). Poor growth at low pH could reflect less favorable nutrient levels or balances. Also, alum- inum toxicity in the mineral soils is a possibility at the low pH levels used in this experiment. Any chemical or biological factor which is pH dependent could cause differences in responses of soils to pH changes. The degree of control exhibited by arsenic for annual bluegrass was markedly affected by soil type. All soils were analyzed for percent sand, silt, clay, and organic matter, as well as iron and aluminum (Table 3). Those soils which contained the highest levels of l N NH4OAc extractable aluminum were the least responsive to applied arsenic. Woolson, et al. (117) reported sim- ilar results using corn as an indicator crOp. They in- vestigated several soils, which varied considerably in l N NaOH extractable aluminum, and found that as aluminum content declined arsenic toxicity increased. Jacobs et a1. (45) reported that the ability of a soil to sorb arsenic was proportional to the A1203 and Fe203 contents. 141 Dorman et a1. (33) also found that soils highest in per- cent A1203 exhibited the least arsenic toxicity. There was no apparent relationship between citrate- dithionite-bicarbonate extractable iron and arsenic toxi- city. Woolson et al. (117) reported similar findings for oxalate extractable iron. There are several reports of arsenic-iron interactions in soils, but few relate iron content to arsenic toxicity. Albert (1, 2, 3), Cooper et al. (22, 23), and Dorman et a1. (33) reported that soils high in iron exhibited little toxic effect due to arsenic. However, the soils studied all had very high iron contents. Organic matter was indicated to have little effect on reduction of arsenic toxicity. Compared to other soils the peat soil exhibited the most toxic effects due to arsenic, especially at the low pH. Arsenic was applied on 3-17-72 and within 2 days the grass on the low pH (Treatments 2 and 3) pots exhibited wilting even though all pots were at field capacity. The plants then died as indicated by the reduced yields on subsequent clipping dates. The arsenic levels were apparently high enough to kill the root system or to kill the crown tissue at the soil surface. The arsenic soil test results were very 142 high at the low pH level. Boischot and Hebert (19) re- ported some fixation by humus but in no case was over 18% of the added arsenic was fixed. Other authors have found little correlation between organic matter and arsenic fixation (45, 47). In general, as the percent sand decreased and the percent silt increased, arsenic toxicity was reduced. Increasing clay was less well correlated to arsenic tox- icity. On Midwestern soils, the clay contents vary much less in magnitude than the silt or sand fractions. When correlating particle size fractions to arsenic toxicity the clay fraction may be less correlated than sand or silt due to relatively small changes in clay content between soils. This illustrates that a strong correlation between two factors, such as percent silt versus arsenic toxicity, is not always associated with a cause and effect relation- ship. In any case, texture was not as closely associated to a reduction in arsenic toxicity as was the aluminum content. Many investigators have observed that finer- textured soils generally inactivate more arsenic than lighter—textured soils (19, 22, 25, 30, 32, 33, 45, 56, 65, 84, 89, 95, 98). While texture has a definite 143 influence on arsenic toxicity, the relationship between these factors may be due to associated aluminum and iron. Response of annual bluegrass to arsenic varied not only with soil type but also with pH level for each soil type. All soils except the loamy sand followed the trend of decreased arsenic toxicity with increasing pH. Two possible mechanisms could account for this trend. The first is the possible influence of calcium on the action of tricalcium arsenate. A high calcium ion acti- vity may prevent the alteration of tricalcium arsenate into more soluble arsenate compounds, such as mono- or dicalcium arsenates. Second, the presence of any free calcium carbonate could result in precipitation of ar- senate onto its surface and subsequent reduction in arsenate activity. Several investigators have reported a decrease in arsenate activity due to CaCO (19, 22, 23, 3 65, 108). In the above study free CaCO could be impor- 3 tant at the high pH. However, at the intermediate pH little free CaCO3 would be expected after the incubation period. The loamy sand soil did not follow the pattern of decreased arsenic toxicity with increasing pH. Instead, yields were most affected by arsenic at the intermediate 144 pH even though arsenic soil test results indicated higher arsenic levels at the low pH. This soil required supple- mental phosphorus applied as a foliar spray. Such appli- cations may have upset the normal arsenate-phosphorus balance within the plant, causing an atypical reSponse. The Bray P1 and NaHCO3 soil arsenic tests were correlated with annual bluegrass growth (Table 35, Ap- pendix). The methods were compared on 3 soils used in this experiment. Correlations varied with date but tended to become similar at the later clipping dates. Only on the loamy sand was an appreciable difference apparent between extractants. The correlation coefficients between extractants for all arsenic rates and pH levels were 0.82, 0.85, 0.96, and 0.84 for the loamy sand, sandy loam, loam, and all 3 soils, respectively. The correlation coeffi- cients for the NaHCO3 extraction versus yield for the last clipping date varied from -0.73 to -0.83. Woolson et al. (116) reported an r of -0.82 for NaHCO3 extractable arsenic versus growth of 4-week old corn. The Bray Pl extractant was utilized on all soils. The correlation coefficients for the arsenic soil test against yield at the last clipping date varied from -0.74 to -0.84. However, the correlation coefficients averaged 145 over all soils with all arsenic and phosphorus levels were -0.63 and -0.56 for the four mineral soils and all soils, respectively. If the correlations were separated by pH a higher correlation would be expected. Jacobs et a1. (46) reported r values of -0.91, -0.88, -0.77, and -0.93 for Bray P extractable arsenic and yield of potatoes, 1 peas, snap beans, and sweet corn, respectively. The correlation coefficients of the two extrac- tants compared favorably on all 3 soils examined. How— ever, the quantity of arsenic extracted by NaHCO3 was considerably less than for the Bray Pl extractant. While the ppm-As soil values indicate that NaHCO extracts only 3 30-40% as much, the values are much lower in the extract which is analyzed. The Bray P extraction procedure has 1 a soil to extractant ratio of 1:8. The NaHCO3 procedure requires a 1:20 ratio. This results in low levels of arsenic, which must be read on a less accurate portion of the standard curve. For the soils investigated, all arsenic analyses were 2.8 ppm arsenic in the test solution or less. Thus, all determinations for the NaHCO3 analyses were made on the portion of the standard curve between 86-100 percent transmission. Even very small errors in reading the curve could produce serious errors. 146 Another factor which argues against use of the NaHCO3 extractant was the lack of reSponse to applied arsenic. This was especially apparent at the intermediate pH level on the loamy sand and the high pH level on the sandy loam. Very small differences occurred in arsenic soil tests between treatments even though twice as much arsenic was applied. Such small differences would make arsenic soil test results very difficult to interpret. Soil type significantly affected arsenic retention against the Bray Pl extractant. In general, the order of greatest retention was silty clay loam > loam > sandy loam > loamy sand > peat. This is not in agreement with the general order of arsenic toxicity indicated by yield of annual bluegrass, where the loam and peat soils ex- hibited the greatest toxicity. The above suggests that factors in addition to the Bray Pl extractable arsenic must be considered. Another factor closely associated with arsenic activity as indicated by Poa annua L. growth was extractable aluminum. 147 Greenhouse Experiment 7 Two forms of tricalcium arsenate are available, granular and powder. Both forms have been used in pre- viously discussed experiments, either applied prior to turf establishment or to a relatively mature turf stand. The purpose of the present experiment was to evaluate the action of both forms applied prior to seeding of Poa annua L. and to the mature turf (3-1eaf stage or older). Table 26 contains the clipping weight and soil test data. Table 36, Appendix, contains the analyses of variance for data in Table 26. The description of the soil used in the experiment is in Table 2. Arsenic form was not significantly related to clipping yields. Thus, little difference in arsenic toxicity to Poa annua L. occurred due to arsenic form. Applied arsenic significantly decreased yields on most clipping dates, irrespective of arsenic form. In general, significant differences did not occur be- tween the 0 and 4.9 kg TCA/100 m2 rates, but did between the 0 and 9.8 kg TCA/100 m2 rates. 