- ‘o————_—.— _—..-__. _ __ RESPONSE OF A POND METABOLISM ' no SODIUM ARSENITE Thesis for the Degree of M. S. MICHEGAR STATE UNIVERSETY JACK DEVERICK BAELS 1988 THES‘s . m“ rav‘vn‘szd‘y. o any .- , Q 3"! ’2? 1“. . 5 L1» 1AA i ' Michigan 1-Star; 011ch5. 1y . N e “2.3“ ‘1. J 5‘ Of ABSTRACT RESPONSE OF A POND METABOLISM TO SODIUM ARSENITE by Jack Deverick Bails The metabolic responses of a small pond to a sodium arsenite treatment were investigated. Two ponds were used in the study and one received an application of 8 mg/l sodium arsenite in two successive years. The other pond was utilized as a control. Dissolved oxygen, alkalinity, conductivity and pH were measured in both ponds. The dissolved oxygen was recorded automatically on a twenty-four hour basis and gross oxygen production values were calculated using a computer program. Periphyton and phytoplankton production was also measured during the second year of this study. The dissolved oxygen, pH, and alkalinity in the treated pond all reacted immediately to the sodium arse- nite application. The pH dropped from 8.65 to 7.1 within one week after the herbicide was added in the first year (1963). The total alkalinity increased over 30 mg/l after the herbicide was added in 1963. Dissolved oxygen was Jack Deverick Bails reduced from 8.5 mg/l before treatment to a summer low of 4.8 mg/l within 72 hours after the arsenite was added in 1963. Oxygen production, respiration, efficiency and pro- duction—to-respiration ratios were all depressed following the 1963 treatment. Similar results were obtained in the 1964 applica- tion and the control pond served to confirm the observation of the effects of the sodium arsenite. The degree of changes in 1964 were less than those observed in 1963 indicating that the metabolic reaction was related to the quantity of plant biomass killed. The periphyton and phytoplankton measurements in 1964 indicated the sodium arsenite temporarily interfered with their growth. The immediate effects of the herbicide treatment were short lived (i.e. two to three weeks). However, the removal of the macrophytes did effect production-to- respiration ratios, and the primary efficiencies through- out the study period. RESPONSE OF A POND METABOLISM TO SODIUM ARSENITE BY Jack Deverick Bails A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1968 .21.; U“ H I -‘ _-./ r'k: 1,5 (3 ACKNOWLEDGMENTS I extend my sincere thanks and appreciation to Dr. Robert C. Ball, for his thoughtful guidance and super- vision of my graduate program. A special debt is owed many of my fellow graduate students whose helpful advice and consul made my research much easier. To my wife, Janis, I am grateful for her confident urging and understanding patience. I am also grateful to the Michigan Department of Conservation, Research Division, who made the study for this thesis possible under a graduate fellowship (F-27-R3, WP No. 11, Job No. 8). Thanks is extended to the Agri- culture Experiment Station of Michigan State University within whose facilities this research was conducted. ii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . LIST OF TABLES . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . DESCRIPTION OF STUDY AREA . . . . . . METHODOLOGY . . . . . . . . . . . . . Sodium Arsenite Application . . . . Physical Measurements . . . . . . . Temperature . . . . . . . . . . . Solar Energy . . . . . . . . . . . Air Pressure . . . . . . . . . . . Chemical Measurements . . . . . . . Alkalinity . . . . . . . . . . . Hydrogen Ion Concentration . . . . Conductivity . . . . . . . . . . . Dissolved Oxygen . . . . . . . . . Automatic Monitoring of Chemical Measurements Biological Measurements . . . . . . Periphyton . . . . . . . . . . . . Phytoplankton . . . . . . . . . . Oxygen Metabolism . . . . . . . . Accrual Oxygen . . . . . . . . . . Diffusion . . . . . . . . . . . . Respiration . . . . . . . . . . . Computer Analysis . . . . . . . . P/R Ratios . . . . . . . . . . . . Efficiencies . . . . . . . Conversion of Oxygen Values to Energy Units iii Page ii vi 10 10 ll 11 12 12 13 l3 13 13 13 14 18 18 22 24 24 25 26 27 30 31 34 RESULTS AND DISCUSSION . 1963 Preliminary Study Hydrogen Ion Concentration . Alkalinity (Methyl Orange - Phenolphtalein). Factors in Alkalinity and pH Responses . Dissolved Oxygen . Oxygen Metabolism Temperature Conductivity 1964 Study . Periphyton . Phytoplankton Dissolved Oxygen . Oxygen Metabolism Hydrogen Ion and Alkalinity Comparison Between 1963 and 1964 Comparison of Results with Other Solar Radiation Oxygen Metabolism Efficiencies Periphyton . SUMMARY . . . . LITERATURE CITED . APPENDIX . . . . iv Treatments Studies Page 37 37 37 41 42 49 50 SO 51 61 65 68 71 72 74 75 75 79 86 90 93 99 LIST OF TABLES Table Page I. Analysis of Covariance of Periphyton Production in Pond C (Treated) Versus Pond D (control) 0 O O O I O O O O O O O O 56 II. Analysis of Covariance of Periphyton Production in Pond C (Treated) Versus Pond D (Control) . . . . . . . . . . . . . 57 III. Range of Efficiency and Gross Oxygen Pro- duction Values from Various Aquatic Communities Compared to Lake City Ponds . 81 Figure 10. 11. LIST OF FIGURES Page Map of Lake City ponds showing location of sample devices . . . . . . . . . . . . 9 Diagramatic representation of automatic monitoring equipment and photograph of laboratory apparatus . . . . . . . . . . 16 Correction for respiration in the deter- mination of gross oxygen production . . . 29 Energy flow diagram adapted from E. P. Odum (op. cit.) where: . . . . . . . . . 33 Daily maximum and minimum hydrogen Ion concentrations in Pond C after 1963 treatment . . . . . . . . . . . . . . . . 39 Maximum-minimum dissolved oxygen (D.O.) and production/respiration (P/R) ratios in Pond C following 1963 treatment . . . 44 Available solar energy (La), efficiency (Pg/La), respiration (Rt) and production (Pg) in Pond C following 1963 treatment . 47 Periphyton production on vertical sub- strate in Pond C and Pond D in 1964 . . . 53 Periphyton production on horizontal substrate in Pond C and Pond D in 1964 . 55 Index to standing crop of phytoplankton in 1964 from 16 liter centrifuged samples . . . . . . . . . . . . . . . . . 63 Maximum-minimum dissolved oxygen and production/respiration (P/R) ratios in Pond C during 1964 study . . . . . . . 67 vi Figure 12. l3. 14. 15. Available solar energy (La), efficiency (Pg/La), respiration (Rt) and production (Pg) in Pond C (treated) and Pond D (control) during 1964 study . Range of oxygen metabolism values from several ponds: The inverse, curvilinear relationship between solar energy and efficiency in the Lake City ponds . The linear relationship between optical density of chlorophyll content and dry weight of periphyton in Pond D and Pond C in 1964 vii Page 70 77 85 88 INTRODUCTION The over-abundance of aquatic plants often limits the full use of a water resource. Extensive research has been carried out to develop controls for aquatic weeds and as a result, many chemical herbicides have been developed. Most of the studies involving herbicides have dealt with their effectiveness in removing aquatic weeds; very little work has been done to determine what effects these chemi- cals have on the metabolism of an aquatic community. Many methods have been presented which attempt to measure a part, or the total metabolism of an aquatic com- munity. With the refinements presented by Odum (1956), the diurnal oxygen curve method is one of the most effi- cient means of obtaining a total picture of the metabolic rates of an aquatic community. The purpose of this study was to utilize diurnal oxygen curves and other measures of primary production to determine the effects of sodium arsenite on the metabolism