DYNAMICS OF ARSENIC N THE AQUATIC ENVIRONMENT Thesis for the Degree of Ph. D. MlCHiGAN STATE UNIVERSlTY LEONARD P. SOHACKI 1968 This is to certify that the thesis entitled- Dynamiu 06 Amen/La in the Aquatic Enui/wnmen/t presented by Leona/Ld P. Sohacfz/é has been accepted towards fulfillment of the requirements for Ph.D. Lmnozogy degree in Major professor Date ALLQLUSI 6, 1968 0-169 ‘ n V r; amoma 3y 3'— ‘ Hm & SON ' E I mgjltlllf'fil' mu. .‘ ABSTRACT DYNAMICS OF ARSENIC IN THE AQUATIC ENVIRONMENT by Leonard P. Sohacki The distribution of arsenic, added as the herbicide sodium arsenite, was traced in a small farm pond for four years via radioarsenic and stable arsenic analyses. Shortly after introduction to the pond a major portion of the intro- duced arsenic passed from the water phase into the solids phase (sediments, flora, fauna) of the environment. Es— tablishment of an exchange between these two phases main- tained higher than natural arsenic levels in the water for at least two years after the last sodium arsenite treatment. Most of the arsenic which disappeared from the water accumulated in the sediments within two months after treat- ment. Silt samples consistently exhibited greater arsenic concentrations than sand, and both sediment types showed fluctuations in arsenic content which appeared to originate from biological rather than physical or chemical sources. Subsequent analysis suggested that the arsenic was pene- trating into the deeper sediment layers and thereby losing its active role in the environment. Leonard P. Sohacki All aquatic plants rapidly accumulated arsenic with- in two hours after the introduction of sodium arsenite, al- though the tissue concentrations differed markedly between species. All plants except Chara globularis were killed or suffered temporary setbacks because of the herbicide treat— ment. Chara globularis appeared to be unaffected by the arsenic toxicity in spite of the high arsenic content in its tissues. Accumulation by the aquatic plants persisted throughout the study even though water-arsenic diminished to extremely low levels. Arsenic uptake by the vascular macro— phytes appeared to be associated with the stems and leaves rather than roots. The seston and aufwuchs accumulated higher quantities of arsenic than any other environmental component. As the study progressed the accumulations decreased. At the last sampling, the seston and aufwuchs arsenic content approached natural levels. The highest arsenic concentrations in the pond fauna were detected within the first week after herbicide appli- cation. Subsequently, low arsenic levels prevailed. The arsenic content of edible portions of fishes were judged too low to present a public health hazard. DYNAMICS OF ARSENIC IN THE AQUATIC ENVIRONMENT BY \\ \_. <\ Leonard P. Sohacki A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 1968 \x) ACKNOWLEDGMENT S A study such as this would be impossible to conduct without academic, moral, and financial support from others. I wish to express my appreciation to all who assisted me in any way. To Dr. Robert C. Ball, my committee chairman, I offer special thanks for the academic guidance and financial as— sistance he provided during my graduate career. I am grateful to Drs. Peter I. Tack, Gordon E. Guyer, and Gerald Prescott for serving as members of my guidance committee. I wish to thank Drs. Eugene W. Roelofs and Wilbert E. Wade who provided assistance in time of need. Finally, I wish to thank my wife, Betty, who assisted me by providing encouragement. This study was financed by (1) Atomic Energy Com- mission Grant AT (11-1) 655 administered by Drs. R. C. Ball and F. F. Hooper, and (2), water Pollution Control Administra- tion Fellowship l-Fl-WPb25,7ll-Ol. ii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1 THE STUDY AREA . . . . . . . . . . . . . . . . . . . . 7 METHODOLOGY . . . . . . . . . . . . . . . . . . . . . 13 Preparation of Arsenic-74 . . . . . . . . . . . . l3 Counting Equipment and Procedures . . . . . . . . 13 Decay Rates, Counter Efficiencies, and Activity Units . . . . . . . . . . 14 ,Application of the Tagged Herbicide . . . . . . . 16 Water and Seston . . . . . . . . . . . . . 16 Bottom Sediments . . . . . . . . . . . . . . . . . l7 Aufwuchs . . . . . . . . . . . . . . . . . . . . . l7 Macrophytes . . . . . . . . . . . . . . . . . . . 18 Fishes . . . . . . . . . . . . . . . . . . . . . . 19 Invertebrates . . . . . . . . . . . . . . . 20 Phytoplankton Productivity . . . . . . . . . . . . 20 Stable Arsenic . . . . . . . . . . . . . . . . . 21 RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . 26 Water . . . . . . . . . . . . . . . . . . . . . . 26 Radioarsenic Study . . . . . . . . . . . . . 26 Stable Arsenic Study . . . . . . . . . . . . 31 Sediments . . . . . . . . . . . . . . . 34 Radioarsenic Study . . . . . . . . . . . . . 34 Stable Arsenic Study . . . . . . . . . . . . 39 Macrophytes . . . . . . . . . . . . . . . . 48 Radioarsenic Study . . . . . . . . . . . . . 48 Stable Arsenic Study . . . . . . . . . . . . 61 Seston . . . . . . . . . . . . . . . . 67 Radioarsenic Study . . . . . . . . . . . . . 67 Stable Arsenic Study . . . . . . . . . . . . 74 Aufwuchs . . . . . . . . . . . . . . . . . . . 76 Radioarsenic Study . . . . . . . . . . . . . 76 Stable Arsenic Study . . . . . . . . . . . . 80 Macroinvertebrates and Fishes . . . . . . . . . . 81 Radioarsenic Study . . . . . . . . . . . . . 81 Stable Arsenic Study . . . . . . . . . . . . 84 iii Page SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 88 LITERATURE CITED . . . . . . . . . . . . . . . . . . . 90 APPENDIX . . . . . . . . . . . . . . . . . . . . . . . 95 iv Table LIST OF TABLES Two way analysis of variance testing the effects of dates and depth on the arsenic content of water Stable arsenic concentrations of pond water Summary of a two way analysis of variance testing the effect of time and sediment composition upon activity density in the 1 cm sediment layer . . . . . . . . . Accumulation of A374 by aquatic plants within two hours after treatment . . Accumulation of stable arsenic by submerged plants during the years 1964, 1965, 1966, and 1967 . . . . . . . . . . . . . . . . Summary of the one way analysis of variance testing the stable arsenic concentrations in leaves, stems and seeds of Potamogeton praelongus . . . . . . . . . . . . . . . Means, standard errors and concentration factors of the stable arsenic concen— trations in the seston and aufwuchs sampled from 1963 to 1968 . . . . Raw water activity and the conversions re- quired for estimating the turnover times in water and solids . . . . . . . Page 32 33 35 50 65 67 75 101 Figure 1. 10. ll. 12. 13. LIST OF FIGURES Map of the Lake City experimental ponds and laboratory facilities . . . . . . Distribution of silt and sand in pond C Calibration curve for arsenic showing the 93% confidence intervals for individual estimates Decrease in radioarsenic from the water of the test pond . . . . . . . . . Activity densities of pond sediments showing fluctuations in arsenic content Stable arsenic concentrations in the core samples of sand from the test pond Stable arsenic concentrations in core samples of silt from the test pond Arsenic profiles of a silt and a sand core sample showing the high concentrations at the 3-4 cm layer ~Activity density of_§pirogyra sp showing the fluctuations of arsenic content Activity density of Chara_globularis showing the fluctuations in arsenic content . Activity densities of submerged aquatic plants . . . . . . . . . . Stable arsenic concentrations of submerged macrophytes from the test pond Solar energy and phytoplankton productivity values before and after sodium arsenite introduction 0 C O O O O O O O I 0 vi Page 11 25 28 37 42 44 47 53 56 59 63 71 Figure 74 14. As activity in the seston of the water samples 15. ,Standing crop and activity density of the aufwuchs from the test pond . . l6. Plot of N—B used in determining turnover times of arsenic in the water and solids phases of the experimental pond vii Page 73 79 100 INTRODUCTION .Arsenic is a ubiquitous element whose presence may be demonstrated in practically all living and nonliving matter. Natural occurring arsenic in the free state is scant; it normally exists in combination with sulfur as arsenopyrite (FeAsS), orpiment (As283) or realgar (A3252) (King and Caldwell, 1959). Natural concentrations of arsenic in soils may vary from a fraction of a part per million (ppm) to 40 ppm (Vallee, Ulmer, and wacker, 1960). Studies on the terrestrial environment showed local arsenic levels to in- crease as a result of human influence. Mining and industrial wastes plus widespread use of arsenical pesticides have in- creased arsenic concentrations of soils to many times their natural levels (Vallee, gg_g1., 1960; Bishop and Chisholm, 1962). The study of arsenic in fresh waters has been neglected (Hutchinson, 1956). From the meager data that do exist, we know that natural occurring arsenic in fresh waters is usually low. Concentrations of 0.002 to 0.003 ppm were reported for waters in Germany (Hutchinson, 1956). Fish (1963) measured 0.006 ppm in a New Zealand lake. Doepke (1963) and Leuschow (1964) determined the natural arsenic concentrations of water in twelve Wisconsin lakes; a range of 0.002 to 0.011 was detected. Mineral springs may con- tain sufficient quantities of arsenic to cause cattle poison- ing (Vallee, et_§l. 1960). Measurements of 2.3 and 0.5 ppm arsenic have been made on two hot springs in Nevada and wyoming (Hem, 1959). High levels of arsenic in the aquatic environment usually result from human intervention. Introduction of arsenic laden industrial wastes along with arsenical pesti- cides are potential sources of pollution (Rainwater and Thatcher, 1960). Recent concern over the use of an aquatic arsenical herbicide, sodium arsenite (NaAst), has been ex- pressed by scientist and layman alike (Carson, 1962). Of major concern are the potential hazards to humans posed by the addition of this herbicide to ponds and lakes. Sodium arsenite is an extremely potent poison to humans. In fact, a great majority of organisms, terrestrial as well as aquatic, are susceptible to its toxicity. In addition to causing direct physiological damage from acute or chronic doses, arsenic is also suspected of acting as a carcinogen at sublethal levels (Vallee, §E_§1., 1960). Despite the hazards involved, addition of sodium arsenite to lakes has proceeded for over sixty years. Furthermore, little infor- mation is available pertaining to the distribution and ulti- mate fate of this compound after it is added to the aquatic environment. According to Timmons (1962), sodium arsenite was first used in 1902 for controlling water hyacinth in Louisiana. Details on the extent of utilization from 1902 to 1926 is obscure. However, widespread acceptance for its use in the aquatic environment followed Domogalla's (1926) work in addition to its adoption as a tool in fisheries management by Surber (1929). The low cost of the herbicide coupled with its effectiveness in eliminating nuisance weeds led to its increased utilization, especially in the states onginnesota, Wisconsin, and Michigan (Mackenthun, 1959). In Wisconsin alone sodium arsenite utilization increased from 54,012 lbs A3203 in 1950 to 182,790 in 1963. Some lakes received more extensive treatment than others. For ex- ample, Mackenthun (1964) referred to a lake which received repeated dosages of sodium arsenite amounting to 195,548 lbs of arsenic in twelve years. Following sodium arsenite treatment the arsenic con— tent of the water declines with time, but with different rates of disappearance (Dupree, 1960; Ullman, Schaefer and Sanderson, 1961; Gilderhus, 1966). Increases in the arsenic concentrations of sediments also accompany treatment (Dupree, 1960; Mackenthun, 1964; Gilderhus, 1966). Because of the non-specific toxicity of sodium arse- nite desirable as well as undesirable aquatic organisms are adversely influenced by its introduction. Surber (1931) warned against exceeding the recommended dosage of 2 ppm in order to prevent the elimination of planktonic organisms. subsequent investigations showed 