148 .mndu musums on omaammm oacmmumIIz .GOHUMCHEHOO wam 0“ HOHHQ OOHHQQM UHCOmHMIIUmkc .HmaosmuoIIo .Hmo3omIIm« o.ma vma. maa. ooa. ooo. Noa. omo. ooo. moo. omo. aao. mac. n mo ooa o.vo ama. mmm. aam. mom. mom. IIII IIII IIII IIII IIII IIII o.o z o o o.mm ova. ovm. oam. mmm. omo. IIII IIII IIII IIII IIII IIII o.v E O o v.mo mam. mom. ooa. ooa. mma. ooo. oma. ovo. mao. moo. moo. o.o .O.m O o m.om mom. ooo. ovm. oma. mma. ooo. voa. ooo. omo. ooo. mao. o.v .w.m 0 m o.voa vma. mma. oom. mam. ooa. IIII IIII IIII IIII IIII IIII o.o z m m m.ao ooa. mmm. mmm. mmm. omo. IIII IIII IIII IIII IIII IIII o.v z m v N.mo oma. Noa. aam. mma. maa. moo. moa. ooo. omo. ooo. moo. o.o .0.m m m o.mm mvm. ovm. mvm. ooa. oma. maa. ooa. mmo. omo. moo. moo. o.v .o.m m N v.m vmm. mmm. oam. vmm. ohm. ooo. aov. ama. omo. mao. omo. o I a I I I I I I I .I I I I I 8 .mm mm mm 5 ma o oa m v m am m o m mm v ha v o v a v on m N ooa «« d «Show H z mom mo _ 2mm mm ucmaumoua luoo\oo muooams ooaooaao ooa ox msas Imumm oacwmudv .acoaumoaammd mo msaeIfiuom oasmmum o ucosaammxm monogamouo now mumo umou aaOm oacomum mom unoam3 mcammaaUII.m~ mamoe 149 Time of arsenic application was significant with respect to yields. Arsenic was applied to the mature turf on 5-12. The next 3 clipping dates demonstrated signifi- cant effects due to time of application, probably because the arsenic applied to the mature turf had not had suffi- cient time to exhibit toxicity. On 7-13 and 7-25 no sig- nificant differences were apparent between forms. By this period, the arsenic applied on 5—12 was producing signif- icant growth reduction. However, while not significant, arsenic applied to the mature turf appeared to cause a greater degree of toxicity on 7-25 than arsenic applied prior to seeding. This trend was present for both powdered and granular arsenic forms and may indicate a higher degree of arsenic activity due to recent arsenic application. Arsenic form significantly affected the Bray Pl extractable arsenic fraction. Both forms applied prior to germination produced similar arsenic soil test results at both arsenic rates. However, arsenic soil tests for the granular TCA applied to the mature turf were significantly less than for the powdered TCA applied to the mature turf. The powdered TCA has a considerably greater surface area and could be altered into more active arsenic forms more quickly than granular TCA. 150 The above suggests that powdered TCA may result in an initially high arsenic soil test, while granular TCA may not. Thus, if powdered TCA was applied just prior to a period when the turf is under stress conditions a much higher degree of toxicity than anticipated may occur. How- ever, yield data does not indicate any greater toxicity due to powdered TCA compared to granular. Increasing arsenic significantly increased the arsenic soil tests, regardless of arsenic form. The cor- relation coefficient between arsenic applied and the ar- senic soil test was 0.85. The marked increase in arsenic soil tests by the powdered TCA applied to the mature turf resulted in sig- nificant arsenic rate-arsenic form, arsenic rate-time of application, and arsenic rate-arsenic form-time of appli— cation interactions, as well as, a significant time of application effect on arsenic soil test results. Growth Chamber Experiment 1 Soil moisture has been reported to influence the degree of toxicity exhibited by calcium arsenate. Kerr (58) 151 recommends drainage of low areas prior to arsenical appli- cation on Poa annua L. The literature in this area is not abundant and is often conflicting (6, 37, 59, 89). In these reports of arsenic-water interactions, the source of arsenic is AsZO3 or the possibility of reducing condi- tions exist. Thus, it is difficult to differentiate be- tween effects due to arsenite (+3) or arsenate (+5). Arsenite has been reported to be much more toxic to plants (4, 21, 55). The current eXperiment was designed to study the effects of tricalcium arsenate on annual bluegrass under different moisture regimes. Moisture levels were brought daily to field capacity or lower to decrease the possi- bility of reducing conditions. Water rates were 16.7 (field capacity), 14.1, and 11.6% water. Tables 27 and 28 contain the yield, water consumption, and soil test data. Analyses of variance for Tables 27 and 28 are in Table 37, Appendix. Water rate significantly affected the growth rate of Poa annua L. The clipping yields from the high water rate were significantly higher than for the low water rate on every date. The clipping weights at the 14.1% water rate were intermediate between the 16.7 and 11.7% water 152 .muaommmo camam u oo.ma« ovo. Nmo. omo. omo. oao. mmo. mmo. Nmo. 0 mo ooa omo. omo. mmo. ovo. vmo. mvo. amo. moo. m.aa o.o o amo. ovo. amo. ovo. amo. omo. ooo. avo. m.aa o.v o omo. vvo. ovo. vmo. ovo. omo. moo. vvo. m.aa o o moo. mmo. omo. omo. voo. ooo. oaa. amo. a.va o.o m ooo. ooo. ooo. ooa. voo. aoo. ooa. mmo. a.va o.v m oma. moo. ooo. ooa. ooa. ooo. mma. voo. a.va o v mmo. mvo. ooo. aoa. ooa. maa. ova. ooa. o.oa o.o m moo. omo. moo. ooa. ooa. maa. ova. moa. o.oa o.v m ooa. voo. aoa. mma. oaa. maa. mva. maa. o.oa o a man omIo oan oIo mmIm oaIm aaIm amIm «0mm NEHWWa amnfioz Abom\ov munoam3 moammaao ucmoamm £09 om pcmEummHB .Aoumm o N mImumm oacmmudv a acmEaammxm amQEmnu £p3oau MOM munoaOB ocammaaUII.om mamme 153 .ouaomomo mamam u oo.mam . . . . . . . mo. a o m N m N m N o N o m o N u ooa N.om o.va m.Na o.ma m.oa m.oa m.oa m.aa o.o o o.vN v.ma o.ma v.ma m.oa v.oa N.aN m.aa o.v o v.N v.va v.ma o.ma m.oa m.oa v.oN m.aa o o o.vm o.oa o.ma m.oa o.oN o.aN m.vN a.va o.o m o.oN m.oa o.ma a.oa o.oN N.mN v.mN a.va o.v m N.m o.oa o.ma N.oN v.aN m.oN o.oN a.va o v v.oo m.aN a.oa m.oN m.oN m.oN m.am o.ma o.o m N.om N.aN m.oa m.aN o.oN o.oN v.Nm o.ma o.v N N.m v.NN m.oa m.oN m.oN o.mN m.om o.ma o a I I I I I I E mm aN m oN m oa m m m N m a m «on N uwma Hmoesz 2mm Auom\aEv soaumsdmcou umumz usmoumm moo ox usmEummuB .Amumm onImumm oacmmuav a ucmEaammxm umnsmnu suzoaw Mom mummy aaOm oacmmum cam ooaumESmcoo umumBII.oN mamde 154 levels. Significant differences between the intermediate rate and the other 2 rates varied with date. Arsenic had no significant effect on yields until the last 3 clipping dates. Significant reduction in growth due to arsenic occurred at the 16.7% water level on 7-18 and 8-12. The lack of a significant response on 7-27 could be due to a slow growth rate because of low nitrogen levels. Nitrogen was applied on 7-27 and the subsequent clipping date (8-12) exhibited a very marked decrease in yield due to arsenic. At the 14.1% water rate a significant reduction in growth due to arsenic occurred on 7-18. The decrease in growth on 8-12 between Treat- ments 4 and 6 were not significant. No significant dif- ferences in yields due to arsenic were observed on any date at the 11.6% water level. Arsenic-water interactions with respect to clip- ping yields were not significant on any date until 8-12. Water was apparently more of a limiting factor than ar- senic. On 8-12 yields were decreased markedly by arsenic at the high water rate. The lack of a significant arsenic- water interaction until 8-12 indicates that the length of the experiment should have been extended. 155 Arsenic had no significant effect on water con— sumption on any date. However, the data were obtained on dates where no arsenic-water interaction occurred with respect to growth. Water consumption data obtained on 8-12 or after may have indicated an arsenic-water inter- action with respect to water use. Applied water significantly influenced water con- sumption (transpiration + evaporation) on all dates. Decreasing applied water resulted in less water consump- tion, irrespective of arsenic rate. This trend was in agreement with yield data. With decreasing soil water contents growth would be expected to decrease and result in less water consumption. Increasing arsenic rates caused increased arsenic soil tests, regardless of the water rate. The correlation coefficient between applied arsenic and arsenic soil test was 0.97. Water rate significantly influenced the arsenic soil test. Arsenic soil tests were significantly higher at the 4.9 and 9.8 kg TCA/100 m2 rates at the 16.7% water level compared to the 14.1 and 11.6% water rates. No differences were apparent between the arsenic tests of the two drier soils. The higher water content at the 156 16.7% water level could cause increased conversion of tri- calcium arsenate into more active arsenic forms, resulting in higher arsenic soil tests. The higher arsenic soil tests at the 16.7% water rate were also apparent in clipping yield data. GENERAL DISCUSS ION Factors Affecting Germination and Growth of Selected Species In Greenhouse Experiments 1, 2, and 5 the effects of arsenic and phosphorus on germination of selected turf- grasses were investigated. Phosphorus and arsenic both reduced germination of Penncross bentgrass in Experiment 1. In this experiment ordinary superphosphate was used. The Penncross bentgrass was seeded immediately after applica- tion of the phosphorus and arsenic. The phosphorus-induced reduction in germination could be due to the presence of water-soluble fluorine in superphosphate, as suggested by Kinra et all (58a). The arsenic related reduction in germination could be due to the very high arsenic levels present in the unincubated soil even at the 4.9 kg TCA/ 100 m2 rate (about 200 ppm As). At these very high levels direct injury to the seed may occur. The arsenic and phosphorus treatments were incu- bated prior to turf establishment in Greenhouse Experi- ments 2 and 5. The phosphorus source was mono-calcium phosphate (reagent grade). Germination counts of Poa 157 158 annua L., Merion Kentucky bluegrass, and Penncross bent- grass in Greenhouse Experiment 2 and Poa annua L. in Greenhouse Experiment 5 were not affected by applied arsenic or phosphorus. Incubation reduced the Bray Pl extractable levels of these elements, which could reduce their toxic effects. Also, a phosphorus source free of fluoride was used which negated any possible influence of fluoride. Freeborg (39) found no reduction in 293 annua L. germination due to arsenic. Several soils and arsenic rates were used. All arsenic was incubated prior to seeding. Stapledon (102) also noted no reduction in germination of several grasses due to lead