of a small pond. Sodium arsenite is the herbicide most often used to kill submergent aquatic weeds. It has remained popular for over fifty years, primarily because of its effective- ness in removing weeds and its relatively low cost (Mackenthum, K. M., 1964). It has been well established that sodium arsenite, when applied to an aquatic environment, reduces the dis- solved oxygen. However, little research has been done to: (1) quantify this loss of dissolved oxygen in terms of primary production; (2) determine the duration of this oxygen depletion; or (3) evaluate possible changes in the community production to respiration ratios. Without this information, the long term effects of a herbicide treat- ment on the biota of an aquatic ecosystem are difficult to evaluate. Sohacki (1965) made estimates of the duration of the dissolved oxygen depletion following the application of sodium arsenite on a small pond. Although the dis- solved oxygen determinations were only made once a day during his study, Sohacki reported that the treatment pond exhibited below normal oxygen readings for thirty days after-the herbicide was added. In this same study, he also attempted to make estimates of the loss of primary production due to the sodium arsenite by determining the post and pre-treatment production of macrophytes, plankton and periphyton. His results indicated that all types of primary producers were at least temporarily inhibited by the application of the herbicide. Copeland and Whitworth (1963) attempted to measure the effects of a new chemical herbicide on the oxygen metabolism of a small Oklahoma farm pond. Their analysis of the diurnal oxygen curves, taken before and after the pond was treated, indicated that the decaying vegetation, resulting from the treatment, increased the community respiration and significantly reduced the community pro- duction-to-respiration ratio. With the exception of this single study, the free oxygen (diurnal oxygen curve) method has not been utilized to measure the effects of a herbicide treatment. However, since 1956 diurnal oxygen curves have been used to measure the metabolic rates of lenthic aquatic communities under a variety of other conditions. Minter and Copeland (1962) studied the wintertime oxygen production and respiration relationships of a small lake and found that diffusion of oxygen from the atmosphere played a major role in maintaining the oxygen balance when photosynthetic oxygen production was limited. COpeland and Dorris (1964) investigated the photo- synthetic oxygen productivity in ponds polluted with oil refinery effluent. They found exceptionally high oxygen production values and correspondingly high community oxygen respiration values. They reported that the oxygen produc- tion-to-respiration ratio was often less than unity in the oil polluted ponds. COpeland (1963) reported similar results on several other oil effluent holding ponds. The oxygen metabolisms of four distinctly different unpolluted farm ponds were studied by Copeland and Whitworth (op. cit.). The results of their study indicated that the sources of water for these ponds and the terrain that it passed over, determined to a great extent the rate of oxygen metabolism. Each of these studies demonstrated that diurnal oxygen curves can be utilized effectively to measure the metabolic rates of aquatic communities. They also indi- cated that the effects of a sodium arsenite application could be measured quantitatively by this same method. There is one major disadvantage in determining primary production values with the diurnal oxygen curve method. Periodic dissolved oxygen readings must be taken throughout a twenty-four hour period to determine oxygen production. Obviously, to maintain daily oxygen production records through chemical analysis would be an arduous task, even for a short period of time. The oxygen metabolism studies previously mentioned relied upon two or three diurnal curves for results. The nature of this study made it imperative that a complete daily oxygen metabolism record be kept. Preliminary studies by the author indicated that it was feasible to obtain continuous dissolved oxygen readings through the utilization of automatic sensing and recording equipment. The unique manner in which the data for this study was obtained and compiled is perhaps as significant as the results themselves and consequently these methods will be discussed in detail. DESCRIPTION OF STUDY AREA The research for this study was conducted at the Michigan State University Agricultural Experiment Station, located two miles south of Lake City, Michigan. The ponds used in this study are maintained by the Department of Fisheries and Wildlife for limnological and fisheries research during the summer.. There are six ponds located on the farm, four of which are connected to a five acre impoundment. The im- poundment was built in the early 1940's on a small stream, Mosquito Creek. It serves both as a reservoir for main- taining the experimental ponds and as a source of irriga- tion water. The two ponds used in this study (Pond C and Pond D) have an inlet from the reservoir and an outlet to the Mosquito Creek (Fig. 1). Pond C and Pond D were chosen for this study primarily because of their close proximity to the field laboratory buildings. In previous studies (Sohacki, op. cit.), Pond C had been treated with both sodium arsenite and copper sulfate. Pond D, on the other hand, had a history of having very few rooted aquatic plants and had not been used in previous algalcide or herbicide experiments. The substrate of all the ponds was originally sand. However, the process of eutrophication has depos- ited a layer of mucky organic debris over most of the bottom of the two ponds used in this study. Physically, Pond D and Pond C are quite similar. The maximum depth of both ponds during the study, was 4.5 feet. Their areas and average depths were as follows: Pond C - .17 acres, 3.4 ft. average depth Pond D - .19 acres, 3.2 ft. average depth The research for this study was conducted through- out the summer months of 1963 and 1964. moofi>mo mHmEmm mo QOMDMOOH mcflzonm mocom huwu mxmq mo mm: .H mHsmHm 0‘9 o”o. o... 0...... - I U . . 00.. . at; I. a“..‘|: .§ 0... J “ O‘ayoo» 0“ ...'I.. o 000‘ 05:“ “°_‘ as 0‘”- .. *§. 0 V ‘ a Pump I L 0 h 0 E .2 G) S I. >~ fl. Fig. I METHODOLOGY. Sodium Arsenite Application In the preliminary study in 1963, and in the final study in 1964, a liquid form of sodium arsenite (NaAsOZ) was added to Pond C. Enough sodium arsenite was added to the pond in each treatment to bring the total concentration of arsenic trioxide (A3203) up to 8 mg/l. The liquid form of sodium arsenite used in this study ("Atlas - A", Chipman Chemical Co., Chicago, Illinois) contained four pounds of arsenic trioxide per gallon. In 1963, the concentrated sodium arsenite was diluted with water prior to the application in a SO-gallon drum, lined with polyethylene. This drum was placed on the shore and the diluted arsenic solution was pumped through one-half inch tubing to an applicator on a boat. It took approximately twenty minutes for the llO-volt submersible pump to drain the SO-gallon drum. The sodium arsenite was diluted in a sixteen gallon plastic barrel prior to the 1964 treatment. The dilution was approximately one part sodium arsenite to ten parts water. The barrel was placed on a boat and the diluted arsenic solution was pumped through the applicator. 