2 ppm to be inadequate for controlling most nuisance weeds; present dosage recommen- dations range from 4 to 10 ppm (Davison, Lawrence, and Compton, 1962; Mackenthun, 1964). Bails and Ball (1966) ob- served a reduction in the standing crop of plankton within a few hours after addition of 8 ppm sodium arsenite to a farm pond. The same concentration suppressed phytoplankton pro- ductivity for about seven days (Sohacki, 1965). Zooplankton are also sensitive to the toxicity of arsenic. Significant reductions in the numbers of copepods, cladocera and rotifers resulting from sodium arsenite toxicity were observed by Lawrence (1958), Cowell (1965), and Gilderhus (1966). Plankton are also capable of concentrating arsenic in high quantities (Dupree, 1960). The deleterious influence of arsenic upon bottom invertebrates is well documented. Early investigations by Surber (1931) and Surber and Meehean (1931) showed that midge larvae, dragonfly, damselfly and mayfly nymphs were killed by concentrations of 2.5 ppm (A3203). More recently, Lawrence (1958), Gilderhus (1966) and Sohacki (1965) provided further data on the mortalities of bottom invertebrates following sodium arsenite applications. Wiebe, Gross, and Slaughter (1931) observed increases in the arsenic content of large-mouth bass following exposure to high ambient arsenic levels. On the other hand, Ullman, §£_§l. (1961), analyzed fish from a lake before and after treatment with sodium arsenite, but were unable to detect any significant increases in arsenic. The absence of a comprehensive ecological study deal- ing with the environmental dynamics of arsenic may be due to the tedious and insensitive methods of arsenic analysis which were formerly required. The present availability of arsenic radioisotopes and newly developed analytical techniques for the detection of arsenic now make such a study feasible. (In this study, both radioarsenic and stable arsenic analyses were used to trace the movements of arsenic follow— ing sodium arsenite introduction. The first application of sodium arsenite was conducted on July 7, 1963. In this initial application As74 labeled sodium arsenite along with a commercial brand of stable sodium arsenite was added to a small pond in sufficient quantities (8 ppm) to control aqua- tic vegetation. The 17.5 day half-life of the radioarsenic permitted detection via radioactive emissions for approxi- mately two months. Stable arsenic detection by means of standard analytical methodology was used to follow the arsenic distribution during the summers of 1964, 1965, 1966 and 1967. A second application of 8 ppm sodium arsenite was administered to the pond on July 19, 1964, about one year after the initial application. In this treatment only stable arsenic was used. Arsenic analyses in the radioarsenic and stable arsenic phases of this study were conducted on the living and non-living components of the pond. The former category included the seston, aufwuchs, macroscopic water plants (macrophytes) macroinvertebrates and fishes. The latter category included the water and sediments. The main objectives of the study were to determine the following: (1) the distribution of arsenic in the en— vironmental components following sodium arsenite application, (2) the persistence of arsenic in the environment, (3) the potential hazards posed to humans by the addition of sodium arsenite or an arsenic radioisotope to the aquatic environ- ment, and (4) the ultimate fate of arsenic when added as sodium arsenite to a pond. THE STUDY AREA This study was conducted at the Lake City Experi- mental Farm which is located about two miles south of Lake City, Michigan. The Farm, owned and operated by Michigan State University, is used mainly for agricultural research studies. The Department of Fisheries and Wildlife maintains experimental ponds and research facilities where limnological work is conducted each summer. The experimental ponds were constructed during the period 1943-1945 on a marshy area adjacent to Mosquito Creek, a small stream which rises on the Farm property. ‘A dam situ- ated on the Creek forms an impoundment which is used as a reservoir for filling the ponds. Each pond has its own inlet and outlet, and hence may be drained or filled independently (Figure 1). From east to west the ponds are designated A, B, C and D. Only pond C was used in this study. Pond C has a surface area of approximately 0.17 acres and an average depth of 3.4 feet. The pond was originally constructed on a sand base, but introduction of silty ma- terial over the years has produced a marked contrast of the sediments (Figure 2). The silt stratum overlaying the sand is approximately one foot deep at the pond spillway. water chemistry determinations made during the past few years .mofluflaflomm wuoumuonma paw mpcom Hmucoeflnmmxo huflo oxmq on» mo mm: .H onsmflm " ‘I'-“ 0'0 B‘ ‘-“ now—3233 5:33. 10 .o econ ah ecmm new page mo coausnfluumfln .N musmflm ll 12 reveal the pond waters to be moderately hard and alkaline. Dissolved oxygen concentrations normally range between 8-10 mg/liter; total alkalinity, 65-75 mg/liter and pH, 8.5-9.0. METHODOLOGY Preparation of Arsenic—74 The arsenic-74 used in this study was purchased from the Oak Ridge National Laboratory (ORNL). Fifty millicuries of As74 were prepared by bombarding a powdered germanium 74 74. Chemical sepa- target with protons - 32Ge (p,n) 33As ration and purification of the target material was conducted by the ORNL personnel. The radioisotope was ultimately de— livered in the form of labeled sodium arsenite (NaAs7402). Details on the production and separation of A374 are dis- cussed by Beard (1960). The decay scheme of As74 shows three types of emis- sions. These emission types and their energy levels in mega electron volts (mev) are as follows: positron - 1.53, 0.92 beta - 1.36, 0.69 gamma — 0.596, 0.635 (USDHEW, 1960). Counting Equipment and Procedures Beta and gamma detection equipment were used in analyzing samples for radioactivity. Those samples which could be reduced to a zero thickness layer (water and 13 l4 aufwuchs) were counted with a Tracerlab Omni/Guard low back- ground proportional counter which responded mainly to beta radiation. Samples such as fish, sediments and plants which could not be reduced to zero thickness were analyzed for gamma activity with a scintillation detector. The scintil- lation detector consisted of a 3"x3" Harshaw NaI (Tl) well crystal, Tracermatic Spectrometer and a Nuclear Measurements Corporation scaler. The spectrometer, a single channel analyzer, permitted selective detection of a single As74 photopeak with the exclusion of extraneous emissions. In this study, the spectrometer was adjusted to exclude all emissions-i 0.05 mev from the 0.596 mev photopeak. Decay Rates, Counter Efficien- cies and Activity Units When determining the activity levels of an isotope in any sample, corrections must be made for background activity, decay rate and counter efficiency. Background counts were conducted three times a day. The daily averages of these counts were subtracted from all raw counts of samples processed on the corresponding date. Two standards, made up of raw isotope, were used to estimate machine efficiencies and decay rates. Each standard 74 contained approximately one microcurie of As , and was counted intermittently during the study. The efficiency observed activity) was 0 023 for actual activity . the Omni/Guard and 0.013 for the scintillation counter. estimates for the counters ( 15 Half-life estimates for A374 determined by other in- vestigators ranged from 16 to 19 days (USDHEW, 1960). The most widely accepted value for this isotope is 17.5 days (Comar, 1955). Results from this study indicated a dis- crepancy between decay rates as measured by the scintillation counter and the Omni/Guard; 18.1 days half-life for the former and 23.0 for the latter. The 18.1 day half-life most closely approximates the literature values and is considered as the most reliable estimate. The 23.0 day half-life was based upon the detection of beta emissions, and obviously is too lengthy to be attributed to As74 doubtedly the As74 preparation was contaminated by some emissions alone. Un— longer lived, unidentified radioisotOpe. Radioactive emis- sions from the contaminant did not interfere with the results from the scintillation detector because the discriminatory features of the spectrometer excluded all counts except those of the A374 photopeak. No such discriminatory ability was possible with the beta detector. Because of the five day discrepancy in half-life determinations between the counters, two different decay rates were used in correcting the sample counts. 74 All final results of the As phase of this study were corrected to time zero, the day of the herbicide appli- cation. Results were expressed in terms of picocuries 12 (micromicrocuries or 10- curies) per unit of weight or volume. 16 Application of the Tagged Herbicide Prior to application, the tagged sodium arsenite was diluted with stable sodium arsenite and distilled water to a volume of 45 gallons. Enough stable sodium arsenite was used to bring the concentration of the pond water to 8 ppm. After complete mixing, the solution was pumped from the reservoir via a submersible pump through Tygon tubing to a floating applicator. The applicator, constructed of 3/4 inch copper pipe, had 18 nozzles along its 16 foot length. The herbicide mixture was pumped through the nozzles while the applicator was towed across the pond‘s length three times. The rate of application was timed to provide a uni- form distribution of herbicide to all areas of the pond. Treatment of the pond, which was conducted on July 7, 1963, was completed in approximately one hour. water and Seston Water samples were obtained with a 1 liter poly- ethylene sampler from the surface, 15 and 30 inch depths. Three samples were obtained from each depth; all sampling sites were selected randomly. Processing of the samples im— mediately followed sampling. Each sample was usually divided into two 100 ml aliquots for analysis, but during the latter stages of the study when the radioactivity levels were low, 200 ml aliquots were used. One aliquot was evaporated to dryness and counted without any further processing. This 17 fraction of the sample was referred to as unfiltered water and contained the radioactivity of the soluble plus sestonic arsenic. The second aliquot was filtered through a 0.45 micron Millipore Filter to remove the sestonic components. The radioactivity contained in this sample represented the soluble arsenic fraction alone. The difference between the radioactivity in the two samples represented the activity retained by the seston, or the living and nonliving particu- late matter in the pond water. All samples were counted in a Tracerlab Omni/Guard low background counter. The water samples along with all other samples which required drying were dried in a forced air oven maintained at 60°C. Low temperature drying was necessary to avoid volatilization of arsenic (Lanz, Wallace, Hamilton, 1950). Bottom Sediments Core samples of bottom sediments were obtained with a polyethylene sampler which extracted a core 2.8 cm in diameter. Three cores comprised a sample. The cores were frozen immediately after sampling; subsequently the cores were sectioned into 1 cm slices. After drying and weighing were completed, the slices were counted in a well counter. Aufwuchs Aufwuchs, or the assemblage of plants and animals which are attached to but do not penetrate the substrate, were collected on plexiglass substrates exposed at the 15 18 and 20 inch depths. At least seven days accrual was allowed prior to sampling. Each sample usually consisted of the pooled growth from four substrates, however, eight were pooled towards the end of the study. After removal from the pond, the aufwuchs was scraped from the substrates and filtered out on a preweighed Millipore Filter (0.45 micron pore size). Filtration was allowed to proceed for 30 seconds beyond the time when the water gloss disappeared from the filter surface. Wet weight of the aufwuchs was determined immediately following filtration. After weighing, the filter was glued to a planchet, dried, and the aufwuchs activity assayed in the Tracerlab Omni/Guard counter. Macrophytes The native pOpulation of macroscopic aquatic plants (macrophytes) included Anacharis canadensis, Potamogeton natans and Chara globularis. In order to increase the va- riety and abundance of macrophytes in the test pond, trans- plants were introduced from a nearby lake. The transplants included Myriophyllum.