arsenate. In contrast, Juska (50) reported substantial reduction in germination of Poa annua L. and other grasses after ar- senate application. However, he counted only live plants and not the number of seeds which germinated. Phosphorus responses by several turfgrasses were observed in Greenhouse Experiments 1, 2, 3, and 4. In Greenhouse Experiment 1 Penncross bentgrass exhibited a strong positive response to the application of 0.22 kg P/100 m2 on a phosphorus deficient soil. A less dramatic but still significant increase in growth occurred between the 0.22 kg P/100 m2 and 3.44 kg P/100 m2 rates. On the 159 same soil, Poa annua L., Penncross bentgrass, Cohansey bentgrass, and Merion Kentucky bluegrass all exhibited significantly increased growth between the 0 and 1 kg P/100 m2 rates in Greenhouse Experiment 2. No signifi- cant response to phosphorus by any grass beyond 1 kg P/100 m2 was apparent. The soil used in Greenhouse Ex- periments l and 2 was a sandy loam soil very low in phos- phorus (11.2 lbs P/A). Thus, a growth response to phos- phorus would be expected. The soil used in Greenhouse Experiments 3 and 4 was a sandy loam soil collected from a long term phos- phorus study area. Soil was obtained from selected plots, which gave a range of high soil phosphorus levels (80 lb P/A to 178 1b P/A). Poa annua L. and Penncross bentgrass responded positively to phosphorus up to the 113 lb P/A level in Greenhouse Experiment 3. Merion exhibited in- creased growth up to 113 lb P/A then decreased in growth at 178 lb P/A. In Greenhouse Experiment 4, Poa annua L. demonstrated increased growth due to phosphorus on the early clipping dates, but the trend reversed on later dates. The magnitude of the phosphorus responses were much less in these experiments than in Greenhouse 160 Experiments 1 and 2. The high phosphorus levels in this soil would account for this observation. Even at very high levels, phosphorus generally is not reported to decrease yields (51). However, both Merion (Experiment 3) and Poa annua L. (Experiment 4) exhibited reduced growth at high phosphorus levels. A phosphorus-induced micronutrient deficiency, possibly iron, may have resulted in decreased growth. The soil used in these experiments was high in pH and phosphorus levels, both which are reported to induce iron defi- ciency (l4). Arsenic reduced the growth of the grasses used in all experiments. Numerous investigators have reported similar findings on grasses (21, 26, 27, 28, 32, 34, 36, 38, 39, 50, 51, 54, 63, 69, 73, 77, 87, 88, 96, 100, 101, 112). Response of grasses to arsenic varied with species. In Greenhouse Experiment 2 decreased growth due to arsenic application was on the order of Poa annua L. > Merion Kentucky bluegrass a Penncross bentgrass > Cohansey bent- grass. A similar trend for_Poa annua L., Penncross bent- grass, and Merion was observed in Greenhouse Experiment 3. 161 Selectivity of arsenic for Poa annua L. has been noted by other investigators (26, 27, 28, 36, 50, 52, 100). In general, an equivalent degree of control, based on a 50% growth reduction, was achieved at 50-70, 140-200, 140-200, and 300 ppm arsenic in the soil for Poa annua L., Penncross, Merion, and Cohansey, respectively, for the soil used in the above 2 experiments. A similar control level on annual bluegrass was observed at 30-60 ppm ar- senic in other experiments, depending on the soil. In Greenhouse Experiment 6 the growth of 393 agnga L. at 3 pH levels on 5 soils was investigated. On all soils growth was least at the low pH (4.28 to 5.15). Three of the soils exhibited no difference in growth of Poa annua L. between the intermediate pH (6.00 to 6.52) and high pH (7.15 to 7.82) levels. On one soil no dif- ference in growth between pH 4.71 and 6.18 occurred, but growth increased at pH 7.37. Another soil exhibited a substantial yield increase between pH 4.22 and 5.12, then a slight decline at the pH 7.42 level. Differences in the nutritional status of the soils at the 3 pH levels could explain the pH growth response. The best growth of annual bluegrass is reported to be between pH 5.5 to 6.5 162 (13, 14), which is generally supported by the data in this study. Soil Factors Affecting Arsenic The arsenic program suggested by Kerr (58) for gradual control of Poa annua L. takes into consideration such soil factors as moisture, acidity, presence of ex- cess CaCO phosphorus, organic matter, and soil type. 3: However, information on the relative importance of each factor is scarce. One purpose of the research reported in this thesis was to determine the degree to which some of these soil factors may influence the effectiveness of arsenic. Phosphorus Greenhouse Experiments 1, 2, 3, and 4 were de- signed to study arsenic-phosphorus interactions with respect to clipping yields and soil tests. In Greenhouse Experiment 1 the arsenic and phOSphorus treatments were applied at seeding time. Thus, the activities of these elements were markedly higher than in the later studies 163 where incubation was allowed. Some reduction in arsenic toxicity may have occurred between the 0 and 0.22 kg P/100 m2 rates. However, the grass on the low phosphorus treatments was quite phOSphorus deficient and may have masked any arsenic-phosphorus interactions. Only at the very high phosphorus rate (3.44 kg P/lOO m2) did a sig- nificant reduction in arsenic toxicity occur. The same low phosphorus soil was used in Green- house Experiment 2 except all treatments were incubated. Only at the high phosphorus level (4 kg P/100 m2) was a significant reduction in arsenic toxicity apparent for any of the grasses. Penncross bentgrass and Merion Ken- tucky bluegrass exhibited a greater positive response to phosphorus than Poa annua L. and Cohansey bentgrass. In Greenhouse Experiment 3 a soil with a range of high phosphorus levels was used. In this study Merion Kentucky bluegrass exhibited a significant reduction in arsenic toxicity as phosphorus soil test increased from the 80 lb P/A to 178 lb P/A level. Penncross bentgrass tended to respond positively to phosphorus, but the re— sponse was not significant. Poa annua L. showed no ten- dency toward less arsenic toxicity at higher phosphorus levels, which was also observed in Greenhouse Experiment 4. 164 While these studies indicated that phOSphorus may decrease arsenic toxicity, the magnitude of the interac- tion was considerably less than expected. Complete elimi- nation or Sparse use of phosphorus is commonly recommended when TCA is used for annual bluegrass control. Data from these experiments suggest that elimination of phosphorus from the fertilizer program may not be essential for arsenic control, especially at the rates normally applied (1 lb P205/1000 ft2 or less) to turf. The grasses which exhibited the greatest positive response to phosphorus were the ones which are often desired in place of £93 33223 L. Annual bluegrass, on the other hand, exhibited less reduction in arsenic toxicity when phosphorus was added. The results of the above experiments are not in agreement with previous studies on arsenic-phosphorus relations reported in the literature. Nutrient solution studies have provided ample evidence that phosphorus can counteract arsenic toxicity (21, 38, 41, 55, 87, 91), but they do not simulate soil solution conditions. Soil investigations on turfgrasses have also indicated signif- icant reduction in the effectiveness of arsenic due to phosphorus (16, 26, 27, 43, 51). However, in every case 165 these studies used very high applied phosphorus rates (5 lb PZOS/lOOO ft2 to 23 lb P205/1000 ftz). The highest phosphorus rate recommended for application to turf by Michigan State University is 4 lb P205/1000 ft2 on the most phosphorus deficient soils. Also, in some instances the investigators were using soils already very high in phosphorus. Thus, while phosphorus can reduce the ef— fectiveness of aqsenic, the magnitude of the response indicates that it may be of less importance than previ- ously suggested. Phosphorus did not influence Bray Pl extractable arsenic except at very high phosphorus levels in conjunc- tion with high arsenic levels. Under these conditions, phosphorus reduced arsenic soil tests to some extent (Greenhouse Experiment 1). Calcium phosphate and calcium arsenate were used. At these high levels the common ion effect due to calcium may reduce the solubility of TCA. At normally used rates of these elements no significant interaction on arsenic soil tests was observed in any experiment. 