10 11 A wet-cell storage battery located on the boat, provided the power to operate the 12-volt submersible pump used for this treatment. For both treatments, a boom-type c0pper applicator was used to distribute the sodium arsenite evenly over the entire surface of the pond. The boat, with applicator attached, was pulled over the pond's surface with ropes in 1963. And in 1964, the boat was simply rowed while the arsenite was being applied. No additional effort was made to mix the herbicide into the pond water, after either application. Physical Measurements The physical measurements of solar energy, atmos- pheric pressure and temperature were obtained in the same manner in 1963 and 1964. Temperature A continuous record of air temperature was main- tained using a Taylor air thermograph. Water temperatures in Pond C were continuously recorded on a Taylor water thermograph. Numerous determinations of the water tempera- tures in Pond D indicated that there was no significant difference between the water temperatures of Pond D and Pond C. Thus, one measurement of water temperature was assumed to be indicative of both ponds. 12 The thermographs, used to obtain air and water temperatures, were located near the spillway of Pond C (Fig. 1). Solar Energy Solar energy was measured with an Epply pyrhelio- meter which was mounted on a permanent concrete cylinder between ponds C and D (Fig. l). The pyrheliometer is a thermopile type radiation detector which generates an elec- tromotive force (emf), through a series of thermocouples, proportional to the incident radiation. A Bristol strip-chart integrated recorder was used to record, and convert the signal from the pyrheliometer to gram-cal/cmz/day. The recorder assembly was located in the laboratory and the signal from the pyrheliometer was carried by an underground cable to the recorder. The recorder was automatically timed to go on at sunrise and turn off at sunset. The timer was preset manually and adjusted peri- odically to approximate, as closely as possible, the actual hours of sunrise and sunset. Air Pressure Barometric pressure was measured with a continuously recording Taylor barograph. Air pressure readings were maintained primarily to record sudden storms which might 13 account for unusually low pyrheliometer and/or dissolved oxygen readings. Chemical Measurements Alkalinity Total alkalinity was measured by titrating samples with 0.02 sulfuric acid, using methyl-orange and phenol- phthalein as indicators (American Public Health Association et a1., 1960). Hydrogen Ion Concentration Hydrogen ion concentration was determined using a Beckman expanded scale pH meter (Model 76) and continuously recorded on a Bausch and Lomb recorder. Conductivity Conductivity was obtained with an Industrial Instruments continuously recording conductivity meter (Model RQ). Dissolved Oxygen Dissolved oxygen was determined with a Beckman polarographic oxygen analyzer (Model 777) and a Beckman polarographic oxygen analyzer adapter in conjunction with a Model 76 pH meter. Dissolved oxygen values were continu- ously recorded on Sargent recorders (Models MR and SR). 14 The polarographic oxygen analyzers were calibrated daily using the Alsterberg (Azide) modification of the Winkler method (American Public Health Association, et al., op. cit.). Automatic Monitoring of Chemical Measurements During the two year study period, changes in dis- solved oxygen, pH, and conductivity were simultaneously analyzed and recorded inside the laboratory on a closed extension of the pond water. A llO-volt submersible pump was located in each pond. The water was pumped from the pond through 1/2 inch Tygon tubing to a plexiglass cylinder (18" long, and 6" in diameter), and eventually returned to the pond. The sensing probes for the numerous analyzers were inserted in the t0p of the cylinder (Figure 2). The system was entirely closed and continuously running such that the chemical measurements made of the water passing through the cylinder were at any given moment, representa- tive of the water in the pond. A diagramatic representa- tion of the system is illustrated in Figure 2. Although instruments which measure conductivity and pH have been in general use for some time, dependable dissolved oxygen analyzers have only recently been developed and utilized in biological research. Odum and Hoskin (1957) attempted to adapt the gold-zinc electrode developed by Ohle (1953) to make 15 msumHMQQm muoumuonma mo gmmnmouonm cam newsmasqm OCHHODACOE oaumfiousm m0 coaumucmmmumwu oflumfimummfla .N musmflm 16 Luemc< xtznuavcou «a: i; 0>_U>. 501.090: Lou»_uc< NO LovLOuou 17 polarographic oxygen measurements in a laboratory stream microcosm. However, their efforts were unsuccessful and their results indicated that the presence of zinc in the system inhibited the growth of algae. Odum and Hoskin (op. cit.), reported that Ambuhl (1955), and Lynn and Okum (1955) also had failed to obtain reliable dissolved oxygen readings in a flowing system with either gold-zinc or solid platinum polarographic analyzers. In England, Gameson and Griffin (1959) were able to obtain a six-month record of dissolved oxygen in a polluted stream using photographic equipment to record the dial readings on a dropping mercury electrode. Elec- trical recorders have also been used with mercury elec- trodes to record dissolved oxygen (Briggs, Dyke and Knowles, 1959). Newly developed polarographic dissolved oxygen analyzers were evaluated by Macklin, Baumgartner, and Ettinger (1959), and used successfully by Bartsch (1959) to measure diurnal oxygen in Ohio streams. Sneed and Dupree (1962) developed an oxygen- temperature meter designed particularly for fishery biologists. Their polarographic oxygen analyzer consisted of a platinum, silver oxide electrode pair covered with a plastic membrane. Sneed and Dupree credit Clark, Wolf, Granger and Taylor (1953) with the development of plastic 18 films to cover the electrodes which prevents "poisoning" (i.e., plating of electrodes with other metalic ions in the water). This development allowed continuous recording of oxygen with polarographic analyzers over long periods of time. Very recently, Copeland and Duffer (1964) used polarographic oxygen analyzers and automatic multi-channel recorders to simultaneously measure both diurnal dissolved oxygen and oxygen diffusion. As indicated previously, in this study a Beckman polarographic dissolved oxygen analyzer was used. The sensing probe consisted of gold and silver electrodes covered with a conductive salt gel which was held in place by a semi-permeable Teflon membrane. This same type of analyzer was used by Pamatmat (1965) to measure the metabolic activity of benthic communities. Biological Measurements Periphyton In many recent publications, periphyton has been used interchangeably with the term aufwuchs (Odum, E. P., 1959). Aufwuchs, as prOposed by Ruttner (1953), includes all organisms (both plant and animal) attached or clinging to stems and leaves of rooted plants or other surfaces projecting above the bottom (Odum, E. P., loc. cit.). In this study, periphyton is used to describe only the 19 autotrophic material growing on, but not penetrating the surface of organic or inorganic objects below the surface of the water (Sohacki, op. cit.). Thus, in this study periphyton includes all primary producers not defined as either phytoplankton (free-floating autotrophs) or as macrophytes (rooted aquatic plants). Recent advances in the quantitative measurement of periphyton have generally been attributed to original work done by Newcombe (1950). In his research work on Sodon Lake, Newcomb (loc. cit.) reported that studies of the growth of attached materials had been made as early as 1915 by Hentschel. Until the early 1950's, most studies of aufwuchs had been conducted in Europe. The method used to measure periphyton in this study-was a combination of techniques developed by several authors (Welch, 1959; Grzenda