§p., Potamogeton praelongus, Potamo— geton zosteriformis, and Potamgqeton Robbinsii. Those plants which survived the transplanting appeared to be healthy by the time the study commenced. In order to facilitate the sampling of plants in the latter stages of decomposition, potted plants were set out on a submerged platform located at the pond spillway. 19 Because of their accessability, the pots could be removed from the pond with minimal disturbance to the plant contents. Following removal from the pond, plants were trans- ferred to the laboratory for immediate processing. The plants were blotted, weighed and rinsed in 0.01N HCl to re— move any adsorbed activity. Subsequently, these samples were dried and reweighed. Plants were counted in the well counter. Fishes Fishes studied in this project included the following: 1. Yellow Perch - Perca flavescens 2. Green Sunfish - Lepomis cyanellus 3. Pumpkinseed - Lepomis gibbosus 4. NOrthern Black Bullhead - Ictalurus melas 5. NOrthern Fathead Minnow -_gimephales promelas 6. Golden Shiner - Notemigonus crysoleucas. Fish were normally captured by trapping, however hook and line methods were sometimes considered necessary. The captured fishes were washed with 10 m1 of 0.01N HCl to re- move the protective mucus along with any adsorbed activity. Total length and weight was then determined. After these procedures were completed, the smaller fishes were counted intact. The larger fishes were eviscerated and the internal organs counted separately from the musculature and skeleton. All fish samples were counted in the well counter. 20 Invertebrates Invertebrates included in the study were Anodonta ‘§2., a fresh water mussel; Cambarus sp., a crayfish, and a variety of immature aquatic insects. Dragonfly naiads and midge larvae predominated the latter group. All inverte- brates except the mussels were counted without any further processing. The valves of the mussels were removed prior to counting. Phytoplankton Productivity Phytoplankton productivity was measured with the 014 method as described by Steeman Nielsen (1952) with modifi- cations by Rhode (1958). Rhode (1958) recommended that light and dark bottles be exposed from sunset of one day to sundown of the next. Water was collected with a polyethylene sampler and transferred to two light and one dark bottles. Two microcuries of C14 as sodium carbonate were injected into each of the samples which were then stoppered and suspended at the 32 inch depth. After exposure, the samples were taken to the laboratory, filtered through 0.45 micron pore size Millipore Filters, attached to planchets and stored for future counting. All samples were counted with a gas flow pro- portional counter and the final results expressed in terms of milligrams of carbon per square meter per day (mg C/mZ/day)- 21 Stable Arsenic Because of its short half—life the distribution of As74 could be traced effectively for about 60 days. In order to further pursue the dynamics of arsenic in the en- vironment, analysis of stable arsenic residues was employed. Samples for stable arsenic analysis were first collected prior to the initiation of the radioisotope study; the arsenic content of these samples served as a baseline for, post-treatment comparison. Subsequently, samples were ob- tained during the summers of 1964, 1965, 1966 and 1967. With few exceptions, the sampling procedures used in the stable arsenic phase of the study were the same as those just described. Three notable differences in sampling pro- cedures are as follows: (1) seston samples were obtained by centrifugation of pond water, (2) aufwuchs accumulation was allowed to proceed for one to two months before sampling, and (3), core samples were sectioned immediately after sampling. All samples except the fishes were dried at 60°C.; the fishes were stored in the frozen state. Stable arsenic concentrations of the samples were de- termined by using the silver diethyldithiocarbamate (SDDC) method as proposed by Vasak and Sedivec (1952). Because of its sensitivity, the SDDC method is considered superior to the older analytical methods for detecting arsenic (Ballinger, Lishka and Gales, 1962). Further evidence of its reliability 22 is attested by the inclusion of this method in Standard Methods (1965). Because all samples analyzed in this study contained organically bound arsenic, conversion into the free inorganic state was imperative for analysis. In order to achieve this, a digestion procedure using equal parts of nitric and per- chloric acid was employed. Samples were digested in 125 ml Erlenmeyer flasks with 10 ml of the digestant. Additional nitric acid was added if the samples exceeded the weight of one gram. The flasks containing samples and digestant were heated on a hot plate until dense white fumes of perchloric acid evolved. If the sample appeared clear and colorless at this point 20 ml of distilled water was added and the sample again boiled to perchloric acid fumes. Samples exhibiting color were allowed to boil until the color disappeared. Oc- casionally a light yellow tint persisted in the sample de— spite prolonged boiling with perchloric acid. This color- ation may have been produced by iron (Analytical Methods Com- mittee, 1960) and appeared to have no effect upon the de- termination. After boiling the samples to white perchloric acid fumes the second time, sufficient distilled water was added to bring the volume up to 35 m1. This completed the digestion procedure. The SDDC methodology as recommended by the Fisher Scientific Company (1960) and Ballinger, Lishka and Gales (1962) were used in this study. Standards ranging from 0.5 23 to 10 micrograms arsenic were included in each series of determinations. The calibration curve computed from 168 analyses was fitted via the least squares method (Figure 3). Using the computed constants, the equation describing the curve may be expressed I = 0.0096 + 0.0314X. The 95 per cent confidence limits for individual estimates of X was calculated to be-i 1.25. Theoretically, in 100 measurements of a sample (X) at least 95 will fall within the limits X-1 1.25 micrograms arsenic. In the past, many investigators expressed their arsenic data in terms of arsenic trioxide (As203). Results presented here are expressed in terms of elemental arsenic (As). Only one deviation from the rule exists; the 8 ppm treatments of sodium arsenite are expressed as A3203. 24 .moumEHumo Hmspfl>flpcw How mam>noucfl oocopflmcoo fimm on» mCHBOnw UHcomum How o>uso COHOMHQHHMO .m ousmflm 25 48 3.3.- 323.3: .0 4 d —.o «.0 9.0 Kauuqlosqv' RESULTS AND DISCUSSION Water Radioisotope Study In the past, a number of investigators used stable arsenic analyses to study the persistence of arsenic in the water of lakes and ponds following sodium arsenite appli- cation. About one-half the arsenic disappeared from pond water after thirty days in Dupree's (1960) study. Doepke's (1963) data show about a 60 per cent decrease in thirty days. Gilderhus (1966) observed very slow losses in cement tanks, about 60 to 80 per cent after 16 weeks. .In this study, radioarsenic disappeared from the water rapidly for the first 19 days following treatment and then slowed considerably for the remainder of the study (Figure 4). Similar observations were made by investigators who used radiophosphorus in the aquatic environment (Hayes, McCarter, Cameron and Livingstone, 1952; Rigler, 1956; Harris, 1957; Whittaker, 1961). Bahr (1966) noted a similar pattern of A574 loss in his aquarium project. According to Hayes, g£_§1. (1952), an element which is added to the aquatic environment will disappear at an exponential rate and enter into the solids (sediment, flora 26 27 Figure 4. Decrease in radioarsenic from the water of the test pond. The mean and standard error for each observation was calculated from nine samples. The line was calculated. 28 10" v.3 u 532?. . Days 29 and fauna) of the ecosystem. When a significant fraction of the element has entered into the solids, there is a feed- back into the water. The feed-back becomes more pronounced as the concentration in the solids increases. During the initial stages loss from the solids is negligible since very few atoms have entered this phase. At this time, the curve of decline from the water is indistinguishable from a loga- rithmic curve. This is shown during the first 19 days of arsenic loss from the water (Figure 4). As the arsenic concentration of the solids increased, exchange with the water became more pronounced and tended to bolSter the levels in the water. This appeared as a retardation in the rate of water-arsenic loss during the latter thirty days of the study (Figure 4). The increase and feed-back continue until a steady- state is established between the water and solids phase. At this stage (the asymptote of Figure 4) the amount of isotope leaving the water and entering the solids is balanced by the amount returned to the water from the solids. Hayes,_g£_gl. (1952), used the following model to describe this phenomenon: dN '5? = -N(Xfu) + Nb. N is the amount of element present at time t, and No is the concentration in the water at time to. The symbols ).and‘A represent the losses as percentage per day from the water and solids respectively. By using this equation the exchange 3O rates of arsenic between the water and solids were determined. These rates are expressed in terms of half-lives and turn- over times (see Appendix 1 for calculations). Turnover time refers to the time required to replace a quantity of arsenic equal to the total amount in the water or solids phase. The turnover time for arsenic in the experimental pond water was 9.8 days (half-life 6.8 days) and 72.5 days (half—life 50.4 days) in the solids. Bahr (1966) observed different rates of disappearance in his A574 aquarium study; the differences appeared to be related to substrate type and the presence or absence of organisms. Turnover times calcu- lated for phosphorus in lake waters range from a few days to a month, whereas the solids range from five to ten times longer (Harris, 1957). The contrast in turnover times and rates of disap— pearance of phosphorus and arsenic measured by radioisotopes or analytical methods is undoubtedly related to the abiotic and biotic composition of the environment. The biotic in- fluence, which has not received too much emphasis, is never- theless of extreme importance (Lee and Hoadley, 1967). The presence or absence of bacteria greatly affected cycling of phosphorus (Hayes and Phillips, 1958). Rapid turnover rates have been shown to exist for phosphorus in phytoplankton, periphyton and microcrustacea (Whittaker, 1961). These studies indicate that all organisms in the ecosystem in— fluence the cycling of an element by differential uptake, 31 retention and elimination. In essence, turnover times re— flect the combined interactions of the living and nonliving components of the ecosystem with the element under study. Since all environments differ in abiotic and biotic compo— sition, arsenic cycling would be expected to vary between each. Consequently, the exchange rates measured here may differ markedly from the rates in other ecosystems. In fact, results from other studies (based on stable arsenic analyses) showed slower rates of arsenic loss than measured here. Dupree (1960) and Fish (1963) observed a layering of arsenic after sodium arsenite treatment. In order to detect any evidence of stratification in this study, comparisons were made between water samples collected at the surface, 15, and 30 inch depths. A two way analysis of variance was used to test the effects of depths and dates (Table 1). A signifi- cant difference in times reflects the decreasing concen- trations of arsenic in the water. However, no significant difference existed between depths indicating a uniform distri- bution of arsenic at all levels in the pond. Stable Arsenic Study Chemical analyses of the stable arsenic concentrations in the experimental pond were conducted on water sampled prior to the A374 study and during the four ensuing years. During this time the pond received one more treatment of 8 ppm sodium arsenite. 