166 Soil Reaction and CaCO3 Greenhouse Experiment 6 was designed to investigate the possibility of an arsenic-soil reaction interaction. On four of the soils arsenic toxicity significantly de- creased with increasing pH as measured by clipping yields. The greatest response was generally between the low and intermediate pH levels. The Bray Pl arsenic soil tests on these soils followed the same trend. The fifth soil exhibited little significant difference in yields among pH levels even though arsenic soil tests decreased with increasing pH. However, this soil was apparently defi- cient in phOSphorus and required frequent foliar applica- tions for adequate growth.. This could alter the phosphorus- arsenic relationship within the plant. The reduced activity of arsenic with increasing pH may be due to a greater calcium activity at higher pH. The arsenic was applied as relatively insoluble tricalcium arsenate. The presence of calcium could decrease the solubility rate of TCA and reduce formation of more avail- able forms. Any excess calcium carbonate present in the soil could reduce the activity of arsenic by precipitation (19, 167 22, 23, 65, 71, 112). However, free CaCO3 would not be expected to be present at pH 6.0 to 6.74, which was the intermediate pH level for these soils. Also, the pH of the initially acid soils was altered by thorough mixing of a 1:1 ratio of find powdered Ca(OH)2 and Mg(OH)2. The soils were then incubated through several wetting and drying cycles and remixed periodically for 2 months. Grass was then established. Arsenate treatments were not added until 7 weeks after seeding. Thus, all soils had 15 weeks of near ideal conditions for alteration of ap- plied lime into exchangeable calcium (or magnesium). In acid soils the activity of calcium would be low. The TCA could be more easily degraded into more soluble arsenic forms, causing an increase in active or extractable arsenic. Recommendations for the use of TCA (58) state that arsenicals are less available at low pH and above pH 7.8. Application of lime for adjustment of pH to 6.0 is sug- gested on acid soils to make TCA more available. Also the presence of excess CaCO is suggested to cause a reduction 3 in water soluble arsenate. No recommendations for the ‘use of reduced rates of TCA at low pH are made. 168 Results of Greenhouse Experiment 6 indicate that pH is an important factor to consider when TCA is used for control of £22.22222 L. A higher level of toxicity can occur at pH 4.5-5.0 than with the same arsenic rate at pH 6.0-6.5. The data of this experiment and others also suggests that pH can have a greater effect on the degree of arsenic toxicity obtained at a given TCA rate than does phosphorus level. Texture, Iron, and Aluminum The effect of texture on the toxicity exhibited by arsenate was investigated in Greenhouse Experiment 6. ‘While percent sand, silt, and clay were all related to some degree with the general level of arsenic toxicity, the correlations are not always meaningful. The percent clay in many Midwestern soils is small (often less than 10%, especially on golf greens) compared to percent sand and percent silt. Thus, because the values for clay percentage generally are within a narrow range, they may correlate poorly with arsenic toxicity. However, percent sand or silt (which themselves are related to percent clay for a given soil) may exhibit a better correlation 169 because their values cover a wider range, and may be more accurately determined than percent clay. Freeborg (39) surveyed several golf course soils using arsenic and found that silt and sand contents were related to the degree of 223 EEBEE.L' control. At our present level of knowledge of the chemistry of arsenic in soils, such relationships have little basis. Thus, while they may correlate, a cause and effect relationship may not be present. The texture-arsenic toxicity relationship may be more empir- ical in nature. Finer-textured soils have generally been observed to exhibit a lesser degree of arsenic toxicity at a given arsenic rate than do coarser-textured soils. While this trend was apparent in Greenhouse Experiment 6, there were exceptions. The relative level of arsenic toxicity which de- veloped in a soil was quite closely related to l N NH4OAc extractable aluminum in Greenhouse Experiment 6. Similar findings were reported by other authors (33, 45). Citrate-dithionite-bicarbonate extractable Fe203 Showed no apparent relationship to arsenic toxicity as measured by clipping yields or Bray P extractable arsenic. Many 1 of the studies (1, 2, 3, 22, 23, 33) concerning iron 170 effects on arsenic toxicity have been conducted on high iron soils. Jacobs et a1. (45) reported that retention OAc or Bray P extractants was of arsenic against NH 1 4 related to sesquioxide content. However, whether A1203 or Fe203 or both was the active factor was not determined. The results of Greenhouse Experiment 6 indicate that soil type has an important effect on arsenic activity. However, soil texture was a less accurate measure of the ability of a soil to decrease arsenic activity than was aluminum. The fact that arsenic retention against Bray Pl extractant is related to aluminum content suggests that aluminum-arsenates are an active form of arsenic. The Bray P extractant attacks aluminum phosphates (44). A 1 similar mechanism for aluminum-arsenates may exist, which could contribute to the good correlation between Bray Pl extractable arsenic and plant yield. Organic Matter The limited data on organic matter-arsenic rela— tionships obtained in Greenhouse Experiment 6 indicate that organic matter has a relatively low ability to fix 171 arsenic. This is especially evident if weight of the peat soil is considered when reporting As (values in Table 25 are not adjusted for weight differences for field condi- tions). Similar results have been reported by other in- vestigators (19, 45, 47). Recommendations for use of TCA on Poa annua L. (58) suggest that soils low in organic matter may require less arsenic. The data from the above experiment and that of other investigators suggest it may not be necessary to adjust arsenic levels with respect to organic matter con- tent. Water Drainage of wet areas prior to the use of TCA is suggested (58). A higher degree of toxicity is generally observed in these sites. Limited useful information on arsenic-water inter- actions was obtained in Growth Chamber Experiment 1. How- ever, two trends were observed. One was that at field capacity Bray P extractable arsenic was higher than at 1 lower water levels. Another trend was that arsenic toxi- city, as exhibited by growth of Poa annua L., was greatest 172 at field capacity compared to drier conditions. However, the limiting factor in this experiment was water not arsenic. In the above experiment a water rate above field capacity would have been desirable. Physical Form of Arsenic Greenhouse Experiments 4 and 7 were designed to compare powdered and granular TCA forms. The powdered TCA produced higher Bray P arsenic soil tests and in general 1 high toxicity on Poa annua L. While powdered TCA resulted in higher initial arsenic soil tests, data in Greenhouse Experiment 7 indicated that with time both forms produced similar soil tests. However, the initially high arsenic levels due to powdered TCA could give a greater degree of control than desired soon after application. Time of Application The effect of TCA on control of Poa annua L. with respect to application time was studied in Greenhouse Ex- periment 7. No significant differences were observed between yields of the turf grown on pots which received arsenic prior to germination compared to pots which 173 received arsenic when the turf was mature. However, clip- ping yields on the 7-25 date were appreciably lower at similar arsenic rates on the pots which received arsenic to the mature turf. Continuation of the study beyond 7-25 would have clarified the results. Further studies would need to be conducted before any definite conclusions could be made. Feasibility of a Routine Arsenic Soil Test One of the initial objectives of this thesis re- search was to determine the feasibility of the development of a routine arsenic soil test. This required (1) inves- tigation of a suitable procedure with which to analyze for arsenic, (2) investigation of a suitable extractant for arsenic which would be well correlated to plant growth, and (3) investigation of soil factors which significantly influence the activity of arsenic with respect to plant growth. The distillation and colorimetric method of Small and McCants (97) has been sensitive, consistent, and relatively free from interferences. Other authors have 174 reported similar results (45, 46, 47, 115, 116, 117). A selective solvent and colorimetric method (80) was also investigated but this proved to be unreliable. A correlation study between the procedure of Small and McCants (97) and an atomic absorption procedure fOl- lowed by Freeborg (39) indicated a correlation coeffi- cient (r) of 0179. However, at the lower arsenic levels the correlation was less. On sub-samples of the same Bray P1 extract the atomic absorption procedure always indicated lower arsenic levels than the distillation procedure (Table 6). This could result from a lack of sensitivity. Another reason could be the problem of determining arsenic directly in the Bray P soil extract, 1 since the possibility of interferences due to substances in the extract exist. The distillation procedure selec- tively removes arsenic from the extract prior to analyses. The Bray P1 soil extractant was well correlated to applied arsenic with r values of about 0.90 (except where arsenic form was a variable). This indicates that the Bray P extractant is highly responsive to applied 1 TCA. The correlation of Bray P1 extractable arsenic versus Poa annua L. clipping weights by date on five soils 175 was investigated in Greenhouse Experiment 6. The corre- lation coefficients varied from -0.74 to -0.84 on the last date (Table 35). Similar values have been reported by other investigators (46) on other cr0ps. These values reflect all 3 pH levels used in this experiment. Better correlation would be expected if the pH effect was elim- inated. The r values for all mineral soils and all soils were -0.63 and 6.057, reSpectively. This indicates that while the Bray P extractant correlates quite well on one 1 soil, the correlation is less when other soil factors (i.e. soil type, pH) are considered. Thus, while Bray P1 extractable arsenic may determine whether arsenic soil levels will be toxic or not to Egg appp§_L., use of the soil test alone is insufficient for predicting how much afiSenic to apply for a given degree of control. Arsenic had a decided influence on the phosphorus analysis procedure. At phOSphorus levels normally encoun- tered in soils, arsenic caused increased phosphorus soil test results due to interference in color development. The degree of interference declined with increasing phos- phorus but was very significant at low