and Brehmer, 1960; Kevern, 1962; and Sohacki, op. cit.). The sampling period and choice of substrates closely paralleled those used by King (1964). Periphyton production was measured as the accumu- lation of chlorophyll on plexiglass substrates (140 cm2 exposure area) suspended 18 inches below the pond's surface. The substrates were removed at periodic intervals, placed in individual plastic bags, and frozen to facilitate re- moval of the attached algae. After freezing, the plexiglass 20 substrates were scraped clean with a rubber spatula and washed with 95% ethanol. The ethanol wash and scraped materials were placed in a 2 oz. bottle, and ethanol was added to bring the total volume of the sample to 50 ml. Each sample was then shaken vigorously and placed in a dark box for at least twenty-four hours to complete the chlorophyll extraction. A number of trials revealed that if kept in the dark, extracted chlorOphyll samples would retain a con- stant absorbency reading for at least thirty days from the time they were processed. Similar results were re- ported by Grzenda and Brehmer (op. cit.). Absorbency readings of the chlorophyll extract were made on a Klett-Summerson Colorimeter using a filter in the 643-700 mu range. To prevent cellular particulate matter from interfering with the absorbency readings, a 25 ml aliquat of the chlorophyll extract was carefully pipetted from the top of each sample for the colorimetric readings. An ethanol-chlorophyll solution absorbency curve conforms to the Lambert-Beer Law up to a reading of 200 (Klett Units with a 600-700 mp filter) (Grzenda and Brehmer, op. cit.). All readings taken during this study were under 200 and no correction factor was applied. Klett units were multiplied by .004 to obtain Optical density, which is equivalent to phytopigment units used 21 by other authors cited (i.e., Grzenda and Brehmer; Kevern; Sohacki; and King). For all graphical and tabular pres- entations, all periphyton phytopigments units were multi- plied by a factor of 103. Possible sources of error in measuring the absorbency of the extracted chlorophyll were carefully avoided. Samples which were accidentally shaken during the pipetting process, were allowed to settle before the Klett readings were made. The colorimetric cell was carefully rinsed and cleansed between samples and zero absorbency instrument readings were made with pure ethanol in the colorimetric cell. All periphyton samples were dried in pre-weighed crucibles at 55°C., after chlorophyll absorbency readings had been made. A representative number of the dried samples were placed in a muffle furnace at 550°C. and the loss of weight due to muffling was used as an estimate of the organic weight of the periphyton samples. King (op. cit.) and others have shown that periphyton growth on plexiglass plates placed vertical with respect to the surface of the water is significantly different than periphyton growth on plexiglass plates placed horizontally. Both horizontal and vertical sub- strates were used in this study and the results from each type of substrate were evaluated separately. 22 Periphyton growth was measured during three periods. At the beginning of each period twenty-four substrates were placed in the pond and every four days four substrates were sampled. To obtain significant colorimetric readings, the four substrates were pooled into two paired samples each consisting of two substrates (a total of 280 cm2 exposure area per sample). This sampling procedure was followed in measuring periphyton growth on both horizontal and vertical sub- strates in the two study ponds during each of three periods. Phyt0plankton Several methods have been developed to quanti- tively measure the production of phytoplankton. The light and dark bottle methods (Odum, 1956) and physical removal- of phytoplankton by filtration (COpeland and Dorris, op. cit.) have been used by numerous authors to measure phytoplankton productivity. However, both of these methods depend upon a relatively large quantity of plankton for accuracy. The ponds used in this study have comparatively small amounts of phytOplankton and attempts by the author to utilize either one of these methods failed to produce reliable results. The carbon-l4 technique was used with some success in two previous studies to measure the phytoplankton 23 production in the Lake City ponds (Knight, et al., 1962; and Sohacki, op. cit.). Both of these studies indicated that phytoplankton contributed very little of the total energy fixed by the primary producers. Thus, during the course of this study, only the standing crop of phyto- plankton was estimated. Periodically throughout the summer of 1965, 16 liter samples of water were collected and passed through a Foerst centrifuge (at a rate of 1 liter per 4 minutes). The precipitant was then scraped into a 2 oz. bottle and mixed with 50 ml of 95 per cent ethanol. The resultant chlorophyll-ethanol mixture was shaken vigorously and allowed to settle for 24 hours in the dark. The phyto- pigment concentration was read on a Klett colorimeter in the manner described previously for periphyton. Chlorophyll extracts of concentrated phytoplankton have been used by other authors to measure the standing crop of phytoplankton (Kosminski, 1938 from Prescott, 1951; and Welch, 1948). Recent studies lead E. P. Odum (op. cit.) to state, chlorOphyll may be, in some cases, a good measure of primary productivity when it is expressed on a square meter basis. Several authors have used ex- tracted phytoplankton chlorophyll as an index to potential productivity (Odum, 1957; Odum, Burkholder and Rivero, 1959; Minter, Copeland and Dorris, 1964; and Knudson and Dorris, 1963). 24 Oxygen Metabolism Diurnal changes in dissolved oxygen were used to estimate gross production by methods similar to those presented by Odum (1956) and Odum and Hoskin (1958). The free-water gas curves (diurnal curves) are particularly well suited to measure productivity in shallow homogeneous aquatic ecosystems, such as the ponds used in this study. Many factors may contribute to the observed dis- solved oxygen readings which are not a direct result of photosynthetic production. Diffusion from the atmosphere and oxygenated water entering the system from another source (accrual oxygen) may increase the dissolved oxygen in the water. While other factors, such as oxygen dif- fusing out of the water and respiration, may decrease the dissolved oxygen. If these factors, which add to or sub- tract.from the photosynthetic oxygen, are accounted for, the rate-of-change between two dissolved oxygen readings during the daylight hours will be a measure of gross photosynthetic production. Accrual Oxygen Accrual oxygen was discounted as a possible source of oxygen to the ponds as they were essentially closed systems isolated from any source of outside water, including run-off. 25 Diffusion Diffusion was determined indirectly by methods presented by Odum (1956). Initially several oxygen curves in the control and treated ponds were corrected for dif- fusion by the indirect method. The rate of diffusion depends on the degree of saturation of the water. Odum (loc. cit.) presented the following formula for determin- ing the rate of diffusion: D = KS where D is the diffusion rate of an area basis, S is the saturation deficit between water and air and K is the gas transfer coefficient (i.e., g/mz/hr for 0% saturation). To make practical use of this formula, K must be deter- mined or at least reasonably estimated. Although several K values were determined by a method presented by Odum (loc. cit.), this author questioned the reliability of these values because they were based on the assumption