32 Table 1. Two way analysis of variance testing the effects of dates and depth on the arsenic content of pond water. Degrees of Sum of Mean sum Source freedom squares of squares F Depths 2 10191795 5095897 0.357X * Dates 15 89510040322 5967336021 418.065 Interaction 30 573585476 19119515 1.339x Subtotal 47 90093817593 Within 96 1370275816 14273706 Total 144 91464093409 * denotes significant difference at the .05 level. xdenotes no significant difference at the .05 level. Pretreatment stable arsenic concentrations in the pond existed in trace quantities (Table 2). Sediment samples acquired at the same time also showed low arsenic levels indicating that the pond had never received any previous ap— plications. Post-treatment levels in the water ranged from 1.8 to 6.6 pg/liter during the four year study, with the ex— ception of two sampling dates. During September, 1964 an ex- tremely high arsenic concentration was measured; this re- sulted from the treatment on 7/19/64. On the other hand, extremely low levels were measured in September, 1967. Be- cause of a leak in the spillway, I cannot ascribe the re- duction in arsenic to natural causes; the low levels may 33 have been produced by dilution via inflowing water. In spite of this unfortunate incident, several conclusions may be reached from these data. Table 2. Stable arsenic concentrations of pond water. Sampling Mean Standard error date pg/liter yg/liter 7/63 tr 6/64 1.76 0.70 9/64 281.64 39.12 6/65 2.05 0.48 7/65 6.59 1.16 9/65 3.91 0.74 6/66 3.44 0.47 9/66 4.21 0.10 9/67 0.68 0.05 tr = trace The introduction of sodium arsenite to a pond de- finitely provided arsenic with an active and persistent role in the environment. The treatments increased the arsenic concentration of the water to above natural levels for at least two years. Although these concentrations were low, arsenic availability to organisms via the water medium was evident throughout the entire study. Specific details on this tOpic will be covered in later sections. 34 Sediments Radioarsenic Study Two distinct types of bottom sediments existed in the experimental pond, sand and silt (Figure 2). Because of the unequal particle range exhibited by these sediments, a stratified sampling technique was employed to obtain repre— sentative samples from both areas of the basin. A two way analysis of variance was used to test the effects of sediments and dates upon radioarsenic levels (Table 3). The signifi- cant difference (0.005 level) shown by dates indicated a change of arsenic concentration with time, as was expected. However, the sediments (sand and silt) also showed a signifi- cant difference (0.005 level), indicating a contrast between the arsenic content of each sediment type. This information coupled with the fact that silt samples consistently ex— hibited higher activity levels than sand implicated silt as having the greater affinity for arsenic. .According to Sayre, Guy and Chamberlain (1963) the most important characteristic of sediments which influences ion absorption capacity is particle size. The surface area of one gram of silt averaging 0.01 mm in diameter has roughly 25 times the surface area of sand 0.1 mm in diameter. The area per unit weight of sedi- ment is determined by particle size and the absorption ca— pacity is approximately proportional to area. Silt, because of its greater area, will sorb a greater quantity of ions. 35 than sand. This fact was borne out by the result of this study. Table 3. Summary of a two way analysis of variance testing the effect of time and sediment composition upon activity density in the 1 cm sediment layer. Sum of Mean Source Squares df Square F Dates 11994304.97 17 705547.35 3.69* Sediments 6189974.28 1 6189974.28 32.42* Interaction 4157377.38 17 244551.61 1.28 Error 13744702.44 72 190898.64 Total 36086359.07 107 * Denotes significance at the 0.005 level. Past experiments with radioisotopes in the aquatic environment established relationships between the activity of the water and that of the sediments. Hayes, et al. (1952) and Whittaker (1961) working with radiophosphorus, also Bahr (1965) with radioarsenic witnessed a gradual increase in sediment radioactivity accompanying the decrease in water radioactivity. In this study no such relationship was ob— served, instead sediment radioactivity fluctuated irregularly (Figure 5). The results of Figure 5 suggest that the pond sediments were involved in arsenic exchange with other en- vironmental components. However, these results pose an Figure 5. 36 Activity densities of pond sediments show- ing fluctuations in arsenic content. Mean and standard error for each observation calculated from at least three samples. Curve drawn by freehand. -3 pt]: (dry weight) x 10 1.0;- 200 r- I.5 r- O 0 an Silt llillllltllltllltlllltLlllJllllJlllljlllljllllljtlu N o O 1 '05 '- 1.0 F‘ 0.5 "‘ lo ' 20 30 40 50 Days 38 interesting question. What causative factors were responsible for the abrupt changes in sediment radioactivity? Sorption of ions by the upper layers of the sediments have been attributed to inorganic as well as organic ex- change mechanisms. Sayre, et al. (1963) maintain that physi— cal sorption and ion exchange are the predominant mechanisms responsible for sorption in river sediments. The degree of sorption is governed by (l) the specific nature of the sor- bent solid and sorbed substance, (2) available surface area, (3) concentration and types of solute in solution, (4) time, and (5), temperature. .Furthermore, the sorption process is usually reversible; changes in the ambient pH may result in a release of ions back into the water (Macpherson, Sinclair, and Hayes, 1958). However, Hayes (1964) concluded that the ion exchange at the mud-water interface is regulated not only by inorganic exchange mechanisms, but organic as well. Bacteria were especially influential in regulating the ion exchange in his study. Two observations led me to believe that biological influence was mainly responsible for the sediment activity fluctuations. Firstly, if the fluctuations were caused by inorganic reactions such as pH or redox changes in the sedi- ments, such alterations would be reflected in the water chemistry of the pond. No such changes were observed. Secondly, the activity of the aufwuchs community, which is closely associated with the sediment surface, fluctuated 39 irregularly like the sediments (Figure 15). Since the aufwuchs population of the artificial substrates may have differed in composition from the sediment aufwuchs, exposure and reaction to the addition of arsenic may have also differed. Consequently, it was not surprising to observe the lack of synchronization in the radioactivity cycles. These data suggest that the epipelic aufwuchs may play an extremely important role in the exchange of arsenic and possibly other elements in the aquatic environment. Future investigations of a similar nature should definitely explore the possible influence these organisms may wield in the exchange of elements. Stable Arsenic Study Soils of the terrestrial and aquatic environments manifest a remarkable ability to concentrate arsenic when ex- posed to arsenical pesticides. Bishop and Chisholm (1962) reported concentrations of 9.8 to 124.4 ppm arsenic in orchard soils which were exposed to lead arsenate (PbHAsO4) treatments, whereas the non-treated soils ranged from trace concentrations to 7.9 ppm. Arsenic concentrations remain relatively constant in the top soil layers despite discon- tinued use of the pesticides (Bear, 1957; Vallee, et a1. 1960; MacPhee, Chisholm and MacEachern, 1960). Most of the arsenic is retained in the upper five inches of the soils (Arnott and Leaf, 1967). All of the investigators cited 40 above reported inhibition of plant growth as a result of arsenic accumulations. Accumulations of arsenic in the sedi— ments of ponds and lakes have been reported by Dupree (1960), Doepke (1963), Mackenthun (1964), Leuschow (1964) and Gilderhus (1966). Arsenic appears to persist for an extended length of time in the top layers of terrestrial soils; how- ever, its long-term distribution in the aquatic sediments has not been investigated. The stable arsenic concentrations in silt continued to exceed the levels in sand for the duration of the study. Although the first analysis of core samples showed arsenic residues to be significantly greater than the pretreatment estimates, a further increase accompanied the July, 1964 treatment (Figures 6 and 7). All sand layers reflected this increase in arsenic by September, 1964 (43 days after treat- ment) indicating that the arsenic penetration in sand was more rapid than in silt. Silt did not respond similarly. Arsenic levels in the sand gradually declined during the following years. Arsenic diminishment in the silt was only apparent in the top 1 cm layer. Any gradual trends of di— minishment in the 2 and 3 cm layers of the silt were obliter- ated by the variability of arsenic concentration, which in turn was caused by the variation in sediment composition of the silted areas. Because the silt was superimposed upon a sand base in varying thicknesses (l to 30 cm), many of the lower layers consisted of an admixture of silt and sand. 41 .monEmm wmunu ummoa um Eoum popmasoamo coHum>HomOo some How Houuo pumpcmum cam :60: .vo\ma\h paw m©\h\h pmuospcou mucoEumoHu ouflcomum Edfipom .Ucom umou ecu Eonm pcmm mo moHQEmm ouoo 029 CH mcoHumnucoocoo oesomum magnum .o ousmflm 42 If 19-6 PF ”-1 ' :9-9 79-6 L... 79-9 lg ! l9 -6 99-9 It 99-1 99 -9 79-6 , L79-9 l 1 . , l l 1 L ‘ .gcq-l " . 99-1 S9 -9 79-6 v9-9 . , , , I , . . ifi—Egco-z l O in c F 13“!“ UP) 3! Iowan! In: J [29-6. Sampling Periods (math-you) tr :trau 43 .monEmm woman uwmwa um Eoum pmumasoamu cofium>uomno some now Honuo pumpnmum paw amoz .¢©\ma\h Ucm m©\>\n pouospcoo mucoEumoHu ouflcomum Seapom .pcom umou may Eoum uHHm mo moHQEmm ouoo CH mcoHumHucoocoo oacomum magnum .h musmflm 3cm 2cm 1F 44 1cm 1r. r 150 125 o In 0 In 0 Q N In N ”as!“ up) SIIGQUOSJ! 3m 19-6 99-6 59-1 99-9 79-6 89-1 ‘29-6 99-6 99-9 99-1 99-9 79-9 89-1 19-6 99-6 99-9 99-1 99-9 79-9 89-1 Sampling Periods (month- yaarl tr = true 45 Since the sorption properties of these two soil types differ markedly, samples containing a larger percentage of sand yielded less arsenic than those comprised wholly of silt. In contrast, the sand area of the pond appeared to be more uniform in composition and exhibited far less variation in arsenic content. The steady decline in the surface layers of the sedi- ments implies either an exit from the sediments to the water and/or plants, or penetration into the deeper layers of the sediments. Evidence indicated that both processes occurred. As has already been mentioned, some arsenic moved from the solids (sediments included) into the water. Plants also acquired concentrations of arsenic. However, a considerable quantity also moved into the lower layers of the sediments. Deep core samples from the last sampling series (September, 1967) showed high arsenic concentrations at the 10 cm depth. Furthermore, about one-third of all core samples taken during this study exhibited higher concentrations at the lower levels than at the surface (Figure 8). This suggests that the arsenic may be either leaching into the deeper sediments or being overlayed by shifting sediments. Because of the wind protection afforded by the dikes that surround the pond, plus the minimal quantity of sediment introduction the latter possibility was rejected. The porous nature of the basin permits a slow seepage of water into the underlying strata. Perhaps this movement promoted ion exchange processes with 46 .pcmnooum ma c3mup mo>uso .Ahmma .HoQEoumomv weapon UCHHQEmm umma ecu ca pocflmuno 0603 monEmm nuom .Homma EU «Im on» um mCOHumuucoocoo goes on» mcflzonm meEmm ouoo pawn m can pawn m mo mmaflmoum oesomufl .m musmflm 2.3:. :3 9. 