phosphorus. Thus, 176 caution should be used when interpreting phosphorus tests on arsenic toxic soils. The studies conducted in this thesis suggest that important soil factors to consider when using TCA are soil reaction and extractable aluminum (or soil texture but with less reliability). Phosphorus and organic matter are reported to influence arsenic activity, but data from the experiments in this thesis indicate these to be of minor importance at normal phosphorus levels, especially when compared to soil reaction or soil type influences. However, short term effects from foliarly applied phos- phorus may be appreciable. Other factors may also influ— ence the degree of control achieved by TCA, such as drainage, temperature, light, time of application, and presence of stress conditions. The complexity of the arsenic—Poa annua L. rela- tionship precludes the development of a routine arsenic soil test with our present understanding. To obtain sufficient data to predict the amount of arsenic to apply for achievement of a given level of control of Poa annua L. would require an extensive set of field experiments under a wide variety of conditions. While the students reported in this thesis have been useful in determining some of 177 the soil factors influencing arsenic activity and their relative importance, further studies are needed to ascer- tain the quantitative influence of each on the arsenic soil test and selectivity for Poa annua L. CONCLUSIONS Methods The arsenic distillation procedure of Small and McCants (97) was sensitive, reproducible, and rela- tively free from interference. The Paul (80) arsenic determination procedure did not prove reliable. Phosphorus did not affect the Small and McCants (97) arsenic determination procedure. Bray Pl extractable arsenic was well correlated with applied arsenic and with growth of Poa annua L. and other grasses for any given soil. Increasing applied arsenic increased the levels of Bray Pl extractable arsenic and reduced plant growth. Arsenic extractable with 0.5 M NaHCO3 was correlated with applied arsenic and plant growth, but to a lesser degree than the Bray Pl extractant. The 0.5 M NaHCO3 178 II. 179 extractant was found to be less acceptable as an arsenic extractant compared to Bray Pl' Arsenic interfered with phosphorus determination, especially at levels of phosphorus normally encoun- tered in soils. Growth No effect on germination of Poa annua L., Penncross bentgrass, or Merion Kentucky bluegrass was observed when arsenic was mixed with the soil and incubated prior to establishment. This was consistent even at the extremely high arsenic rate of 19.5 kg TCA/100 m2 (with a Bray P soil test of 800 lb As/A). Without l incubation arsenic levels of 400 lb As/A decreased germination of Penncross bentgrass. Phosphorus, as mono-calcium phosphate, did not affect the germination of Poa annua L., Penncross bentgrass, or Merion Kentucky bluegrass at any rate. The highest soil phosphorus tests in this study were 556 lb P/A (4 kg P applied 100 m2) on one soil and 178 lb P/A soil test on another. PhOSphorus, as ordinary 10. 180 superphosphate, reduced germination of Penncross bent- grass when applied at the seeding time on a soil very low in natural phOSphorus. Arsenic reduced the growth of all grasses used in the various experiments. A definite selectivity of arsenic for Poa annua L. was found. Bray Pl arsenic soil test levels which reduced clipping yields by approximately 50% were 50-70, l40-200, l40-200,and > 300 ppm soil arsenic for Poa annua L., Penncross bentgrass, Merion Kentucky bluegrass, and Cohansey bentgrass, respec- tively. Penncross bentgrass exhibited a slight positive growth response to phosphorus up to a Bray P phosphorus soil 1 test of 804 lb P/A (3.44 kg applied P/lOO m2) in one experiment. In a second experiment Penncross bent- grass, Poa annua L., Cohansey bentgrass, and Merion. Kentucky bluegrass demonstrated increased growth up to a phosphorus soil test of 130 lb P/A (1 kg applied P/100 m2). However, the increases in growth were small in magnitude. In a third similar experiment the same trend occurred up to 113 lb P/A soil test for Penncross bentgrass, Poa annua L., and Merion Kentucky bluegrass. ll. 12. III. 13. 14. 181 High phosphorus levels (> 180 lb Bray Pl extractable P/acre) tended to reduce arsenic toxicity. However, the magnitude of the effect was not great. Control of Poa annua L. by arsenic was less affected by in- creasing phosphorus than Penncross bentgrass, Cohansey bentgrass, and Merion Kentucky bluegrass. Soil reaction significantly affected the growth rate of Poa annua L. on all five soils used in Greenhouse Experiment 6. In every case, growth was best in the pH 6.00 to 7.69 range compared to pH 4.28 to 5.15. Soil Analysis Phosphorus did not influence the level of Bray Pl ex- tractable arsenic, except at extremely high phosphorus (> 400 ppm Bray P extractable phosphorus) and arsenic 1 levels (> 1000 ppm Bray Pl extractable arsenic) together. Soil reaction significantly affected arsenic activity in soils as measured by Bray Pl extractable arsenic and plant growth. In general, increasing pH reduced ar- senic activity in the pH range from 4.28-5.15 to 7.14 to 7.82. The greatest decrease in arsenic toxicity 15. 16. 17. 18. 19. 182 occurred between the low and intermediate pH levels. The magnitude of this effect with respect to plant growth was greater than the phosphorus-arsenic inter- action. Soil texture was correlated with Bray Pl extractable arsenic and plant yield data. However, the correla- tion was lower than that exhibited by extractable aluminum. Soils with the highest 1 N NH OAc (pH 4.8) extractable 4 aluminum exhibited the lowest arsenic toxicity and Bray Pl extractable arsenic levels. Citrate-dithionite-bicarbonate extractable iron was not correlated with arsenic toxicity or Bray Pl ex- tractable arsenic. The peat soil used in Greenhouse Experiment 6 demon- strated little arsenic retention against the Bray P1 ex- tractant. Plant yield data indicated similar results. Bray P extractable arsenic levels were highest on a 1 soil maintained at field capacity, compared to the same soil maintained at 85% and 70% of field capacity. APPENDIX 183 TABLE 29.-~Analyses of variance for data in Table 9, Greenhouse Experiment 1. Source d.f. Mean Square F Approximate Significance Clipping Weights 9-ll A(P) 5 .0115 4.703 *** B(AS) 4 .1040 42.518 *** AB 20 .0049 2.001 * Error 60 .0024 Clipping Weights 9-20 A 5 .0062 7.567 *** B 4 .0891 109.503 *** AB 20 .0042 5.214 *** Error 60 .0008 Clipping Weights ll-7 A 5 .0544 16.574 *4. B 4 .3015 91.810 4.. AB 20 .0160 4.884 *** Error 60 .0033 Germination 8-24 A 5 64944.2070 16.447 *** B 4 162770.3170 41.223 *** AB 20 5956.5570 1.509 n.s. Error 60 3948.5780 Phosphorus Soil Test A 5 283345.6811 533.707 *** B 4 29046.7484 54.712 *** AB 20 1394.7548 2.627 ** Error 60 530.9013 184 Table 29.--Cont. Source d.f. Mean Square F Approximate Significance Arsenic Soil Test A 5 79552.9 9.344 *** B 4 17767159.4 2086.856 *** AB 20 29337.2 3.446 *** Error 60 8513.8 pH Determination A 5 2.2412 488.161 *** B 4 0.0424 9.240 *** AB 20 0.0357 7.772 *** Error 60 0.0046 185 TABLE 30.--Analyses of variance for data in Tables 10, ll, 12, 13, and 14 of Greenhouse Experiment 2. Source d.f. Mean Square F Approximate Significance Clipping weight 8-1 A(P) 3 .0151 45.775 *** B(As) 4 .0206 62.297 *** AB 12 .0024 7.170 *** C(Grass) 3 .0838 253.661 *** AC 9 .0064 19.338 *** BC 12 .0101 30.667 *** ABC 36 .0009 2.596 *** Error 160 .0003 Clipping weight 8-6 A 3 .0570 114.576 *** B 4 .0503 101.162 *** AB 12 .0019 3.901 *** C 3 .1344 270.298 *** AC 9 .0112 22.473 *** BC 12 .0105 21.162 *** ABC 36 .0010 2.072 *** Error 160 .0005 Clipping weight 8-18 A 3 .0776 119.070 *** B 4 .0347 53.242 *** AB 12 .0013 2.014 * C 3 .2712 416.096 *** AC 9 .0155 23.842 *** BC 12 .0087 13.299 *** ABC 36 .0011 1.721 * Error 160 .0007 186 Table 30.--Cont. Source d.f. Mean Square F Approximate Significance Clipping weight 8-28 A 3 .4861 151.230 *** B 4 .2858 88.922 *** AB 12 .0055 1.699 n.s. C 3 .0644 20.034 *** AC . 9 .0204 6.346 *** BC 12 .0238 7.405 *** ABC 36 .0063 1.945 ** Error 160 .0032 Clipping weight 9—12 A 3 .9095 139.554 *** B 4 .7085 108.701 *** AB 12 .0213 3.262 *** C 3 .4108 63.037 *** AC 9 .0455 6.974 *** BC 12 .0776 11.907 *** ABC 36 .0115 1.770 ** Error 160 .0065 Germination 7-24 A 3 20151.13 4.616 ** B 4 972.35 0.223 n.s. AB 12 4279.98 0.980 n.s. C 3 9490004.85 2173.684 *** AC 9 8885.99 2.035 * BC 12 4724.47 1.082 n.s. ABC 36 3250.25 0.744 n.s. Error 160 4365.86 Table 30.--Cont. 187 Source d.f. Mean Square F Approximate Significance Water Use 8-25 A 3 20045.10 90.259 *** B 4 7265.10 32.713 *** AB 12 341.98 1.540 n.s. C 3 9121.22 41.071 *** AC 9 943.07 4.246 *** BC 12 819.13 3.688 *** ABC 36 320.73 1.444 n.s. Error 160 222.08 Arsenic Soil Test A 3 2631.91 29.84 *** B 4 1122597.06 12726.27 *** AB 12 2189.85 24.83 *** Error 220 88.21 Phosphorus Soil Test A 3 710669.71 9883.52 *** B 4 6973.82 96.99 *** (AB 12 1343.13 18.68 *** Error 220 71.91 pH Determination A. 3 2.57920 10279.70 *** B 4 0.05160 205.83 *** .AB 12 0.02580 102.75 *** Error 220 0.00025 188 TABLE 31.