that respiration remained a constant at night. Gross oxygen production values were calculated from both diffusion corrected and uncorrected diurnal curves for several days and the greatest difference between any two of the values was less than 5 per cent. Because of the questionable reliability of this indirect measurement of diffusion and the small differences realized when diffusion corrections were made, oxygen 26 production values were calculated from diurnal oxygen curves not corrected for diffusion. Diffusion into or out of the ponds was assumed not to be a major source of error in determining either oxygen production or respira- tion values from the observed diurnal curves. This assumption was reinforced by two recent reports by other authors. Further studies of reaeration by Odum and Wilson (1962) lead them to state that because of changes in nighttime respiration the diffusion correc- tion had been previously miscalculated and when the diurnal curves were corrected as outlined by Odum (1956) the total respiration was often overestimated. Copeland and Duffer (1964) used a plexiglass dome to make diffusion estimates. They found that indirect measurements of diffusion in lentic communities were gross overestimates of the actual diffusion. They also illustrated that observed diurnal oxygen rate-of-change curves from shallow ponds changed very little when they were corrected for diffusion by the direct dome method. Respiration The rate of oxygen consumption (respiration) in an aquatic community may vary considerably from one day to the next. However, changes in the respiration rate during any twenty-four hour period are generally gradual and the average respiration rate can be closely 27 approximated from the nighttime dissolved oxygen rate-of- change curve. Figure 3 illustrates the correction made for respiration. Respiration was estimated by averaging the nighttime dissolved oxygen readings from the uncorrected rate-of-change curve. This average respiration was then added to each of the daytime rate-of-change values and a respiration-corrected rate-of-change curve was calculated. This area shaded in the corrected rate-of-change curve (Fig. 3) represents the gross photosynthetic production for one day (gm 02/m3/day). The average respiration rate (gm 02/m3/day) times twenty-four was used as an estimate of the total community respiration (gm 02/m3/day). For comparative purposes the volume production and respiration values were multiplied by the average depth to obtain values on a square meter or area basis. Since both ponds were maintained with an average depth of close to one meter, the correction factor for conversion to an area basis was 1.0. Computer Analysis Diurnal oxygen curves were analyzed through the use of the Control Data Corporation 3600 Digital Computer at the Michigan State University Computer Center. The computer program was designed by the author with assistance 28 Figure 3. Correction for respiration in the determination of gross oxygen production (T1 equals sunrise and T2 equals sundown). Shaded area in corrected rate-of-change curve is gross oxygen production for one day. Oxygen Gm/M3 Oxygen — Grams / Meter3 / Hour (Gm/ M3/Hr) 10.5 $0 U1 8.5 '0 mo (:10 Diurnal Oxygen 9 Fig.3 29 {>13 Noon 17 21 O .Yfi‘w nu»- ~;!- 1' v.“ A. o- nu- hove-l 30 from Computer Center personnel. Details of this program can be found in the appendix. The computer program calculated average respira- tion, gross production, and total respiration from observed hourly diurnal oxygen readings in generally the same manner that was previously described. Although this program was designed for a specific use, it could be utilized to ana- lyze diurnal oxygen curves from any aquatic community in which diffusion was not a significant contributor to the observed dissolved oxygen values. In some respects, the computer program used in this study is similar to the program designed by Armstrong (1963) to calculate production using carbon dioxide curves. Armstrong's program has been used successfully by Butler (1964) and others. P/R Ratios Ratios of gross oxygen production to community respiration (P/R ratios) were proposed by Odum (1956) to logically classify aquatic communities into autotrophic and heterotrophic types. When oxygen production and respiration values are used as coordinates of a graph, quantitative comparisons of a wide variety of aquatic communities can be made (Fig. 13). In this study, P/R ratios were used to illustrate changes in community metabolism which were a result of the 31 sodium arsenite application. They were also used to compare the Lake City Ponds with similar aquatic eco- systems analyzed in other studies. Efficiencies Ecological efficiencies are ratios of energy at different points in the food chain, expressed as percent- ages. Since there are many types of efficiencies, an abbreviated energy flow diagram has been used to define exactly what ratios are being considered (Fig. 4). In Figure 4, the abbreviations are defined as follows: L = Total light La = Absorbed light (Visible range) reaching surface of plant Pg = Total photosynthesis (gross production) Pn = Production of biomass (net production) R = Respiration Rt = Total Respiration (community respiration) S & E = Storage and export Consumer-decomposer organisms as used in Figure 4 include herbivores, carnivores, bacteria and any other non- producers. These abbreviations and definitions are presented by H. T. Odum (op. cit.), and they have been widely used by a number of authors to express ecological efficiencies. 32 uuomxm cam mmMHODm u m a m Acoflumuflmmmu >DMCSEEOUV cowumufimmmu Hmuoa H um Asofluosooum umcv mmmfiofln mo :ofluosooum N am Acofluosooum wmoumv mHmeUQMmovonm Hmuoa H mm mpcmam mo momeSm mcflnommu .Aomsmu wanflmw>v unmfla Connemng n ma Bnmfiq Hence n q "mums; A.uflo .mov Esoo .m .m Eoum @mpmmom EnumMHo 30am amumcm .v musmflm «a: 33 9.050230 nsonouEOuon udm I 09 ‘9... V --r A n ~' bl”- . . n: 1 .1_. I r . ("-l .uu d ulls ; \(J n “7". 5"“ .' 11‘ 5.. .. .‘u x. L. 34 Trophic level energy intake efficiency (Lindeman's Efficiency) for the primary level is then 1:901:23 L La where L is total sunlight energy at the surface of the water and La is the total visible light energy available at the surface of the plants. To calculate efficiencies, oxygen production values (gm OZ/mz/day) were converted to energy units (kg-cal/mz/day) by an approximate conversion factor of 4 kg-cal/gm O2 metabolized. Total light energy (L) as determined with the pyrheliometer (g-cal/cmz/day), was converted to kg—cal/mZ/day and absorbed light energy (La) was estimated to be approximately 20 per cent L (Odum, Burkholder, and Rivero 1959; Reid 1961). Conversion of Oxygen Values to Energy Units Several factors are involved in the conversion of oxygen production values to energy units. These factors vary depending on the composition of -(CH20)-(the photo- synthetic product) in the following reaction: CO 0+0 +H 0 = -(CH20)-+H 2 2 2 2 The caloric content of the photosynthetic product can be determined using bomb calorimetry. Kevern (Op. Cit.) demonstrated that the caloric content of the photosynthetic product is greater than that of carbohydrate 35 alone, thus indicating the presence of either protein and/or fats which have higher caloric values. By direct analysis Kevern (op. cit.) found that the photosynthetic product of certain algae was 76% carbo- hydrate, 21% protein, and 3% lipid with an average caloric value of 4522. cal/gm dry weight. The synthesis of protein indicates a PO (photo- synthetic quotient, moles of 0 produced/ moles C02 2 consumed) greater than one (Strickland, 1960). The higher the caloric content of the organic product the greater