3.3:. a... O— Q 0 Q N 00— On 00 0* ON 2.3 :5 an n! mdau 48 accompanying movement of arsenic into deeper areas of the sediments. Regardless of the process involved, the important aspect of this observation is the decrease of arsenic from the surface layers of the sediments. The arsenic added via sodium arsenite treatment appears to become "buried“ in the deeper sediments of lakes and ponds, and thereby loses its active role in the environment. The rate of disappearance into the sediments for any lake or pond would depend mainly upon the sediment characteristics, varve formation and seepage. Macrophytes Radioisotope Study All macrophytes except Chara globularis, a resistant alga, were killed by the sodium arsenite treatment. Within three days after application, about 95 per cent of the vascu— lar macrophytes slumped to the bottom. Within a few days thereafter all plants yellowed and showed signs of decompo- sition. Some plants, such as Anacharis canadensis, decom- posed more rapidly than others. Sampling of such plants terminated early because of the difficulty experienced in re- -trieving samples. On the other hand, Myriophyllum EB decom- posed more slowly and was sampled throughout the entire study. Two weeks before the study ended, new growth de— veloped from the roots of the "dead" Myriophyllum sp and 49 Potamogeton praelongus. Apparently these species only suffered a temporary setback. Uptake of A374 by plants proceeded rapidly following the addition of sodium arsenite to the pond. This is at- tested by the concentration factors calculated for the plants sampled within two hours after treatment (Table 4). The rapid uptake suggests that the leaf and stem surfaces were more instrumental in arsenic sorption than the roots. Sedi- ment activity at this time was too low to account for the high plant accumulations via the roots. Bahr (1966) observed similar responses by plants in his A574 study. Few other studies have been conducted to determine the mode of entry by arsenic into aquatic plants. Rigg (1955) found arsenic was sorbed near the soil level of Myriophyllum verticillatum and Anacharis canadensis stems. She also suggested that arsenic was absorbed mainly on free surfaces such as the cuticle of leaves, epidermal walls of stems and walls of air chambers and lacunae. .Many aquatic biologists consider sodium arsenite toxiCity to be effective towards vascular macrophytes only. However, it also is used for controlling filamentous algae (Surber, 1943; Lawrence, 1958). The algicidal potential of -sodium arsenite was manifested in this study. A luxuriant growth of Spirogyra Sp prevailed in the test pond prior to treatment. Within 24 hours after treatment only a few scattered fragments of the plant remained and these were in 50 Table 4. Accumulation of As74 by aquatic plants within two hours after treatment. * Plant Concentration factor Algae Spiroqyra_§p. 136.3 137.6 171.1 Chara globularis 31.8 34.5 38.0 Submerged vascular plants Myriophyllum.§p. 38. 34. 33. Elodea canadensis 3. 10. 11. Potamogeton praelongus ll. 14. Potamogeton zosteriformis 24. 19. 21. ** Potamogeton Robbinsii 10. 12. 17. bOl-J bow \le \OO‘H comm Floating leaved plants Potamogeton natans 42.3 13.0 30.6 * _ . Concentration factor = pc AS 74/9 plantgjwet weight) pc As-74/ ml water ** Concentration factors determined 26 hrs after treatment. 51 the advanced stages of decomposition. After 48 hours, no traces of Spirogyra could be found. New growth appeared twice before the study terminated. In addition to exhibiting an extreme susceptibility to sodium arsenite, Spirogyra sp also showed an unusual af- finity for arsenic (Table 4). Activity densities of Spigo— .HYEE sp on day zero were on the average six or more times greater than other macrophytes (Figure 9). The high affinity for arsenic was once again demonstrated when new growth ap— peared on day 16. The new growth never persisted for any length of time but disappeared after accumulating arsenic. Whether or not arsenic toxicity was responsible for these disappearances could not be established. On the other hand Chara globularis, a non—filamentous alga, reacted quite differently to the toxicity of sodium arsenite. Despite accumulation of high arsenic concen— trations this plant appeared to suffer no adverse effects (Table 4). The resistance of Chara species to sodium arsenite is mentioned in most herbicide application manuals and has been demonstrated several times in laboratory and field ex- periments (Rigg, 1955; Bails and Ball, 1966; Sohacki, 1965). The latter investigator demonstrated that the alga continued to grow during the recovery stages of a sodium arsenite application, indicating that physiological impairment, if any, was slight. 52 m m u n o I O O .m onsmam 53 )- . a O M I,” 'I 0 g 2 o In £_DI X (mm Lip) and 20 10 Days 54 The reaction of g. globularis to the presence of arsenic bears special significance since it was the only macrophyte to survive the herbicide treatment without suffer- ing any adverse effects. The pattern of uptake and loss by this plant may reflect the reactions of healthy plants to sublethal doses of arsenic. The initial uptake and rapid loss of arsenic during the first five days resembled the re- action of a plant to the increased availability of a de- ficient nutrient (Figure 10). That is, the plant incorpor- ated the element in quantities greater than was necessary for metabolic sustenance (luxury consumption). Although arsenic is not an essential element, it is commonly con- sidered to act as a biochemical mimic of phosphorus. The active incorporation of arsenic by the plants possibly re- sults from a physiological inability to distinguish between arsenic and phosphorus. In fact, the initial gain and loss of arsenic by_g. globularis was very similar to the incorpor- ation and elimination of radiophosphorus in water plants (Ball and Hooper, 1963). After the initial decrease, the arsenic concentration of g. globularis remained relatively stable for the remainder of the study. I believe the plant established a steady state interrupted slightly by minor shifts in the equilibrium which produced slight alterations in the activity density. Because .9- globularis had a larger standing crop (65.6 g.wet weight/ m2) than any other organism in the pond, shifts in the 55 .CCmnoon an CBmHm.o>Hsu .ucoucoo oacmwuw ca wcofiumsuosam onu mcflzonm mammasnoaw mumzo mo muamnop >ua>HpU< .OH musmflm «as: 06 an ac on o« o— “aqua-a-aq-qdaW—q«11—11-1....JSJ-uda-du—unq-—..qu—u-d .1 ~qiuq O P In F ON s_Ol x (when up) 3/0d 57 activity density, however slight, must have had a signifi- cant effect upon the distribution of arsenic in the pond. Each of the susceptible vascular macrophytes which eventually died from arsenic toxicity absorbed and retained different quantities of arsenic (Figure 11). Myriophyllum sp exhibited signs of luxury consumption with subsequent loss; however, no other vascular macrophyte behaved similarly. One unusual phenomenon observed was the increase in activity density by some plants after death. As mentioned earlier, nearly all plants lost tur— gidity after three days, suggesting that extreme unhealthi- ness or death prevailed at the time. Presumably the cell integrity was impaired or completely disrupted. Nontheless, arsenic concentration continued to proceed (Figure 11). In Anacharis canadensis which exhibited the most pronounced post— mortem increases, arsenic concentration continued although the plants were transformed into a brown, viscous, amorphous mass. A possible explanation for the increased arsenic concentrations by the dead plant tissues may reside in the decompositional sequence of plant material. According to Allgeier, Peterson and Juday (1934) decomposition of plants in the aquatic environment proceeds rapidly for 6-8 days and then slows down to a fairly constant rate. During the first phase of decay the more readily decomposable tissues are mineralized, followed later by the more resistant types. 58 newcoaomnm.couow05muom .m mm ESHHMBQOHHNZ .w awmcflnnom couomoamuom .m wHEHOMHHOHmON COflOUOEMHOfi . N mflwcopmcmu maumnumc¢ .H .muCMHQ oaumsvm pomumfinsm Mo mofluamcop mufl>wuo¢ .HH ousmflm sure.» to: o guxehu _n=_u_.o . was: an ac an cu o— On ov an on o— _.:._...._...._.:._...._ ._ _ _ _ .._u..._m.., 13.3... J ._ .._ .H._...._n..._...._...._...._....,o ” m I a a o annamnwm n a” m 0 H H "a a u a a a m.....:..”” . . . Up...” a u . . . .......H.n... my... 0 o— m— an a as 1 a an. o on 0 an on. u... u a a. oo o u s t 6 l o 1 o I U l 1 l l n n l cpl x (wagon up) and 60 Assuming the decomposition progressed as described, then the increased activity exhibited in the A374 study was probably due to the increased ratio of decay resistant tissues. This suggests that the readily decomposable plant constituents such as the cytoplasm retains less arsenic than the more re- sistant tissues such as the cellulose. The persistence of arsenic in the decomposition resistant tissues indicates that the macrophytes, although dead, continue to influence the distribution of arsenic in the pond. Before the study terminated, new growth developed from the roots of Myriophyllum sp and Potamggeton praelongus. Some of these living specimens incorporated arsenic concen- trations which were equal to or greater than that required to "kill" the original growth. Yet these plants were apparently unaffected (Figure 11). Perhaps plants are capable of de- veloping a tolerance to arsenic as suggested by Fish (1963). Potamogeton natans, the sole floating-leaved macro- phyte in the pond, initially concentrated high levels of arsenic in its tissues (Table 4). Because of the limited abundance of this plant, sampling completely depleted the standing crop before any definite trends in arsenic uptake were established. Consequently, I do not know whether these plants suffered any setback as a result of the treatment. Floating plants usually resist the toxicity of sodium arse- nite. New 3. natans plants appeared the following year, so evidently the roots survived. 61 Marginal vegetation, which included Scirpus sp, Carex sp and Phalaris arundinacea, was analyzed to detect the possibility of lateral arsenic movement from the pond to the terrestrial environment. Significant counts were oc- casionally manifested in a very few plants. Results of these analyses provided no evidence of a lateral arsenic movement from the pond. Stable Arsenic Study The herbicide treatments of 1963 and 1964 selectively altered the species composition of the pond flors. Anacharis canadensis, Potamogeton Robbinsii, and Potamogeton zosteri- formis were totally eliminated, whereas Myriophyllum sp and Potamogeton praelongus recovered from both treatments. In addition to altering the species composition, a complete change in dominance was manifest. During the years 1963, 1964 and 1965 g. globularis maintained dominance in the pond despite gradual encroachment of Myriophyllum sp and Potamo- geton praelongus. By September 1967, dense beds of the latter plants covered the entire basin and a very scant growth of_§. globularis prevailed. Stable arsenic concentrations of the macrophytes con- sistently exceeded the pretreatment values during the four year study (Figure 12). However, a great deal of variability in arsenic concentrations was evident between sampling periods. The most marked increase, shown by_g. globularis 62 Figure 12. Stable arsenic concentrations of sub- merged macrophytes from the test pond. Sodium arsenite treatments conducted 7/7/63 and 7/19/64. 2O 15 IO 20 15 IO In; arseaicllt: (dry weight) 20 15 10 63 2 12 Chara llohularis " t‘. 