--Analyses of variance for data in Tables 15, 16, 17, and 18 of Greenhouse Experiment 3. Source d.f. Mean Square F Approximate Significance Clipping weight 4-16 A(As) 3 .0688 54.324 *** B(P) 3 .0207 16.310 *** AB 9 .0072 5.703 *** C(Grass) 2 .0324 25.603 *** AC 6 .0088 6.977 *** BC 6 .0035 2.741 * ABC 18 .0025 1.940 * Error 96 .0013 Clipping weight 4-26 A 3 .1990 63.883 *** B 3 .0224 7.189 *** .AB 9 .0144 4.627 *** C 2 .0166 5.329 ** .AC 6 .0152 4.880 *** BC 6 .0032 1.028 n.s. .ABC 18 .0082 2.646 *** Error 96 .0031 Clipping weight 5-7 A. 3 .2043 47.763 *** SB 3 .0205 4.789 ** AB 9 .0065 1.520 n.s. C 2 .0399 9.340 *** AC 6 .0119 2.784 * ZBC 6 .0065 1.529 n.s. ABC 18 .0048 1.118 n.s. Error 96 .0043 189 Source d.f. Mean Square F Approximate . Significance Clipping weight 5-21 A 3 .3682 96.192 *** B 3 .0297 7.768 *** AB 9 .0061 1.583 n.s. C 2 .0014 .366 n.s. AC 6 .0187 ,4.892 *** BC 6 .0057 1.491 n.s. ABC 18 .0072 1.897 * Error 96 .0038 Clipping weight 6-5 A 3 .3826 82.699 *** B 3 .0084 1.817 n.s. AB 9 .0081 1.753 n.s. C 2 .0416 9.000 *** AC 6 .0223 4.810 *** BC 6 .0135 2.910 * ABC 18 .0060 1.303 n.s. Error 96 .0046 Clipping weight 6-18 A 3 .5937 170.831 *** B 3 .0136 3.922 * AB 9 .0079 2.276 * C 2 .0118 3.400 * AC 6 .0198 5.689 *** BC 6 .0144 4.147 *** ABC 18 .0055 1.590 n.s. Error 96 .0035 Table 31.-~Cont. 190 Source d.f. Mean Square F Approximate Significance Clipping Weight 7-8 A 3 .9271 251.482 *** B 3 .0132 3.569 * AB 9 .0125 3.401 *** C 2 .5318 144.242 *** AC 6 .0371 10.063 *** BC 6 .0192 5.208 *** ABC 18 .0080 2.175 ** Error 96 .0037 Arsenic Soil Test A 3 399794.07 3195.28 *** B 3 998.44 7.98 *** AB 9 581.09 4.64 *** Error 128 125.12 Phosphorus Soil Test A. 3 9559.89 867.44 *** B 3 13183.44 1196.23 *** 4A3 9 150.33 13.64 *** Error 128 11.02 191 TABLE 32.--Ana1yses of variance for Table 19 of Greenhouse Experiment 4. Source d.f. Mean Square F Approximate Significance Clipping weight 5-21 A(As Form) 1 .00037 0.1129 n.s. B(As Rate) 2 .01913 5.7763 ** AB 2 .00543 1.6390 n.s. C(P Rate) 1 .11674 35.2523 *** AC 1 .00828 2.5007 n.s. BC 2 .00606 1.8299 n.s. ABC 2 .00290 0.8768 n.s. Error 24 .00331 Clipping weight 6-4 A 1 .02544 8.0525 ** B 2 .02592 8.2036 ** AB 2 .00731 2.3123 n.s. C 1 .01017 3.2182 n.s. AC 1 .00483 1.5289 n.s BC 2 .00382 1.2097 n.s. ABC 2 .00230 0.7280 n.s. Error 24 .00316 Clipping weight 6-19 A 1 .02035 9.3250 ** B 2 .02079 9.5267 *** AB 2 .00791 3.6238 * C 1 .00075 0.3423 n.s. AC 1 .00292 1.3360 n.s. BC 2 .01796 8.2277 ** ABC 2 .00167 0.7630 n.s. Error 24 .00218 Table 32.--Cont. 192 Source d.f. Mean Square F Approximate Significance Clipping weight 7-13 A 1 .00344 1.6891 n.s. B 2 .02933 14.3954 * * AB 2 .00109 0.5372 n.s. C 1 .00865 4.2447 * AC 1 .00009 0.0428 n.s. BC 2 .00321 1.5733 n.s. ABC 2 .00124 0.6095 n.s. Error 24 .00204 Clipping weight 7-27 A 1 .01699 4.7531 * B 2 .32631 91.3050 *** AB 2 .00437 1.2237 n.s. C 1 .10956 30.6564 *** AC 1 .00071 0.1990 n.s. BC 2 .00033 0.0910 n.s. ABC 2 .00021 0.0591 n.s. Error 24 .00357 Arsenic Soil Test A 1 13548.96 45.354 *** B 2 66411.00 222.305 *** AB 2 6472.05 21.665 *** C 1 100.00 0.335 n.s. AC 1 21.78 0.073 n.s. BC 2 213.81 0.716 n.s. ABC 2 30.10 0.101 n.s. Error 24 298.74 193 Table 32.--Cont. Source d.f. Mean Square F Approximate Significance Phosphorus Soil Test A 1 421.62 21.758 *** B 2 1077.19 55.589 *** AB 2 186.52 9.626 *** C 1 35746.20 1844.701 *** AC 1 0.07 0.004 n.s. BC 2 25.88 1.336 n.s. ABC 2 0.07 0.004 n.s. Error 24 19.38 194 TABLE 33.--Ana1yses of variance for data in Table 20, Greenhouse Experiment 5. Source d.f. Mean Square F Approximate Significance Germination 9-24 A(As) 3 479.56 .665 n.s B(P) 1 73.50 .102 n.s. AB 3 1213.50 1.683 n.s. C(pH) 3 2485.44 3.446 * AC 9 1277.00 1.771 n.s. BC 3 431.17 .598 n.s. ABC 9 648.50 .899 n.s. Error 64 721.25 pH A 3 .3778 34.352 *** B 1 .9107 82.802 *** AB 3 .0020 .180 n.s. C 3 16.9830 1544.197 *** AC 9 .0572 5.202 *** BC 3 .1750 15.914 *** ABC 9 .0138 1.254 n.s. Error 64 .0110 Arsenic Soil Test A 3 200666.97 4663.06 *** B 1 40.56 .94 n.s. AB 3 18.92 .44 n.s. C 3 3515.40 81.69 *** AC 9 1134.98 26.37 *** BC 3 73.15 1.70 n.s. ABC 9 47.51 1.10 n.s. Error 64 43.03 Table 33.--Cont. 195 Source d.f. Mean Square F Approximate Significance PhOSphorus Soil Test A 3 14344.77 479.84 *** B 1 20627.21 688.84 *** AB 3 26.57 .89 n.s. C 3 18542.55 619.22 *** AC 9 235.93 7.88 *** BC 3 1306.94 43.64 *** ABC 9 87.47 2.92 ** Error 64 29.95 196 TABLE 34.--Ana1yses of variance for data in Tables 21, 22, 23, and 24 of Greenhouse Experiment 6. Source d.f. Mean Square F Approximate Significance Clipping weight 3-17 A(Soil) 4 .72074 824.63 *** B(pH) 2 .01420 16.22 *** AB 8 .01261 14.44 *** C(As) 2 .00001 .02 n.s. AC 8 .00001 .02 n.s. BC 4 .00001 .02 n.s. ABC 16 .00001 .02 n.s. Error 135 .00087 Clipping weight 3-28 A 4 .71786 621.88 *** B 2 .09454 81.90 *** AB 8 .01904 16.50 *** C 2 .02145 18.58 *** AC 8 .00506 4.38 *** BC 4 .01157 10.03 *** ABC 16 .00431 3.73 *** Error 135 .00115 Clipping weight 4-4 A 4 .36784 237.44 *** B 2 .07144 46.11 *** AB 8 .01528 9.86 *** C 2 .05635 36.38 *** AC 8 .00523 3.37 *** BC 4 .00840 5.42 *** .ABC 16 .00240 1.55 n.s. Error 135 .00155 Table 34.--Cont. 197 Source d.f. Mean Square F Approximate Significance Clipping weight 4-11 A 4 .13871 208.92 *** B 2 .11351 170.97 *** AB 8 .11842 27.74 *** C 2 .03527 53.12 *** AC 8 .00137 2.07 * BC 4 .00359 5.41 *** ABC 16 .00185 2.78 *** Error 135 .00066 Clipping weight 4-20 A 4 .30145 226.29 *** B 2 .18497 138.86 *** AB 8 .03139 23.57 *** C 2 .16209 121.68 *** AC 8 .00862 6.47 *** BC 4 .01587 11.92 *** ABC 16 .00655 4.92 *** Error 135 .00133 Clipping weight 4-27 A 4 .59132 470.65 *** B 2 .20013 159.29 *** AB 8 .01567 12.47 *** C 2 .17441 138.87 *** AC 8 .00511 4.07 *** BC 4 .01505 11.98 *** {ABC 16 .00776 6.18 *** Error 135 .00126 Table 34.--Cont. 198 Source d.f. Mean Square F Approximate Significance Clipping weight 5-4 A 4 .65751 591.07 *** B 2 .10638 95.63 *** AB 8 .01347 12.11 *** C 2 .23226 208.79 *** AC 8 .00852 7.66 *** BC 4 .00860 7.73 *** ABC 16 .00262 2.36 ** Error 135 .00111 Clipping weight 5-11 A 4 .41816 118.44 *** B 2 .31718 89.84 *** AB 8 .03466 9.82 *** C 2 .77271 218.86 *** AC 8 .03419 9.68 *** BC 4 .03043 8.62 *** ABC 16 .01552 4.40 *** Error 135 .00353 Clipping weight 5-18 A 4 .99633 567.42 *** B 2 .12623 71.89 *** AB 8 .02895 16.49 *** C 2 .66787 380.36 *** AC 8 .03751 21.36 *** BC 4 .00889 5.06 *** ABC 16 .00420 2.39 ** Error 135 .00176 199 Table 34.--Cont. Source d.f. Mean Square F Approximate Significance Clipping weight 5—25 A 4 .33803 303.91 *** B 2 .09067 81.52 *** AB 8 .01904 17.12 *** C 2 .52265 469.90 *** AC 8 .05186 46.63 *** BC 4 .01702 15.31 *** ABC 16 .00900 8.10 *** Error 135 .00111 Bray Pl-Arsenic Soil Test A 4 11166.02 189.95 *** B 2 19046.84 324.02 *** AB 8 2929.21 49.83 *** C 2 177040.23 3011.77 *** AC 8 3317.66 56.44 *** BC 4 7184.02 122.21 *** ABC 16 848.48 14.43 *** Error 135 58.78 NaHCO3-Arsenic Soil Test A 4 15.030 389.65 *** B 2 2.001 51.87 *** .AB 8 .590 15.29 *** C 2 11.237 291.31 *** {AC 8 2.083 54.00 *** BC 4 .958 24.84 *** .ABC 16 .220 5.71 *** Error 135 .039 200 50.: 00.: 00.: av.: mv.: 00.: 00.: 00.: HN.: 00. m 5000 0 .A00 .0 .Am .00 mawow Add 00 000H0>¢ 00.: 00.: 00.: mm.: mm.: 00.: 0m.: 0N.: 00.: H0. H0 >0um 000 .4 .Am .00 maflOm H0chflz mo 000u0>0 m 00. m5.: 00.: 00.: 00.: 50.: «0.: 00.: 00.: 0m.: 00.: H 00002 q .Am .04 mo 000um>< 50.: 00.: H0.: 50.: 00.: 00.: 00.: 5N.: NH.: H0.: 0 50um 0 .Am .00 mo 000H0>< v5.: 55.: 00.: 05.: 00.: 00.: 00.: 00.: 00.: 0a. Hm >0um Amv u0mm :: 05.: 00.: 00.: 00.: «0.: 5a.: Hm.: 00.: 5a.: 5a. am 5000 A0000 E000 >0H0 huafim 0m. 00.: 00.: 05.: m0.: m5.: m0.: 00.: 00.: H0.: 0H. Hmoum0z 5000 00.: 00.: 05.: H0.: «5.: 00.: 00.: 00.: 00.: HH. 0 m0um Adv E000 00. Hm.: H0.: 05.: 05.: 05.: 05.: 00.: 50.: 00.: 00.: 000002 8004 50:00 00.: 00.: m0.: m0.: 00.: 50.: 0~.: 30.: mm.: 50.: am swam iqmv smog 50cmm mm. 05.: 05.: 05.: 55.: v0.: 00.: 00.: m0.: HH.: 00.: 000002 0000 >E0oq 00.: 001: 00.: N0.: 05.: 05.: m0.: 0a.: mm.: m0.: Hm >0um Away 0:00 55000 0u00 an A00 musmfloflmmmoo soau0amuuoo 0N:0 0H:0 HH:0 0:0 5m:v om:v HH:¢ «:0 0N:m 5H:m muonuwe ma .po0uuxm cmm3umn H mmu0o mcflmmwau ma mafiom .Awu0m oacmmud:mm:mnsuxmev 0 pawsflummxm mmdoscwmuw .0000 up m0ama> 0cflmmwao 000 .mHHOm .owcmmu0 man0uo0nuxm 000002 .OHcmmH0 man0uo0uuxw Hm >0um Cmmzumn Any mucwHOHmmmoo :oHu0Hmuu00::.0m mqmde 201 TABLE 36.—-Analyses of variance for data in Table 26. Greenhouse Experiment 7. Source d.f. 'Mean Square F Approximate Significance Clipping weight 3-26 A(As Form) 1 .000002 .0141 n.s. B(As Rate) 2 .000597 4.7345 * AB 2 .000017 .1329 n.s. C(Time) 1 .002304 18.2616 *** AC 1 .000002 .0141 n.s. BC 2 .000597 4.7345 * ABC 2 .000017 .1330 n.s. Error 24 .000126 Clipping weight 4-1 A 1 .000005 .1089 n.s. B 2 .000006 .1476 n.s. AB 2 .000010 .2403 n.s. C 1 .000023 .5419 n.s. AC 1 .000005 .1089 n.s. BC 2 .000006 .1476 n.s. ABC 2 .000010 .2403 n.s. Error 24 .000043 Clipping weight 4-8 A 1 .000001 .0073 n.s. B 2_ .001060 7.7166 ** AB 2 .000048 .3514 n.s. C 1 .004182 30.4560 *** AC 1 .000001 .0073 n.s. BC 2 .001060 7.7166 ** ABC 2 .000048 .3514 n.s. Error 24 .000137 Table 36.-—Cont. 202 Source d.f. Mean Square F Approximate Significance Clipping weight 4-17 A 1 .000034 .0882 n.s. B 2 .008304 21.4199 *** AB 2 .000484 1.2531 n.s. C 1 .033185 86.0017 *** AC 1 .000034 .0882 n.s. BC 2 .008304 21.5199 *** ABC 2 .000484 1.2531 n.s. Error 24 .000386 Clipping weight 4-26 A 1 .000100 .0352 n.s. B 2 .052489 18.4770 *** AB 2 .000046 .0163 n.s. C 1 .209764 73.8403 *** AC 1 .000100 .0352 n.s. BC 2 .052489 18.4770 *** ABC 2 .000046 .01631 n.s. Clipping weight 5-7 A 1 .000069 .0607 n.s. B 2 .012740 11.1418 *** AB 2 .000692 .6055 n.s. C 1 .050925 44.5380 *** AC 1 .000069 .0607 n.s. BC 2 .012740 11.1418 *** ABC 2 .000692 .6055 n.s. Error 24 .0011434 203 Table 36.--Cont. Source d.f. Mean Square F Approximate Significance Clipping weight 5-21 A 1 .013417 3.6460 n.s. B 2 .027893 7.5795 * AB 2 .006704 1.8218 n.s. C 1 .038090 10.3505 * AC 1 .000047 .0127 n 3. BC 2 .009761 2.6524 n.s. ABC 2 .000216 .0587 n.s. Error 24 .003680 Clipping weight 6-4 A 1 .002368 1.1153 n.s. B 2 .138532 65.2359 *** AB 2 .002888 1.3602 n.s. C 1 .032761 15.4774 *** AC 1 .001296 .6103 n.s. BC 2 .008374 3.9435 * ABC 2 .001225 .5770 n.s. Error 24 .002124 Clipping weight 6-19 A 1 .000765 .1847 n.s. B 2 .007603 1.8346 n.s. AB 2 .000822 .1984 n.s. C 1 .022300 5.3812 * AC 1 .000427 .1031 n.s. BC 2 .005596 1.3505 n.s. ABC 2 .000192 .0464 n 5. Error 24 .004144 Table 36.