the P0. P0 values ranging from 1.0 to 1.6 have been reported by various authors (Kevern, op. cit.; Strickland op. cit.; Ryther, 1956; and Odum, 1957). Thus, if the caloric con- tent of the photosynthetic product is determined, the PQ value can be reasonably estimated. An estimate of the P0 is essential to establish the relationship between photosynthetic oxygen production and the organic product. If the PQ is 1.0, 6 moles of CO2 will produce 6 moles of O2 and 1.0 mole of -(CH20)-. To convert oxygen gas production to energy units, you must: (1) establish the number of moles of 02 released for every mole of -(CH20)- produced using the estimated PQ value; (2) divide the molecular weight of the organic product by the weight of the moles of 02 released to obtain the organic weight equivalent of the gas; and (3) multiply the organic weight equivalent by the average caloric 36 content of the organic product. The result of these steps is a conversion factor by which gm Oz/mZ/day can be con- verted into kg-cal/mz/day. Time did not allow the actual determination of caloric values of the photosynthetic products of the Lake City ponds so the factor used to convert oxygen production values to energy units was estimated from those found in the literature. Duffer and Dorris (1966) used a factor of 3.75 kg-cal/ngz; Copeland (op. cit.) and Butler (op. cit.) used a factor of 3.5 k-cal/ngz; and Odum and Wilson (op. cit.) used 4.0 k-cal/ngZ. Kevern (op. cit.) used several different factors depending on the caloric con- tent of the organic product and the estimated PQ. Since the comparisons to be made later are primarily with results obtained by other authors using 4.0 k-cal/ngz, this same conversion factor was used in this study. Actually, Kevern's (op. cit.) work indicates most conversion factors should fall in the range between 3.5 and 3.9, unless caloric values or PQ's are actually determined, however, the factor chosen is somewhat arbitrary. RESULTS AND DISCUSSION 1963 Preliminary Study The results of the 1963 study reported here were one phase of a larger more comprehensive study involving the ecological effects and translocation of radioactively tagged (arsenic 74) sodium arsenite in the Lake City ponds (Ball and Hooper, 1966). Automatic monitoring of the pH and dissolved oxygen was begun on the day of treatment as were the periodic measurements of alkalinity. Hydrogen Ion Concentration The continuous pH measurements in Pond C were initiated some four hours after the arsenite was added in 1963. Subsequently, little is known about the immediate response of the pH to the treatment. However, grab samples indicated afternoon pH values in Pond C were 8.8 or greater before the pond was treated. The maximum pH recorded on the day of treatment was 8.65. Within twenty-four hours after the application, the pH had been reduced to 8.0. The minimum pH in Pond C (7.1) occurred a week after the treatment (Figure 5). 37 38 Figure 5. Daily maximum and minimum hydrogen Ion concentrations in Pond C after 1963 treatment 39 pH maximum 9.0 , 1 minimum 8.0 . 7 15 23 31 . . July August Sodium Arsenite Added Fig. 5 40 It was two weeks after the herbicide application before pH values greater than 9.0 were recorded in Pond C (Figure 5). Periodic samples from adjacent ponds during the same period ranged in pH from 8.4 to 9.8. Within four weeks the pH values in the treated pond returned to normal when compared to adjacent ponds. Alkalinity (Methyl Orange - Phenolphthalein) Periodic sampling of the treated pond, indicated the total alkalinity increased a minimum of 30 mg/l within two days after the arsenite was added. Pretreatment total alkalinity in Pond C ranged below 70 mg/l and a maximum of 106 mg/l was recorded two days after the addition of the herbicide. Adjacent ponds ranged below 70 mg/l total alkalin- ity throughout the study period. The total alkalinity in Pond C appeared to return to normal within three weeks after the treatment. Phenolphthalein alkalinity in Pond C declined with corresponding decreases in pH following the application. Four days after the treatment, phenolphthalein alkalinity was completely absent. For two weeks following the appli- cation of the sodium arsenite, phenolphthalein alkalinity remained low (below 6 mg/l) and periodically was absent when the pH drOpped below 8.0. 41 In adjacent ponds phenolphthalein alkalinity averaged above 6.0 mg/l and ranged in values from 1.0 mg/l to 19 mg/l during the three weeks following the applica- tions. Factors in Alkalinity and pH Responses As indicated by the results, pH and alkalinity are interdependent. In Pond C the rapid decay of aquatic vegetation produced large quantities of carbon dioxide (C02) which initiated changes in both pH and alkalinity. Free carbon dioxide released into an aquatic environment as a product of respiration or bacterial.de- composition of organic matter, immediately combines with available monocarbonate to form bicarbonate. No change in pH or total alkalinity occurs if sufficient monocarbon- ate is available to combine with all the carbon dioxide (Welch, 1952; Welch 1948). In the absence of monocarbonates, free carbon dioxide combines with water to form carbonic acid and the pH of the aquatic environment is lowered (Ibid.). Under normal conditions the Lake City ponds are slightly alkaline (pH 8.0 to 9.0) and have sufficient monocarbonates in solution to resist large fluctuations in pH. When the sodium arsenite was added to Pond C rapid decomposition of organic material followed, producing 42 large quantities of carbon dioxide. Available carbonate (C03), measured as a function of phenolphthalein alkalin- ity, was quickly converted to bicarbonate (HCO3). However, the production of carbon dioxide continued and eventually exceeded the buffering capacity of the available monocar- bonates. The excess carbon dioxide formed carbonic acid, thus lowering the pH. Total alkalinity of the treated pond was increased when bound carbonates in the bottom muds and plant encrustations were brought back into solu- tion as bicarbonates. The reverse process was initiated when production of carbon dioxide decreased and photosynthesis of the remaining plants increased. The pH returned to the normal range and some of the bicarbonates were precipitated as insoluble carbonates lowering the total alkalinity to the pretreatment levels. Dissolved Oxygen The dissolved oxygen (D.O.) in Pond C showed an immediate response to the addition of the sodium arsenite. The pond was treated early in the morning when D.O.* normally begins to increase with increasing photosynthetic activity. Just prior to treatment the D.O. was 8.5 mg/l. Just after the arsenite was added, it began to decrease and in less than twenty-four hours, it had dropped to 5.8 mg/l (Figure 6). 43 Figure 6. Maximum-minimum dissolved oxygen (D.O.) and production/respiration (P/R) ratios in Pond C following 1963 treatment 44 A\ l \\ [I v n 5 \ A l 5", III-II J \1 x A.“ / A \l' .19 L \ v . // s. I l ‘I’\ u . N \ ’4 A “U. \ \/ R l|n|||I|I|I\ 1.l D A \" I I'II'" / /. m I L \4 \ U V 11 \. .al 1..\ n \ 1., .I 1.\ I / III, 1 d \ \\ I y l ‘I' a / I IIHHHII 1 A M // |||l\i.1 m .IU: ‘5‘ n I _ r b _ p p p h i p . S 0 8 6 0 J 2 0. 8. 