63 r' Myriophyllum :9 L... F“ L. )— tr ‘3 r" Potamogeton praelongus r— tr 1 3 33333333 'lllltlll 000050.006 Sampling Periods l month-year) 64 on September 1964 resulted from the July 1964 sodium arse- nite treatment. Explanations of the other increases and de- creases would be pure speculation. Pendleton (1965) ob- served similar variability in the concentration of cesium-137 by submerged plants; he attributed the changes to seasonal and environmental alterations. The submerged plants were capable of accumulating arsenic in levels greater than 1,000 times that which ex- isted in the water (Table 5). Myriophyllum sp and Q. globu- laris consistently exhibited higher arsenic levels than_g. praelongus. Despite the fact that high levels of arsenic were incorporated into their tissues, the plants appeared as healthy as those growing in non-treated ponds. Other in— vestigators made similar observations. Doepke (1963) measured as high as 38 ppm in Ceratophyllum demersum and 56 ppm in Anacharis canadensis which grew healthily in a sodium arsenite treated lake. Fish (1963) measured 23.5 ppm natural occurring arsenic in the tissues of Lagarosiphon sp. The concentrations in the latter rose to 76.0 ppm after a sodium arsenite treatment, but no ill-effects were shown by the plants. Fish (1963) concluded that aquatic plants, like man, may develop a tolerance to lethal doses of arsenic. The notable reduction in arsenic concentration of all plants in 1967 warrants further discussion. As mentioned earlier, the arsenic content of the water fluctuated from about 2-7 pg/liter during the years 1964 to 1966. The 65 Auoum3 pcomv Houaa\oficomno oanmum m6 M AunmfloB uozq Osman wx\0flcomum oanmum m8 M." Houumm coflumuuco0:00« v.ma N.mH H.m® hw\m m.wam m.mm¢ m.m¢o oo\m m.mm m.a¢o H.mmm oo\o m.ona m.mmo h.¢mm mm\m m.mmm o.mMMH m.m¢m mm\h m.Hm m.onNH m.nmo mm\© «.ho vo\m o.mov N.N¢HH m.mmv vo\o nouomw .ocoo nouomm .ocou snouomm .ocoo msmdoaomum.couom05muom .mm Esaamnmowmwz mammasnoam mumno oumn .homa 0cm coma .moma .voma mnmom one meansp magmam oaumsvm pomnoensm an oacomum oanmum mo coaumasfisoo< .m oanme 66 arsenic content of the plants also fluctuated but maintained high levels, especially in g._globularis and Myriophyllum sp. In 1967, the concentration of arsenic in the water de- creased to 0.68‘pg/liter. A marked reduction in plant arsenic was also apparent at this time. These data suggest that (1) these aquatic plants lose their ability to accumu- late high levels of arsenic when the ambient concentrations fall below 2,u9/liter, and (2), arsenic uptake is dependent mainly upon stem and leaf surfaces rather than roots. If root absorption were primarily responsible for arsenic up— take, the decrease in tissue concentrations should not have been so severe because of the ample arsenic supply in the sediments. In Chara globularis the thallus and rhizoids have an equal absorptive capacity (Littlefield and Forsberg, 1965). From the above observations, the thallus was judged to be the most influential in arsenic uptake. Seeds, stems and leaves of_g. praelongus were analyzed separately to determine whether equal concentrations of arsenic prevailed throughout the plant. A one way analysis of variance showed these plant parts to differ significantly in arsenic concentration (Table 6). Further analysis via a multiple range test revealed that the leaves contained sig- nificantly higher arsenic levels than the seeds and stems. The latter exhibited no differences. 67 Table 6. Summary of the one way analysis of variance test- ing the stable arsenic concentrations in leaves, stems and seeds of Potamogeton praelongus. Source of variation Sum of Mean df F squares square Among groups 2.329 1.164 2.0 11.472* Within groups 1.218 0.101 12.0 Total 3.547 14.0 * Significantly different at the 0.05 level. Leaves Stems Seeds *9: Means 1.00 0.252 0.278 ** rMeans not underlined by common line significantly different at 0.05 level. Seston Radioisotope Study An increasing accumulation of data dealing with sodium arsenite implicates the herbicide as an inhibitor of planktonic as well as macrophytic organisms. Sodium arsenite is capable of reducing the standing crop of phytoplankton within a few hours after treatment (Bails and Ball, 1966). It is also able to suppress phytoplankton productivity for about seven days following application (Sohacki, 1965). High concentrations of arsenic are incorporated into the plankton 68 of treated ponds. Dupree (1960) measured as high as 7,200 ppm of arsenic in the planktonic constituents of a pond which was treated with 4 ppm sodium arsenite. The zooplankton appear to be more severely affected by arsenic toxicity than the phytoplankton. Lawrence (1958) and Cowell (1965) reported drastic reductions in the numbers of rotifers, cladocera and copepods in waters treated with sodium arsenite. Gliderhus (1966) observed decreases in numbers of microcrustaceans in pools which received treat- ments of 1 ppm sodium arsenite or more. Crosby and Tucker (1966) determined the mean immobilization concentration for Daphnia magna to be 6.5 ppm. The extreme susceptibility of the zooplankton to arsenic toxicity thwarted my attempts to measure the arsenic concentration potential of these animals. In this study, submerged light traps were used to collect zooplankton samples. The traps were set out after darkness and retrieved at dawn. During the two nights preceding herbicide addition, 0.7 and 1.8 grams (wet weight) of zooplankters were captured. These samples were comprised mainly of cladocera with lesser quantities of copepods and a few hydracarina. During the night following sodium arsenite application only 0.07 grams of organisms were collected, and these were all species of hydracarina. All microcrustaceans were either killed or rendered immobile by the arsenic. The former was assumed to be true since intermittent sampling for the following two 69 months and a half showed that the quality of the pretreatment samples was never re—established. Carbon-l4 estimates of the phytoplankton productivity were made before and after the herbicide introduction. The results showed a nine day decrease in productivity followed the treatment (Figure 13). Previous studies dealing with the productivity of the test pond revealed that low incident solar energy was the only other natural factor which formerly caused abrupt reductions in productivity. Normally, produc- tivity was suppressed on days when the solar energy values were less than 125 gram calories per square centimeter. As may be observed from the data in Figure 13, the solar energy values were always around 200 gram calories per square centi- meter or considerably higher during the recovery period. Hence, I assumed that light had little, if any, affect upon the depression of phytoplankton productivity. Since the pond remained undisturbed except for the introduction of the herbi- cide, I concluded that arsenic toxicity was the main factor responsible for the productivity suppression. Sohacki (1965) observed a similar response following sodium arsenite application. An estimate of the A574 activity in the seston was achieved by subtracting the mean activity of the filtered from the non-filtered samples. The difference obtained re- flected the sorbed activity of the phytoplankton, zooplankton and tripton (Figure 14). Like all other environmental 70 Figure 13. Solar energy and phytoplankton produc- tivity values before and after sodium arsenite introduction. o 8 ‘ D: Dal/inzlday 3 8 100 S I: D/mz/day 3 71 Solar Enorgy LllJLlllllJlllJ — Productivity. 1 233232 lllllllllllllJ Days 72 .QCmnooum an pouuflm mm3 o>uso one .monEmm pououaaw mo >DH>Huom CmoE ozu mocaE pououaawucoc mo >ua>au0m Coos onu coozuon oocouommap onu ucowoumou mucaom onB .moHQEmm Houm3 mo coumom on» Ca >DH>Huom mm fin .wa ousmflm 73 gas: .00 On 0? as ON O— qa.___aa_a____a________e.fiea_a__aead__al__4____ea___ae_____a #— @— «N 9.01 x l/otl 74 components analyzed, the arsenic concentrations in the seston fluctuated throughout the course of the study. However there appeared to be a gradual reduction in A574 activity with time. The abrupt changes in A574 content of the seston may have been produced by increases in the standing crop or en- hancement of arsenic uptake by a fairly stable population of planktonic organisms. Nontheless, these data show that sig- nificant amounts of arsenic were taken up by the seston, es- pecially at the initial stages of the experiment. Stable Arsenic Study Since plankton and aufwuchs occupy a prominent po- sition at the base of the food chain, they are capable of transferring elements to all trophic levels. This is es- pecially true if the primary producers (phytoplankton and periphyton) of the forementioned communities are biological amplifiers of any specific element. Concentrated arsenic in phytoplankton theoretically could be deposited in the tissues of fishes and be consumed by humans. Consequently, sur- veillance of uptake and amplification by aufwuchs and plankton were emphasized in this study. The seston, comprised primarily of planktonic forms, manifested an affinity for arsenic even though low concen- trations prevailed in the pond waters (Table 7). Although the levels measured here did not compare to those measured by Dupree (1960), nevertheless, they were abnormally high. 75 Table 7. Means, standard errors and concentration factors of the stable arsenic in the seston and aufwuchs sampled from 1963 to 1967. Sampling Mean Standard Concentration dates mg/kg error factor (dry weight) Seston 7/63 16.3 2.3 6/64 144.5 45.0 82,019 6/65 129.0 ** 62,804 7/65 88.1 ** 13,368 9/65 42.7 *** 10,929 6/66 119.2 35.5 34,691 9/66 31.5 5.0 7,485 9/67 17.6 7.8 25,997 Aufwuchs 6/64 11.0 1.5 6,243 6/65 250.7 36.6 122,293 7/65 147.1 12.4 22,321 9/65 100.6 14.1 25,749 9/66 54.4 4.5 12,927 9/67 16.7 9.5 24,668 ** Arsen1c content of one sample. *** Mean of two samples 76 A high value of 144.5 mg/kg and a concentration factor of 82,019 was measured in samples acquired within two months after the 1964 herbicide application. Despite fluctuations in arsenic content from one sampling to another, the seston showed a tendency to decrease in arsenic content with time. By 1967 pretreatment arsenic levels were once again shown by the seston. Because the dilution effect could have been re- sponsible for a premature reduction in sestonic arsenic, I cannot attribute this to natural arsenic diminishment. [NOn- theless, these data do show that the seston concentrate arsenic in high quantities for at least two years following sodium arsenite application. Aufwuchs Radioisotope Study A literature survey produced no information dealing with the influence of sodium arsenite upon the plants and animals which are attached to but do not penetrate the sub— strates upon which they grow (aufwuchs). Some investigators have dealt with the algal components of the aufwuchs, the periphyton. Sohacki (1965) observed a short-term inhibition of periphyton following sodium arsenite treatment. Bails and Ball (1966) noted an inhibition of periphytic colonization following treatment. Uptake of A574 by the aufwuchs differed in two re- spects from the other organisms studied. First, the initial 77 uptake was slow despite the high arsenic content of the water. Secondly, there was a general tendency for the arsenic concentrations to increase with time during the course of the As74 phase of this study (Figure 15). Like the seston and g. globularis the aufwuchs activity fluctuated irregularly. However, it should be noted that the standing crop also exhibited fluctuations which were out of phase with the radioactivity. That is, the activity density of the samples decreased when the standing crOp increased and vice versa. The inverse relation between standing crop and arsenic content manifested the failure of the aufwuchs to maintain a steady state during periods of rapid growth. The similarity between the behavior of arsenic uptake by the sediments and the aufwuchs may offer a plausible ex- planation for the fluctuations exhibited by the former (Figure 5). High concentrations of bacteria are known to thrive in the upper layers of sediments from fresh water lakes (Hayes, 1964). In addition, attached algae may also frequent the upper layer provided sufficient light is avail- able to sustain their growth. However, the role of algae in the sediments during the isotope phase was assumed to be negligible. Although no analysis was conducted on the auf- wuchs population of the sediments per se, chlorophyll ex- traction determinations on the artificial substrate growth revealed a scarcity of algal flora. I concluded from this information that the algae were inhibited, possibly