--Cont. 204 Source d.f. Mean Square F Approximate Significance Clipping weight 7-13 A 1 .007891 1.6674 n.s. B 2 .002628 .5553 n.s. AB 2 .016498 3.4861 * C 1 .000030 .0064 n.s. AC 1 .000010 .0021 n.s. BC 2 .002345 .4956 n.s. ABC 2 .000075 .0159 n.s. Error 24 .004732 Clipping weight 7-25 A 1 .000220 .0350 n.s. B 2 .080316 12.7597 *** AB 2 .008776 1.3942 n.s. C 1 .019460 3.0916 n.s. AC 1 .002256 .3585 n.s. BC 2 .004871 .7739 n.s. ABC 2 .001877 .2981 n.s. Error 24 .006295 Arsenic Soil Test A 1 5675.1 95.83 *** B 2 37307.9 630.01 *** AB 2 2285.8 38.60 *** C 1 4285.9 72.37 *** AC 1 4937.4 83.38 *** BC 2 1434.5 24.22 *** ABC 2 1579.9 26.68 *** Error 24 59.2 205 TABLE 37.--Ana1yses of variance for data in Growth Chamber Experiment 1. Source d.f. Mean Square F Approximate Significance Clipping weight 5-31 A(As) 2 .000207 .5819 n.s. B(HZO) 2 .010665 30.0153 *** AB 4 .000063 .1761 n.s. Error 18 .000355 Clipping weight 6-11 A 2 .000161 .7582 n.s. B 2 .015181 71.4718 *** AB 4 .000381 1.7929 n.s. Error 18 .000212 Clipping weight 6-18 A 2 .000072 .3403 n.s. B 2 .011186 53.1653 *** AB 4 .000059 .2798 n.s. Error 18 .000210 Clipping weight 6—26 A 2 .000221 1.8329 n.s. B 2 .008669 71.8907 *** AB 4 .000162 1.3437 n.s. Error 18 .000121 Clipping Weight 7-8 A 2 .000876 2.8618 n.s. B 2 .009181 29.9914 *** AB 4 .000227 .7431 n.s. Error 18 .000306 Table 37.--Cont. 206 Source .f. Mean Square F Approximate Significance Clipping weight 7-18 A 2 .001321 4.6803 * B 2 .004476 15.8605 *** AB 4 .000151 0.5360 n.s. Error 18 .000282 Clipping weight 7-27 A 2 .001700 4.8782 * B 2 .002827 8.1151 ** AB 4 .000357 1.0245 n.s. Error 18 .000348 Clipping weight 8-12 A 2 .008625 10.7634 *** B 2 .004859 6.0636 ** AB 4 .002444 3.0500 * Error 18 .000801 H20 Consumption 6-1 A 2 2.785 1.018 n.s. B 2 270.055 98.747 *** AB 4 5.217 1.908 n.s. Error 18 2.735 H20 Consumption 6-2 A 2 8.508 2.703 n.s. B 2 210.704 66.937 *** AB 4 1.943 0.6171 n.s. Error 18 3.148 207 Table 37.--Cont. Source d.f. Mean Square F Approximate Significance H20 Consumption 6-5 A 2 2.033 .753 n.s. B 2 233.806 86.583 *** AB 4 .624 .231 n.s. Error 18 2,700 H20 Consumption 6-19 A 2 2.405 1.165 n.s B 2 88.378 42.802 *** AB 4 2.238 1.084 n.s Error 18 2.065 H20 Consumption 6-20 A 2 1.900 .931 n.s. B 2 45.674 22.373 *** AB 4 .881 .432 n.s. Error 18 2.041 H20 Consumption 6-21 A 2 1.200 .569 n.s B 2 114.651 54.309 *** AB 4 1.450 .687 n. Error 18 2,111 Arsenic Soil Test A 2 10276.79 455.41 *** B 2 228.79 10.14 *** AB 4 59.26 2.63 n.s. Error 18 22.57 90a 804 70‘ 60‘ 50~ 40‘ 30' T% 20- 10 4 L I 1 0 5 10 15 20 ppm—As Fig. 1.--Arsenic standard curve for Small and McCants (97) procedure. 25 209 1.5“ Poa annua L. Total Clipping Weight (g/POt) Kg TCA Applied/100 m Fig.2.--Tota1 clipping weights versus applied TCA on Poa annua L. (Greenhouse Experiment 2). 210 1.5—1 Penncross bentgrass Total Clipping Weight (g/pot) 0 L A A A l A 4 l l 1 L l 1 J1 J A l 1 J 0 5 10 15 20 Kg TCA Applied/100 m2 Fig. 3.--Tota1 clipping weights versus applied TCA on Penncross bentgrass (Greenhouse Experiment 2). Total Clipping Weight (g/pot) 211 Cohansey bentgrass Kg TCA Applied/100 m2 Fig. 4.--Tota1 clipping weights versus applied TCA on Cohansey bentgrass (Greenhouse Experiment 2). 212 1-5‘ Merion Kentucky bluegrass I? O .. Cl: \ _ 3‘ fa, '1 o. 1.0— '8 3 1 U3 - .5 Q: ‘1 Q: .,..{ " r-1 0 o 5- I-P 4—P E + Z-P a d \\ “\043 0 11 . 1 i Mi 1 .fi: - I 0 5 10 15 20 2 Kg TCA Applied/100 m Fig. 5.--Tota1 clipping weights versus applied TCA on Merion Kentucky bluegrass (Greenhouse Experiment 2). BIBLIOGRAPHY 10. BIBLIOGRAPHY Albert, w. B. 1932. Arsenic toxicity in soils. S. Car. Expt. Sta. 45:44-46. . 1933. Arsenic toxicity in soils. S. Car. Expt. Sta. 46:44-45. . 1934. Arsenic solubility in soils. S. Car. Expt. Sta. 47:45-46. , C. H. Arndt. 1931. The concentration of soluble arsenic as an index of arsenic toxicity to plants. S. Car. Expt. Sta. 44:47—48. Alexander, M. 1967. Introduction to Soil Micro- biology. John Wiley & Sons, Inc., New York. Pp. 412-420. Arnott, J. T., and A. L. Leaf. 1967. The determina- tion and distribution of toxic levels of arsenic in a silt loam soil. Weeds 15:121-124. Bartlett, N. F., and J. Troll. 1967. Can you work with Poa annua. University of Massachusetts Turf Bulletin 4:5-9. Bastisse, E. M. 1967. Contribution a l'etude des equilibres entre les sols et les solutions de certains composes de l'arsenic et du phosphore. Sci. Sol. No. 1:5—17. Beard, J. B. 1964. Effects of ice, snow and water covers on Kentucky bluegrass, annual bluegrass, and creeping bentgrass. Crop Sci. 4:638-640. . 1966. Direct low temperature injury of nineteen turfgrasses. Mich. Quart. Bull. Vol. 48, No. 3:377-383. 213 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 214 . 1968. Effect of temperature stress on Poa annua. U.S.G.A. Green Section Record. July:lO-12. 1968, Low temperatures and Poa annua. U.S.G.A. Green Section Record. November:10-11. . 1970. An ecological study of annual blue- grass. U.S.G.A. Green Section Record. March: 13-18. . Turfgrass: Science and Culture. Prentice- Hall, Inc., Englewood Cliffs, New Jersey. To be published in 1973. Wmmruflr , and C. R. Olien. 1963. Low temperature injury in the lower portion of Poa annua L. crowns. Crop Sci. 3:362-363. Benson, N. R. 1953. Effect of season, phosphate, and acidity on plant growth in arsenic-toxic soils. Soil Sci. 76:215-224. Bingham, W. S. 1972. The nature of biochemical mechanism of herbicide selectivity. Proc. of Scotts Turfgrass Research Conference. Vol. 3: 27-47. Bishop, R. F., and D. Chisholm. 1962. Arsenic accumu- lation in Annapolis Valley Orchard soils. Can. J. Soil Sci. 42:77-80. Boischot, P., and J. Hébert. 1948. Fixation des arséniates par 1e sol. Ann. Agron. 18:426-448. Chapman, H. D., and P. F. Pratt. 1961. Methods of Analysis for Soils, Plants, and Waters. Univer- sity of California, Division of Agriculture Sciences. Pp. 73-77. Clements, H. F., and J. Munson. 1947. Arsenic tox- icity studies in soil and in culture solution. Pacific Sci. Vol. 1:115-171. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 215 Cooper, H. P., W. R. Paden, E. E. Hall, W. B. Albert, W. B. Rogers, and J. A. Riley. 1931. Effect of calcium arsenate on the productivity of certain soil types. S. Car. Expt. Sta. 44:28-36. I I I I and . 1932. Soils differ markedly in their reSponse to additions of calcium arsenate. 8. Car. Expt. Sta. 45:23-27. I WM" Cockerham, S. T., and J. W. Whitworth. 1967. Germi— nation and control of annual bluegrass. The Golf Superintendent 35(5):10-17. Crafts, A. S. 1935. The toxicity of sodium arsenite and sodium chlorate in four California soils. Hilgardia 9:461-498. Daniel, W. H. 1955. Poa annua control with arsenic materials. The Golf Course Reporter 23(1):5—8. . 1955. Arsenic control of Poa annua points to fertilizing study. Golfdom. Apri1:70-76. . 1972. Midwest current trends in turfgrass weed control. Proc. of Scotts Turfgrass Research Conference. Vol. 3:155-160. Day, P. R. 1969. Particle fractionation and particle- size analysis. Methods of Soil Analysis. Agron- omy Monograph No. 9, Part 1: pp. 545-567. Dean, L. A., and E. J. Rubins. 1947. Anion exchange in soils: I. Exchangeable phosphorus and the anion-exchange capacity. Soil Sci. 63:377-387. Deb, D. L., and N. P. Datta. 1967. Effect of asso- ciating anions on phosphorus retention in soil: I. Under variable phosphorus concentration. Plant and Soil. 26:303-316. Dickens, R., and A. E. Hiltbold. 1967. Movement and persistence of methanearsonates in soil. Weeds 15:299-304. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 216 Dorman, C., F. H. Tucker, and R. Coleman. 1939. The effect of calcium arsenate upon the productivity of several important soils of the cotton belt. J. Amer. Soc. Agron. 31:1020-1028. Duble, R. L., E. C. Holt, and G. G. McBee. 1969. Translocation and breakdown of disodium metha- nearsonate (DSMA) in coastal bermudagrass. J. Agr. Food Chem. 17:1247-1250. Ellis, B. G. CrOp and Soil Science Dept., Michigan State University, Personal communication. 1971. Engel, R. E., A. Morrison, and R. D. Ilnicki. 1968. Preemergence chemical effects on annual bluegrass. The Golf Superintendent 3(2):20-21, 39. Epps, E. A., and M. B. Sturgis. 1939. Arsenic com- pounds toxic to rice. Soil Sci. Soc. Amer. Proc. 4:215-218. Everett, C. F. 1962. Effect of phosphorus on the phytotoxicity of tricalcium arsenate as mani- fested by bluegrass and crabgrass. Diss. Abstracts 23:1851. Freeborg, R. P. 1971. Arsenic concentrations required for Poa annua control as determined by soil tests. Ph.D. Thesis, Graduate School, Purdue University, Lafayette, Ind. Gibeault, V. A. 1971. Perenniality in Poa annua L. Ph.D. Thesis, Graduate School, Oregon State Uni- versity, Corvallis, Oregon. Greaves, J. E. 1934. The arsenic content of soils. Soil Sci. 38:355-362. Hochster, R. M., and J. Quastel, eds. 1963. Meta- bolic Inhibitors. Volume II. Academic Press, New York. Hurd—Karrer, A. M. 1939. Antagonism of certain ele- ments essential to plants toward chemically re- lated toxic elements. Plant Phys. 14:9—29. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 217 Jackson, M. L. 1958. Soil Chemical Analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. Jacobs, L. W., J. K. Syers, and D. R. Keeney. 1970. Arsenic sorption by soils. Soil Sci. Soc. Amer. Proc. 34:750-754. , D. R. Keeney, and L. M. Walsh. 1970. Arsenic residue toxicity to vegetable crops grown on Plainfield sand. Agron. J. 62:588-591. Johnson, L. R., and A. E. Hiltbold. 1969. Arsenic content of soil and crops following use of methane- arsonate herbicides. 'Soil Sci. Soc. Amer. Proc. 33:279-282. Jones, J. S., and M. B. Hatch. 1937. The signifi- cance of inorganic Spray residue accumulations in orchard soils. Soil Sci. 44:37-61. Juo, A. S. R., and B. G. Ellis. 