2 cl cl 52.5 .3: Fig.6 45 Although the D.O. increased slightly during the daylight hours of the next day, the values continued to decline further during the night. The lowest D.O. (4.8 mg/l) was recorded three days after the application of the herbicide. A D.O. greater than the pretreatment value of 8.5 mg/l was not recorded for more than a week after the treatment (Figure 6). During the week following the treatment, grab samples of the three adjacent ponds (ponds A, B, and D) showed no significant decrease in D.O. and they ranged in values daily from 7.0 mg/l to 12.0 mg/l. Throughout July and August of 1963, D.O. values from Pond C averaged much lower than those of the adjacent ponds. Oxygen Metabolism The dissolved oxygen readings suggest that the addition of sodium arsenite and the subsequent removal of higher, rooted aquatic plants severly limited photosyn- thetic activity in the treated pond. Although there were no pretreatment or control oxygen metabolism determinations made in 1963, Figure 7 clearly suggests a major reduction of gross photosynthetic production (Pg) occurred immedi- ately after the herbicide was added. Gross oxygen production increased gradually from a summer low of 1.0 gm-OZ/mg/day on the day of treatment 46 Figure 7. Available solar energy (La), efficiency (Pg/La), respiration (Rt) and production (Pg) in Pond C following 1963 treatment K-Gm 031/lean Percent Gm Oglelan Gm Oleleay d N, O O 400 2.5 47 I _\ I" .— I ’ I\/AV/ \\/ \ A / \\ /v / \ / ’ \ I\ / V \ \’ Solar Energy Efficiency ‘ \ N « \\\/\~I/ v \/\\___.a/ V'V’ ’\ Respiration t it 1 ’ 1H1 ’1 H, I i I\ I\'/ f\ ’1 V I I‘ I I ’ \ I 'l /\ I I I \ I’ 1 / \ J \ II\ I\’ \J L/ \/ \/\V/\/’ v Production A A f [\l \I\ r\ I " \ ,1 II \ /\ I \\ I, \\ I / A /' H \l \I I/ \ / l \, ‘ V V \/ /"I 7 n 15 19 23 27 I - a .~ July August Sodium Arsenite Added I Fig. 7 48 to a high of 6.9 gm-OZ/mZ/day, seventeen days after the application (Figure 7). Community respiration in Pond C showed similar trends. A summer low of just over 1.0 gm-OZ/mz/day occurred three days after the arsenite application. The highest respiration value (7.2) occurred the same day as the highest production value, seventeen days after the treatment (Figure 7). While oxygen production and respiration values closely paralleled each other throughout the summer, they demonstrated very little correlation with solar energy until three weeks after the arsenite was added (Figure 7). The lowest observed gross production-community respiration ratio (P/R ratio) of 0.4 was recorded on the day of treatment. Although the P/R ratios in Pond C increased significantly during a short recovery period following the treatment, less than one third of the observed readings were greater than 1.0, indicating the pond was essentially a heterotrophic community during most of the growing season (Figure 6). The efficiency (Pg/La) in Pond C was less than 1.0 per cent for four days following the treatment. A high of 3.0 per cent occurred a week after treatment and corresponded to lowest observed solar energy reading. During the remainder of the summer the efficiency ranged between 1.0 and 2.5 per cent (Figure 7). 49 Gross production, community respiration, efficiency and P/R ratios all indicate the metabolism of Pond C was at least temporarily inhibited by the addition of the herbicide. The P/R ratios and the efficiency cal- culations suggest the productive potential of the pond may have been significantly reduced by the removal of the higher aquatics. A more thorough discussion of these results will be made later. Temperature The water temperature in Pond C varied between 70° and 90°F. throughout the summer study period in 1963 and corresponded very closely with changes in air temperature. Although increases in solar energy were often reflected in temperature, the oxygen metabolism of the study pond showed no direct correlation with changes in temperature. The role of temperature in aquatic community metabolism regulation was discussed by Odum and Wilson (1962). Their analysis of one hundred and twenty-three diurnal oxygen curves from Texas bays indicated community production and respiration values corresponded very closely with incident radiation and that even rather large fluctua- tions in temperature had little effect. They concluded that while changes in temperature have a pronounced effect on the rate of physical gas exchange, an aquatic system and its organisms are sufficiently organized in their activity 50 to keep total metabolism in phase with food conditions as developed seasonally from light energy, generally independent of temperature. Temperature fluctuations in the near optimum range (70-90°F.) in the study ponds probably played a very minor role in controlling community metabolism. Conductivity When compensations for changes in water tempera- ture were made, fluctuations in conductivity of the study pond water were below the sensitivity of the automatic analyzing equipment. There is little doubt that some changes in conductivity did occur. Immediately after the herbicide was added, the total alkalinity increased, indicating a rise in the total dissolved solids. Although not measured on the conductiv- ity meter, this increase in dissolved solids probably affected the conductivity. 1964 Study An analysis of the 1963 study results uncovered many areas where further information and adjustments in sampling techniques were needed. Thus in the 1964 study, a control pond (Pond D) was selected to provide a compari- son for changes occurring in the treated pond (Pond C). Measurements of phytoplankton and periphyton were made in 51 both ponds in 1964, and simultaneous dissolved oxygen readings were maintained for both ponds during critical periods. Automatic monitoring of pH and conductivity was discontinued in 1964. However, periodic measurements of pH and alkalinity were made in both ponds as a check of the 1963 results. The date of the sodium arsenite treatment in 1964, was moved to July 19 to provide pretreatment information in Pond C. Periphyton The production of periphyton was designated as a rate of accumulation of chlorophyll on artificial sub- strates expressed as phytopigment units (optical density of extracted chlorophyll). In the 1964 study, both hori- zontal and vertical substrates were used and the results were analyzed separately. The slope of each periphyton growth curve (Fig. 8 and Fig. 9) was considered the rate of production. The periphyton production rates of Pond C and Pond D were compared using methods presented by Li (1957) for analysis of covariance with replicate samples (Tab. I and Tab. II). Three intervals were used to compare periphyton production on vertical substrates in Pond C and Pond D. Interval I (June 25 - July 18) was chosen to compare the 52 vmma Ga 9 Unom one 0 boom CH magnumnsm HMUHuHm> so sofluosooum cowhnmfluom mm .m ousmflm 53 a a: :32 3232 528m 532 - :2 2:: 2- .2 T d Jo. III- 253.: o 23.. Ill 28.38 a 23.. tuawaidotIud 000'! X 511”" 54 vmma CH a once one O oaom ca oumuumnsm amusoNHHon so sofiuosooum goumnmfluom .m ousmfim 55 Tm: .522 3E3; 533m «SEE . . , 2.: 2:: m mm em . m. E m e mm 4} i1 1 £ 'T * + J rlll @385 o 28a II 22:88 0 Ban. oN_ om: ruawfirdomud X 5111111 OOO'I 56 . Q GCOQ UCM O Usom aw moumu COHuosooum coosuon COwumaum> mamnqo momoam macaw goflpmaum> «a .Hm>oH «mm map um mogmquMHU usmoflmasmflm i mm.ma m mm.¢ma Hmseummu emaoom HHH Hm>umocH «gmomoam «He.o me.eoa a me.koa macaw soHBMHnm> marm on oath mm.mm m NH.HAG Hmseflmmn emaoom HH Hm>umuaH «emmmon «mo.» om.mmm H om.mmm macaw cofiumflnm> mum on ere 4m.ovm m mm.e~k.~ Hmseflmmn emaoom H Hm>nmucH «xmomoam He.o NA.H4H H NA.H4H macaw aofiumnum> mark on mmrm owumm mumsqm Eooooum moumsvm cOHDMHHm> m coo: mo moonmoo mo 85m mo mousom Hm>umucH suzouw lemma an mopmuumnsm HMUfluum> co usosbflm0phnm mo coapmHsfisoom mo mfiuou CH come mGOmHHmmEOUV Aaouusouv o ocom msmuo> Aomumouav O osom sH sofluosooum coumgmfluom mo mommaum>oo mo mammamsm H mflmdfi 57 .a econ can 0 osom CH mmumu cofiuosooum cmozuon cowumwum> mamsqm momon macaw COHDMHHM> ii .Hm>mH mmm map um mocmanuHo ucmonHamHm ¥ mm.em~ A Hm.mHo.