by 78 Figure 15. Standing crop and activity density of the aufwuchs from the test pond. Mean and standard error for each observation calcu- lated from at least three samples. Curve fitted by freehand. g/m2 (wet weight) no]: (net weight) it 10'4 Standing Crop °_ lllllllllllllllllllllllllllillJlllllllLlllllJlllJ L r— ... ~ A ",//” 1 Activity Density 0 LiluuluilliInllutltnilunlltu[1111111134 10 20 so 40 50 Days 8O arsenic toxicity, and the inhibition was not solely confined to the substrate growth but to the sediments as well. Conse- quently, the arsenic fluctuations in the sediments, if bio- logically originated, must have been caused by the activity of heterotrophs. Stable Arsenic Study By the time the aufwuchs were sampled for stable arsenic analysis, the attached algae had regained their pre- treatment prominence. Like the seston, the aufwuchs acquired extremely high arsenic levels. In fact record levels of 250.7 mg/kg of arsenic and a concentration factor of 122,292 were measured on June, 1965 (Table 7). Subsequently, arsenic levels of the aufwuchs decreased with the passage of time. Although no pretreatment estimates of the aufwuchs arsenic content were made, a natural level of 8.3 mg/kg existed in the aufwuchs of a neighboring untreated pond. I assumed this value to be comparable to the pretreatment concentra- tions in the aufwuchs of the test pond. Comparison of the latter value with the final sampling estimate (September, 1967) showed that the arsenic in the aufwuchs of the test pond was approaching natural levels again. As mentioned previously, the aufwuchs and plankton occupy a prominent position at the base of the food chain, and because of this status are capable of transferring concentrations of elements and compounds to the higher 81 trophic levels. Data presented here and in the preceding section implicate these organisms as being effective bio— logical amplifiers of arsenic. In fact, these organisms ex- hibited higher arsenic concentrations than any other environ- mental component, living or non-living. The persistence of the high arsenic levels for at least two years following the arsenic application provided ample time for the element to pass on to all trophic levels, provided the transfer did, in fact, occur. Macroinvertebrates and Fishes Radioisotope Study Because of the highly toxic nature of arsenic, ad- dition of sodium arsenite to lakes and ponds produces a po— tential hazard to humans. Direct consumption of recently treated waters may produce toxic effects, but another source must not be ignored: consumption of arsenic via contaminated aquatic organisms. Fishes, of course, are the most frequent- ly consumed freshwater organisms, however, crayfish and mussels are sometimes eaten. Biological amplification of arsenic by these animals could render them unsuitable for human consumption. Wiebe, Gross and Slaughter (1931) measured an increase in the arsenic content of fish following an in- crease in ambient concentration. On the other hand, Ullman, Schaeffer, and Sanderson (1960) failed to detect differences in the arsenic content of fishes sampled from a lake before 82 and after sodium arsenite treatment. Utilization of As74 provided a sensitive method for assaying uptake of arsenic in fish as well as invertebrates upon which the fishes depend for food. Because of the scarcity of aquatic insects in the ex- perimental pond, countable quantities were difficult to ob— tain. Pretreatment surveys revealed a small insect popu- lation. Since arsenic effectively eliminates benthic insects (Sohacki, 1965), further reduction undoubtedly transpired after sodium arsenite introduction. Of all the samples ac- quired during the study, only 37 per cent exhibited de— tectable levels of radioactivity, and these were extremely low in arsenic content. The highest counts were obtained from a sample of mayflies collected on the date of herbicide application (1,663 pc/g wet weight). Subsequent bottom samples were nearly devoid of mayflies, consequently this was the only countable mayfly sample obtained. Sohacki (1965) observed severe decreases in the mayfly populations of ponds treated with sodium arsenite. Hence, the post-treatment re- duction in mayflies must have resulted from arsenic toxicity. Dragonfly naiads ranged in activity from 200 to 900 pc/g wet weight. No detectable quantities of radioactivity were found in any of the midge larvae samples. These insects undoubted- ly acquired low arsenic levels, but the samples were too small to provide detectable counts. 83 Mussels of the genus Anodonta sp were found dead on the third and fourth day after the treatment. Tissue concen— trations ranged from 1605 to 1112 pc/g dry weight in six specimens collected. One living specimen collected 25 days after treatment contained 1511 pc/g activity. Although ocean mussels are known to accumulate arsenic (Holmes, 1934) no information is available concerning the fresh water species. Because the mussels feed primarily upon the seston which contain high levels of arsenic, it is not surprising to find high arsenic levels in their tissues. Crayfish (Cambarus sp) in the test pond consistently exhibited radioactivity in their tissues. A range of 800 to 1400 pc/g was measured in five specimens during the first week following treatment. Thereafter, the radioactivity de— creased; subsequent assays on 17 specimens revealed a range of 200-600 pc/g dry weight. ,Although these levels were fair— ly high in comparison to levels in other fauna, no crayfish mortality was evident. Accumulation of arsenic by the fishes was low in com— parison to other organisms. Approximately 75 per cent of the fishes sampled within the first three days exhibited measureable activity levels. After seven days the above figure fell to 36 per cent. Of all the fishes sampled, the golden shiner and fathead minnow concentrated the most arsenic. Their tissues contained between 900 and 2000 pc/g dry weight on the first sampling period (day 1). Since both 84 of these fishes are forage feeders, plankton constitutes a large measure of their diet. Consumption of the arsenic laden plankton probably explains the abnormally high arsenic content of these forage fishes. With the exception of the forage fishes, no con- trasting differences in the arsenic uptake could be attributed to feeding habits. The omnivorous green sunfish, pumpkinseed 74 than the and bullhead did not acquire any more or less As more carnivorous perch. Furthermore, separate analyses of the viscera, fillets and external mucous showed no prefer- ential areas of arsenic concentration. As the experiment progressed the numbers of fishes with detectable radio- activity dwindled. .At the last sampling (53 days) only six fishes out of 21 contained radioactivity, and these were ex- tremely low levels. Stable Arsenic Study Stable arsenic concentrations in the fauna persisted at low levels from 1964 to 1966, when sampling for this phase of the study was discontinued. Arsenic concentrations in dragonfly naiads ranged from 0.3 to 5.1 mg/kg dry weight. Midge larvae concentrated 2.8 to 13.5 mg/kg arsenic in their tissues. Two mussels collected in 1965 contained 0.6 and 0.7 mg/kg arsenic. Low concentrations continued to persist in the fishes. Arsenic analyses of the fillets revealed an arsenic range of 85 trace to 0.65 mg/kg, whereas the viscera samples ranged from trace to 15.0 mg/kg. All viscera samples except one con- tained measurable quantities of arsenic. Ullman, et a1. (1960) also detected more arsenic in the viscera than in the fillets. Results from both the radioisotope and stable arsenic studies show that very low quantities of arsenic were re- tained by the aquatic fauna. These low quantities prevailed despite the extremely high arsenic values in the organisms which comprise the base of the food chain. Ingestion of arsenic tainted food was unavoidable; this is attested by the higher arsenic concentrations in the viscera of the fishes. The absence of high arsenic levels in the fillets suggests that the fishes are capable of eliminating arsenic and thereby prevent a build-up in their tissues. Wiebe, et a1. (1931) noted that fish acquire just so much arsenic and no more, despite increased ambient concentrations. They also concluded that fish were able to rapidly eliminate arsenic. From the standpoint of public health, it is worth- while to review the literature dealing with the restrictions on arsenic in water and in foods. The U.S. Public Health Service restricts arsenic concentrations to 0.05 ppm in drinking and culinary water carriers subject to Federal quarantine regulations (Rainwater and Thatcher, 1960). The world tolerance for arsenic was established as 0.01 grains 86 of A8203 per pound or 1.4 ppm (White, 1933). Natural arsenic concentrations in sea foods greatly exceeds the specified limits. Since the marine environment is rich in arsenic (0.006 to 0.030 ppm), it is not surprising to find high arsenic levels in the organisms derived thereof. Common species of edible fishes (cod, haddock, mackeral, sardines) contain from 1.1 to 4.1 ppm arsenic (Holmes, 1934). Crusta- ceans are noted for their ability to concentrate arsenic; prawns may concentrate as high as 174 mg/kg. Dried kelp sold for human consumption may contain 66 ppm arsenic (Holmes, 1934). Despite the fact that a great percentage of the world population subsists primarily on seafood no symptoms of chronic arsenic toxicity is evident. The low toxicity of organically bound arsenic coupled with the ability of mammals to eliminate the element may ex- plain the apparent absence of toxic effects following the consumption of arsenic tainted sea food. .Rats fed on a diet of arsenic contaminated shrimp rapidly eliminated the arsenic (Coulson, Remington and Lynch, 1935). Overby and Frost (1962) demonstrated that rats retained higher concentrations of inorganic arsenic and for a longer duration than an equiva- lent amount of organically bound arsenic. Humans fed As74 contaminated chicken also eliminated the arsenic rapidly (Calesnick, Wase, and Overby, 1966). From the above information, organically bound arsenic was judged to be relatively innocuous to humans. The arsenic 87 concentrations which prevailed in the pond fishes were much lower than the natural concentrations in marine organisms. Although arsenic is frequently associated with carcinogenic agents, Vallee, et al. (1960) state that there is no experi- mental basis to support any of these conjectures. Conse- quently, the supposition that chronic arsenic poisoning and/ or arsenic induced cancer may result from consumption of organisms derived from sodium arsenite treated waters present— ly appears to be untenable. SUMMARY The dynamics of arsenic in a farm pond was followed for four years via radioarsenic and stable arsenic detection. A continuous exchange between the sediments and water maintained arsenic concentrations above natural levels for at least two years following the last arsenic treatment. Turnover times for arsenic in the water and solids phases of the pond were determined to be 9.8 and 72.5 days respectively. Arsenic sorption and retention by silt was significantly greater than sand, however, both sediments exhibited irregular fluctuations in arsenic content during the radioisotope phase of the study. Arsenic concentrations in the upper layers of the sedi— ments decreased with time; the arsenic appeared to be leached into the deeper sediment layers thereby losing its active role in the environment. High levels of arsenic uptake were measured in the algae, Spirogyra sp and Chara globularis, the former succumbed rapidly after treatment, however, the latter remained unaffected. 88 10. ll. 89 Arsenic uptake in the higher plants varied from one species to another, and the mechanism of sorption ap- peared to be associated with the stems and leaves rather than roots. Zooplankton populations were severely reduced following sodium arsenite application. Phytoplankton productivity was inhibited for approxi- mately nine days as a result of the sodium arsenite treatment. Arsenic levels in the seston and aufwuchs fluctuated during the radioisotope phase of the study. These organisms concentrated the highest levels of arsenic of all the environmental components analyzed. Despite the high concentrations of arsenic displayed by the organisms at the base of the food chain, low arsenic prevailed at the high trophic levels. Macro- invertebrates and fishes showed no tendency to amplify arsenic. LITERATURE CITED Allgeier, W. H., W. H. Peterson and C. Juday. 