1968. Chemical and physical prOperties of iron and aluminum phos- phates and their relation to phosphorus availa- bility. Soil Sci. Soc. Amer. Proc. 32:216-221. Juska, F. V. 1960. Pre-emergence herbicides for crabgrass control and their effects on germina- tion of turfgrass species. Weeds 9:137-144. , and A. A. Hanson. 1967. Phosphorus and its relationship to turfgrass and Poa annua L. California Turfgrass Culture l7(4):27-29. , and . 1967. Factors affecting Poa annua L. control. The Golf Superintendent. , and . 1970. Nutritional require- ments of Poa annpa. The Golf Superintendent. March:42-45. , A. A. Hanson, and A. W. Hovin. 1970. Phytotoxicity of preemergence herbicides. U.S.G.A. Green Section Record. Sept.:2-5. 218 55. Kardos, L. T., S. C. Vandecaveye, and N. Benson. 1941. Investigation of the causes and remedies of the unproductiveness of certain soils following the removal of mature trees. Wash. Agr. Exp. Sta. Bull. 410:25. 56. Keaton, C. M., and L. T. Kardos. 1940. Oxidation- reduction potentials of arsenate-arsenite systems in sand and soil mediums. Soil Sci. 50:189-207. 57. Kelly, J. B., and A. R. Midgley. 1943. Phosphate fixation--an exchange of phosphate and hydroxyl ions. Soil Sci. 55:167-176. 58. Kerr, C. F. 1969. Program for gradual removal of Poa annua. The Golf Superintendent. April:28-29. 58a. Kinra, K. L., H. D. Foth, and J. F. Davis. 1962. Effect of ordinary superphosphate on emergence of oats and wheat. Agron. J. 54:180-181. 59. Leaf, A. L., and R. E. Smith, Jr. 1960. Herbical value of arsenic trioxide in Eastern United States. Weeds 8:374-378. 60. Lehninger, A. L. 1970. Biochemistry. Worth Pub- lishers, Inc., New York. 61. Liebig, G. F. 1966. Arsenic. Diagnostic Criteria for Plants and Soils, H. D. Chapman, ed. Univer- sity of California, Division of Agricultural Sciences. 62. Lunt, O. R., R. L. Branson, and S. B. Clark. 1967. Response of five grass species to phosphorus on six soils. California Turfgrass Culture. l7(4):25-26. 63. Machlis, L. 1941. Accumulation of arsenic in the shoots of sudan grass and bush bean. Plant Phys. 16:521-544. 64. MacPhee, A. W., D. Chisholm, and C. R. MacEachern. 1960. The persistence of certain pesticides in the soil and their effect on crop yields. Can. J. Soil Sci. 40:59-62. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 219 Margulis, H., and R. Bourniquel. 1950. Action de 1'acide arsénieux sur 1e sol. Ann. Agron. 20:550-566. , and J. Gane. 1948. Action de la chaux sur 1'acide arsénieux. Ann. Agron. 18:28-32. , and . 1948. Action du carbonate de chaux sur 1'acide arsénieux. Ann Agron. 18: 175-178. Martin, R., M. J. Masson, C. Duc-Maugé, and H. Guérin. 1959. No. 72--Sur les arséniates d'aluminium. Bull. Soc. Chim. France. 412-418. McBee, G. C., P. R. Johnson, and E. C. Holt. 1967. Arsenic residue studies on coastal bermudagrass. Weeds 15:77-79. Miles, J. R. 1968. Arsenic residues in agricultural soils of Southwestern Ontario. J. Agr. Food Chem. 16:620-622. Misra, S. G., and R. C. Tiwari. 1963. Studies on arsenite-arsenate system. Adsorption of arsenate. Soil Sci. and Plt. Nutr. 9(6):lO-13. Mogendorff, N. 1925. Some chemical factors involved in arsenical injury to fruit trees. N.J. Agr. Expt. Sta. Bull. 419:3-47. Monteith, Jr., J., and J. W. Bengtson. 1939. Arsen- ican compounds for the control of turf weeds. Turf Culture. January:10-43. Morrison, J. L. 1969. Distribution of arsenic from poultry litter in broiler chickens, soil, and crops. J. Agr. Food Chem. 17:1288-1290. Mortland,'M. M. Crop and Soil Science Dept., Michigan State University. Clay Mineralogy 825 notes. 1971. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 220 Murrmann, R. P., and M. Peech. 1968. Reaction prod- ucts of applied phosphate in limed soils. Soil Sci. Soc. Amer. Proc. 32:493-496. Naylor, A. W. 1940. The influence of temperature, calcium and arsenious acid on seedlings of Kentucky bluegrass. Turf Culture. February:28-45. Olson, O. E., L. L. Sesson, and A. L. Moxon. 1940. Absorption of selenium and arsenic by plants from soils under natural conditions. Soil Sci. 50:115-118. Pakalus, P. 1968. Spectrophotometric determination of traces of phosphorus by an extraction method. Anal. Chim. Acta. 40:1-12. Paul, J. 1965. Simultaneous determination of arsenic and phosphorus. Mikrochim. Acta.: 830—835. Pearce, G. W., and A. W. Avens. 1937. Further phase rule studies of the calcium arsenates. J. Amer. Chem. Soc. 59:1258-1261. , and L. B. Norton. 1936. A phase rule study of the calcium arsenates. J. Amer. Chem. Soc. 58:1104-1108. Ramulu, U. S. S., P. F. Pratt, and A. L. Page. 1967. Phosphorus fixation by soils in relation to ex- tractable iron oxides and mineralogical composi- tion. Soil Sci. Soc. Amer. Proc. 31:193-196. Reed, J. F., and M. B. Sturgis. 1936. Toxicity from arsenic compounds to rice on flooded soils. J. Amer. Soc. Agron. 28:432-436. Reymont, T. M., and R. J. Dubois. 1971. Determina- tion of traces of arsenic by c0precipitation and X-ray fluorescence. Anal. Chim. Acta. 56:1-6. Rieke, P. E. 1963. Relationships between aluminum and soil acidity in several Michigan soils and related mineralogy studies. Ph.D. Thesis, Michigan State University, East Lansing, Michigan. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 221 Robert, E. C., and F. E. Markland. 1967. Effect of phosphorus levels and two growth regulators on arsenic toxicity in Poa annua L. Agron. Abst:57. , and J. J. Ptacek. 1962. Pre-emergence crabgrass control. The Golf Course Reporter 30(3):18-26. Rosenfels, R. S., and A. S. Crafts. 1939. Arsenic fixation in relation to the sterilization of soils with sodium arsenite. Hilgardia 12:203-229. Rumburg, C. B., R. E. Engel, and W. F. Meggitt. 1960. Effect of temperature on the herbicidal activity and translocation of arsenicals. Weeds 8:582-588. , , and . 1960. Effect of phosphorus concentration on the absorption of arsenate by oats from nutrient solution. Agron. J. 52:452-453. Salisbury, F. B., and C. Ross. 1969. Plant Physi- ology. Wadsworth Publishing Company, Inc., Bel- mont, California. Schery, R. W. 1968. Bluegrass/bentgrass checks Poa annua. The Golf Superintendent 36(10):32-34. Schweizer, E. E. 1967. Toxicity of DSMA soil resi- dues to cotton and rotational crops. Weeds 15:72-76. Sieling, D. H. 1946. Role of kaolin in anion sorp- tion and exchange. Soil Sci. Soc. Amer. Proc. 11:161-170. Singh, R. K. N., and R. W. Campbell. 1965. Herbi- cides on bluegrass. Weeds 13:170-171. Small, Jr., H. G., and C. B. McCants. 1961. Determi- nation of arsenic in flue-cured tobacco and in soils. Soil Sci. Soc. Amer. Proc. 25:346-348. , and . 1962. Influence of arsenic applied to the growth media on the arsenic content of flue-cured tobacco. Agron. J. 54:129-133. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 222 Smith, C. M. 1920. The arsenates of calcium. 1. Equilibrium in the system arsenic pentoxide, calcium oxide, water at 35 C (acid section). J. Amer. Chem. Soc. 42:259-265. Sprague, H. B., and G. W. Burton. 1937. Annual blue- grass (Poa annua L.) and its requirements for growth. N. J. Agr. Expt. Bull. 630. Stadtherr, R. J. 1963. Studies on the use of arsen- icals for crabgrass control in turf. Ph.D. Thesis, University of Minnesota, St. Paul, Minnesota. Stapledon, P. 1939. What others write on turf. Turf Culture. January:80-87. Steel, R. G. D., and J. H. Torrie. 1960. Principals and Procedures of Statistics. McGraw-Hill Book Company, Inc., New York. Steward, J., and E. S. Smith. 1922. Some relations of arsenic to plant growth: Part 2. Soil Sci. 14:119-126. Swenson, R. M., C. V. Cole, and D. H. Sieling. 1949. Fixation of phosphate by iron and aluminum and replacement by organic and inorganic ions. Soil Sci. 67:3-22. Tandon, H. L. S., and L. T. Kurtz. 1968. Isotopic exchange characteristics of aluminum- and iron- bound fractions of soil phosphorus. Soil Sci. Soc. Amer. Proc. 32:799-802. Thompson, A. H., and L. P. Batjer. 1950. Effect of various soil treatments for correcting arsenic injury of peach trees. Soil Sci. 69:281-290. Vandecaveye, S. C., G. M. Horner, and C. M. Keaton. 1936. Unproductiveness of certain orchard soils related to lead arsenate spray accumulations. Soil Sci. 42:203-215. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 223 , C. M. Keaton, and L. T. Kardos. 1938. Some factors affecting the toxicity of arsenical spray accumulations in the soil. Proc. 34th Ann. Meet- ing Wash. State Hort. Ass. Pp. 150-159. Vincent, C. L. 1944. Vegetable and small fruit grow- ing in toxic ex-orchard soils of central Washington. Wash. Agr. Expt. Sta. Bull. 437:3-31. Vinogradov, A. P. 1948. Arsenic in soils of the U.S.S.R. Soils and Fert. 11:1416. Weton, F. A., and J. C. Carroll. 1938. Crabgrass in relation to arsenicals. J. Amer. Soc. Agron. 30:816-826. Whitnack, D. C., and R. G. Brophy. 1969. A rapid and highly sensitive single-sweep polarographic method of analysis for arsenic (III) in drinking water. Anal. Chim. Acta 48:123-127. Williams, K. T., and R. R. Whetstone. 1940. Arsenic distribution in soils and its presence in certain plants. U.S.D.A. Technical Bull. No. 732:1-20. Woolson, E. A. 1969. The chemistry and toxicity of arsenic in soil. Ph.D. Thesis. University of Maryland, College Park, Maryland. , J. H. Axley, and P. C. Kearney. 1971. Correlation between available soil arsenic, esti- mated by six methods, and response of corn (Egg mays L.). Soil Sci. Soc. Amer. Proc. 35: 101-105. , , and . 1971. The chemistry and phytotoxicity of arsenic in soils: I. Con- taminated field soils. Soil Sci. Soc. Amer. Proc. 35:938-943. ' Youngner, V. B. 1972. Ecological forces affecting weeds and their control. Proc. of Scotts Turf- grass Research Conference. Vol. 3:1-26. L|LL|LLLL|LLL LIL LLLI LL LLLI LL LLL LLLLLLLLLLLLLLLILLLLLLLL