1934. Avail- ability of carbon in certain aquatic materials under aerobic conditions of fermentation. Int. Rev. d. ges. Hydrobiol. u Hydrogr. 30:371-378. Analytical Methods Committee. 1960. Methods for the de- struction of organic matter. Analyst 85:643—656. APHA, AWWA, WPCF. 1965. Standard Methods for the Exami- nation of Water and Wastewater. American Public Health Assn. New York. 769 pp. Arnott, J. T. and A. L. Leaf. 1967. The determination and distribution of toxic levels of arsenic in silt loam soil. Weeds 15:121-124. Bahr, T. G. .1966. Aquaria studies on the uptake of arsenic 74 labeled sodium arsenite by components of an aquatic ecosystem. Unpub. M.S. thesis, Mich. State, Univ., 94 pp. Bails, J. D. and R. C. Ball. 1966. Response of pond metabo- lism to sodium arsenite. Mich. Acad. Sci. Arts and Let. Vol. LI: 193-208. Ball, R. C. and F. F. Hooper. 1963. Translocation of phosphorus in a trout stream ecosystem. Radioecology pp. 217-228. Ballinger, D. G., R. J. Lishka and M. E. Gales. 1962. Ap- plication of silver diethydithiocarbamate method to de- termination of arsenic. Jour. Amer. Water Wks. Assn. 54:1424 Bear, F. E. 1957. Toxic elements in the soil. Soil the 1957 Yearbook of Agriculture. U.S. Govt. Printing Office. pp. 165-171. Beard, H. C. 1960. The Radiochemistry of Arsenic. National Academy of Sciences. Nuclear Science Series 3002. 27pp. Bishop, R. F. and D. Chisholm. 1962. Arsenic accumulation in Annapolis Valley orchard soils. Canad. Jour. Soil Sci. 42:77-80. 90 91 Calesnick, B., A. Wase and L. R. Overby. 1966. ,Availability during human consumption of the arsenic in tissues of chicks fed arsanilic-74 acid. Toxicol. and Appl. Pharm. 9:27—30. Carson, R. L. 1962. Silent Spring. Houghton Mifflin Co. 304 pp. Comar, C. L. 1955. Radioisotopes in Biology and Agri— culture. McGraw—Hill, New York. 481 pp. Coulson, E. J., R. E. Remington and K. M. Lynch. 1935. Metabolism in the rat of naturally occurring arsenic of shrimp as compared with arsenic trioxide. Jour. Nutrition 10:255-270. Cowell, B. C. 1965. The effects of sodium arsenite and silvex on the plankton populations in farm ponds. Trans. Amer. Fish. Soc. 94:371-377. Crosby, D. G. and R. K. Tucker. 1966. Toxicity of aquatic herbicides to Daphnia magna. Science 154:289-290. Davison, V. E., J. M. Lawrence, and L. V. Compton. 1962. Waterweed control on farms and ranches. Farmers' Bulletin 2181. U. S. D. A. Doepke, P. A. 1963. Arsenic distribution in a shallow lake treated with an herbicide of sodium arsenite. Unpub. M.S. thesis, Univ. Wisc. 37 pp. Domogalla, B. P. 1926. Treatment of algae and weeds in lakes at Madison, Wisconsin. Eng. News-Record 97:850. Dupree, H. K. 1960. The arsenic content of water, plankton, soil and fish from ponds treated with sodium arsenite for weed control. Proc. 14th Ann. Conf. SE Assn. Game and Fish Comm. 132—137. Fish, G. R. 1963. Observations on excessive weed growth in two lakes in New Zeeland. New Zeeland Jour. of Botany 1:410-418. Fisher Scientific Company. 1960. Technical Data Bulletin TD—l42. Gilderhus, P. A. 1966. Some effects of sublethal concen- trations of sodium arsenite on bluegills, and the aquatic environment. Trans. Amer. Fish. Soc. 95:289- 296. 92 Harris, E. 1957. Radiophosphorus metabolism in zooplankton and microorganisms. Can. Jour. Zool. 35:769-782. Hayes, F. R., J. A. McCarter, M. L. Cameron and D. A. Livingstone. 1952. On the kinetics of phosphorus ex- change in lakes. Jour. Ecol. 40:202-216. Hayes, F. R. and J. E. Phillips. 1958. (Lake and water sedi- ment IV. Radiophosphorus equilibrium with mud, plants, and bacteria under oxidized and reduced conditions. Limnol. and Oceanog. 3:459-475. 1964. The mud-water interface. Oceanogr. Mar. Biol. Ann. Rev. 2:121-145. Hem, J. D. 1959. Study and interpretation of the chemical characteristics of natural waters. Geological Survey water-Supply Paper 1473. U.S. Govt. Print. Off. 123 pp. Holmes, A. D. 1934. ‘Arsenic content of American cod liver oil. Ind. and Eng. Chem. 26:573-574. Hutchinson, G. E. 1957. A Treatise on Limnology. Volume I Geography, Physics, and Chemistry. John Wiley and Sons. Inc. 1015 pp. King, B. G. and W. E. Caldwell. 1959. The Fundamentals of College Chemistry. Amer. Book Co. New York. 592 pp. Lanz, H. Jr., P. C. Wallace, and J. G. Hamilton. 1950. The metabolism of arsenic in laboratory animals using As as a tracer. Univ. of Calif. Publ. in Pharmac. 2:263- 282. Lawrence, J. M. 1958. ,Recent investigations on the use of sodium arsenite as an algacide and its effects on fish production in ponds. Proc. 11th Ann. Conf. SE Assn. Game and Fish Comm. pp. 281-287. .Lee, G. F. and A. W. Hoadley. 1967. Biological activity in relation to the chemical equilibrium composition of natural waters. Equilibrium concepts in Natural Water Systems. (Amer. Chem. Soc. Pub. pp. 319-337. Leuschow, L. A. 1964. The effects of arsenic trioxide used in aquatic weed control operations on selected aspects of the bio-environment. Unpub. M.S. thesis, Univ. Wisc. 66 pp. Littlefield, L. and C. FOrsberg. 1965. Absorption and translocation of phosphorus-32 in Chara globularis Thuill. Physiol. Plant. 18:291-296. 93 Mackenthun, K. M. 1955. The control of submergent aquatic vegetation through the use of sodium arsenite. NE Weed Cont. Conf. 1955 pp. 545-557. 1958. The chemical control of aquatic nuisances. Committee on Water Pollution, Madison, Wisc. 64 pp. , W. M. Ingram, and R. Porges. 1964. Limnological aspects of recreational lakes. USDHEW, Public Health Serv. Pub. 1167. U.S. Govt. Print. Off. 176 pp. 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. Jour. Soil Sci. 40: 59-62. Macpherson, L. B., N. R. Sinclair and F. R. Hayes. 1958. Lake water and sediment III. .The effect of pH on the partition of inorganic phosphate between water and oxidized mud or its ash. Limnol. and Oceanog. 3:318- 326. Overby, L. R. and D. V. Frost. 1962. NOnavailability to the rat of the arsenic in tissues of swine fed arsanilic acid. Toxic. and Appl. Pharm. 4:38-43. Pendleton, R. C. 1962. _Accumulation of cesium-137 through the aquatic food web. Biological Problems in Water Pollution. Third Seminar. USDHEW Public Health Serv. Pub. 999—WP—25. pp. 355-366. Rainwater, F. H. and L. L. Thatcher. 1960. Methods for Collection and Analysis of Water Samples. Geol. Surv. Water-Supply Paper 1454. U.S. Govt. Print. Off. 301 pp. Rhode, W. 1958. The primary production in lakes: some re- sults and restrictions of the Cl method. Cons. Internat. Explor. de la mer. Vol. 144:122—128. Rigg, H. H. 1955. The effect of herbicides on a series of aquatic plants. Proc. 9th Ann. Meeting NE Weed Conf. pp. 535-544. Rigler, F. H. 1956. A tracer study of the phosphorus cycle in lake water. Ecol. 37:550—562. Sayre, W. W., H. P. Guy, and A. R. Chamberlain. 1963. Up- take and transport of radionuclides by stream sediments. Geol. Surv. Professional Paper 433—A. U.S. Govt. Print. Off. 33 pp. 94 Sohacki, L. P. 1965. Ecological alterations produced by the treatment of pond ecosystems with copper sulfate and sodium arsenite. Unpub. M.S. thesis, Mich. State Univ. 91 pp. Steeman Nielsen, E. 1952. The use of radio—active carbon (C14) for measuring the organic production in the sea. Jour. du Cons., 18:117-140. Surber, E. W. 1929. The utilization of sloughs in the Upper Mississippi Wildlife and Fish Refuge as fish ponds. Trans. Amer. Fish. Soc. 59:106-113. 1931. Sodium arsenite for controlling submerged vegetation in fish ponds. Trans. Amer. Fish. Soc. 61: 143-149. 1943. Weed control in hard-water ponds with copper sulphate and sodium arsenite. Trans. 8th N. Amer. Wildl. Conf. pp. 132-141. Timmons, F. L. 1962. Present and Future of aquatic weed control. Proc. NE Weed Contr. Conf. 16:1-8. Ullman, W. W., R. W. Schaefer and W. W. Sanderson. 1961. Arsenic accumulation by fish in lakes treated with sodium arsenite. Jour. Wat. Poll. Contr. Fed. 33:416— 418. USDHEW. 1960. Radiological Health Handbook. Public Health Service 468 pp. Vallee, B. L., D. D. Ulmer and W. E. Wacker. 1960. Arsenic toxicology and biochemistry. AMA Arch. of Ind. Health. 21:132-151. Vasak, V. and V. Sedivek. 1952. Colorimetric determination of arsenic. Chem. Listy., 46:341. Wiebe, A. H., E. G. Gross and D. H. Slaughter. 1931. The arsenic content of large-mouth black bass (Micropterus salmoides lacepede) fingerlings. Trans. Amer. Fish. Soc. 61:150-160. Whittaker, R. H. 1961. Experiments with radiophosphorus tracer in aquarium microcosms. Ecol. Monog. 31:157-188. White, W. B. 1933. Poisonous spray residues on vegetables. Indust. and Eng. Chem. 25:621-623. APPENDIX 96 Calculation of Turnover Time The following is a detailed account of the mathe- matical steps necessary for the calculation of turnover times. This explanation was kindly provided by Dr. Hayes and was slightly modified for the As-74 study. When excess arsenic is added to the water of a pond, the decrease can be described by the following mechanism. Suppose the water loses arsenic to the solid phase at the rate of l per cent per day during any number of days t. As- sume the arsenic to be returned to the water phase at the rate of‘p per cent per day of the amount No—N present after time t in the solid phase. Nb is the concentration in the water at time to. Then fi—E = -)\N +JJ(No-N) (1) For very small intervals of time, dt, the expression may be . dN _ _ written ET: — N (1+p) +pNo (2) The differential equation (2) may be solved to give N “773‘.— <3) Equation (3) may be rewritten in the form N__’_‘_§2_=_3_b19_0-(7\+»)t (4) AUJ 3+» 97 N N For convenience let -2L—-9-— = A andL—O—— = B (5) 1+ )1 1+ in Under these conditions No = A + B and as t ———)00, N——-9 B so that B represents the asymptote of the curve in Figure x. Equation (4) can be expressed as N - B = 1160“?”t or loge(N - B) logOA-(A +).1)t log (N - B) log A - 0.4343 (A+Ju)t (6) if log (N-B) is plotted against t, (6) is a straight line for which log A is the intercept and -0.4343 ().+‘p) is the slope k. If the experimental data can be fitted to a straight line, the values of A and B plus the slope are sufficient to determine values for A andju. The equilibrium value (B) was determined by averaging the last three points on Figure 4. _A semi-log plot of N-B produced a straight line (Figure 16). The least squares method was used to fit the curve; the slope and intercept were calculated to produce the following equation: log N-B = 4.8662 + (0.4343)(0.0482)t. Using this equation values were calculated to correct the original observations (Table 8). We now have N6 = 56,048 B = 6,698 A = 49,350 98 k From the d‘1scuss1on we know that m = A-I- )1 = 0.11535 N and A = 2+; Solving forA we get A = Jag—£15 = 0.10156. 0 Since A-t-p = 0.11535 and/t = 0.10156, then/.1 = 0.01379, where Ais the rate with which arsenic is lost to the solid phase andp is the rate with which it is returned to the water. Turnover time Tt for water is equal to —%L— for solids Tt = —-1— = 50.36 days. Half—life is related to )u turnover time as follows Tt = TEE-3 loge = 1.4465. 9.85 days and 99 Figure 16. Plot of N - B used in determining turn- over times of arsenic in the water and solids phases of the experimental pond. 1.00 5!) 20 10 Days 101 Table 8. Raw water activity and the conversions required for estimating the turnover times in water and solids. Mean Day activity N-B Calculated Converted density N-B activity pc/liter (B = 6,698) from curve density 0 58,089 51,391 49,350 56,048 1 63,257 56,559 43,970 50,668 3 50,892 44,194 34,920 41,618 5 37,522 30,824 27,720 34,418 7 21,994 15,296 22,040 28,738 10 19,647 12,949 15,570 22,268 13 18,894 12,196 11,010 17,708 16 14,362 7,664 7,793 14,491 19 10,203 3,505 5,514 12,212 23 10,719 4,021 3,476 10,174 27 8,808 2,110 2,191 8,889 31 3,743 —-- 1,381 8,079 35 7,416 718 870 7,568 39 7,420 722 548 7,246 45 6,999 301 274 6,972 52 4,956 —-- 122 6,820 "Iillfllfilfljfllflllllmfllflfllllflr