~ HmseHmmn emHoom HHH Hm>nmpaH «emmmoam mm.m Ho.mem H Hm.mem macaw aoHumHnm> mum on cars mm.hm m Hm.mom HmseHmmn emHoom HH Hm>nmch «emomon «Sk.s os.mm~ H oe.mm~ mcosm coHpmHum> mum on ere om.emo.H m ov.~mm.mH HmseHmmn emHoom H Hm>nmuaH eemomoam Hm.o am.mHm H mm.aHm macaw coHHmHnm> erk on mmrm oflumm oumsqm Eoomoum mmumsqw GOHDMHHM> m coo: mo mmoumoo mo 85m mo condom Hm>uoch £u3ou0 HemmH cH moumuquSm amusoNHHon so ucoEmHmoumnm mo soflpmasasoom mo mEuou sfl coma mCOmflHmmEOOV Aaouusouv o osom msmum> Acoumouev O osom cfl cofluosooum sonagmfluom mo oocmwum>oo mo mammamsd HH mqm<8 58 production rates in ponds C and D under natural conditions prior to any treatment. Interval II (July 7 - Aug. 2) was chosen to illustrate the effects of sodium arsenite on periphyton which had been allowed to establish some growth on the substrates before the treatment. Interval III (July 20 - Aug. 13) began one day after Pond C was treated with sodium arsenite and it was used to determine the effects of the herbicide on the establishment of new peri- phyton communities (Fig. 8). Periphyton production on vertical substrates in the two ponds showed no significant difference during Inverval I. During Interval II the treated pond exhibited significantly greater periphyton production, and in Inter- val III the periphyton production on vertical substrates was significantly greater in the control pond (Fig. 8 and Tab. I). Although at first the results appear to be contra- dictory, a closer analysis reveals a correlation between the arsenite treatment and the observed periphyton production. Under natural conditions the periphyton production in the two ponds is essentially equal. The periphyton which had been allowed to establish itself on the substrates in Pond C before the herbicide treatment, was apparently not affected by the sodium arsenite and eventually utilized the nutrients made available by the decaying macrophytes, to increase production. Colonization 59 of clean substrates placed in Pond C after the treatment was probably inhibited by the presence of the sodium arsenite in the water, and hence, observed Pond C periphy- ton production during Interval III was less than that in the control pond-~even though an increased quantity of nutrients was available. Horizontal substrates were also sampled in the same three intervals indicated for vertical substrates. Periphyton production on horizontal substrates during Interval I and Interval II showed the same results as the vertical substrates. The production of periphyton on horizontal substrates was not significantly different in the two ponds during Interval I, and the periphyton pro- duction was significantly greater in the treated pond during Interval II (Fig. 9 and Tab. II). Periphyton production on horizontal substrates during Interval III was not significantly different in the two ponds. Although this measurement of periphyton pro- duction in Interval III does not correspond to the results obtained from vertical substrates, it does support the contention that colonization of the new substrates was inhibited by the sodium arsenite. If the herbicide had not interfered with colonization, the periphyton production on horizontal substrates during Interval III should have been significantly greater in the treated pond when compared to the control pond because of the increase in 6O nutrients--it was not. The difference between the results on the vertical and horizontal substrates can be explained by the fact that horizontal substrates always "seed on" or colonize quicker than vertical substrates. A certain minimum quantity of plant material on a substrate is required before the logarithmic or arithmetic growth phase of the periphyton community begins (Kevern, op. cit.; King, op. cit.). Normally, this requires less than four days in the Lake City ponds. The accumulation of organic material begins more rapidly on horizontal substrates. Thus, if colonization is inhibited, the hori- zontal substrates could be expected to demonstrate greater periphyton production than the vertical substrates in a short term comparison. In summary, the periphyton production in Pond C, as measured by the artificial substrate method, increased significantly after treatment. Even though the arsenite inhibited the colonization of clean substrates, the natural periphyton growing in the pond was probably capable of utilizing the new source of nutrients to temporarily in- crease production. This increase in production in Pond C was estimated to have lasted for thirty days after the treatment. Large floating mats of Spirogyra Sp; appeared in Pond C ten days after the application of the herbicide and persisted for about one week, which gave further evidence 61 that the decaying macrophytes had released nutrients. Spirogyra blooms of this type were characteristic of fertilization treatments applied to these same ponds in previous studies. Phytoplankton Because of the time involved in sampling, only a limited number of plankton samples were taken. However, samples obtained did provide an adequate index to the standing crop of phytoplankton expressed as phytopigment units. Figure 10 illustrates the effect of the sodium arsenite on the phytoplankton in Pond C compared to the untreated control, Pond D. The most dramatic evidence, showing the immediate effect of sodium arsenite on phyto- plankton, was the difference between the two samples taken on the day of application. One sample was taken in the morning before treatment and yielded .051 phytopigment units. The afternoon sample, after treatment, exhibited only .031 phytopigment units (Fig. 10). The control pond showed no difference between morning and afternoon samples on the same day. The sodium arsenite apparently destroyed a portion of the phytoplankton on contact; however, as the concentration of the herbicide diminished, the phytoplank- ton recovered. Sohacki (op. cit.) reported similar results when he measured the effects of sodium arsenite on phytoplankton 62 mmHmEmm oomswfluucoo Honda ma Eoum woma cfl couxcmHmoumgm mo mono mcflocmum on xoosH .oa musmflm 63 3 2 a: 822 see; 522w .. EN 2 5 .11.... 9385 o 2:: III 28:38 9 23¢ S TIE—1. $3 suun tuawBIdoMud C e X OOO'I O LO 64 productivity using the Cl4 method. He concluded that the slight reduction in photoplankton productivity that oc- curred following the treatment and the short recovery time demonstrated the resistance of algae toward sodium arsenite toxicity. Sohacki (op. cit.) reported that the phytoplankton in the treated pond recovered to pretreatment production levels within six days after the application. As can be seen in Figure 10, the standing crop of phytoplankton in Pond C returned to pretreatment levels within four days after the arsenite was added in 1964. Unfortunately, the Spirogyra sp. blooms which occurred in Pond C after the treatment did not lend them- selves to quantitative measurement by either of the methods being employed to measure changes in periphyton and phyto- plankton production. These blooms were of sufficient magnitude to have significantly increased the total oxygen production in Pond C. With the exception of these blooms, no increase in phytoplankton standing crop was detected in Pond C after the treatment, although periodic samples were taken for a month after the application of the herbicide. Properly considering these Spirogyra sp. floating mats as part of the phytoplankton population, the phyto- plankton production in Pond C certainly increased significantly a short time after the herbicide was added. 65 Although no quantitative measurements of this increase in production were made, visual observations indicated it lasted for about seven days. Dissolved Oxygen The D.O. in Pond C in 1964 again demonstrated an immediate response to the sodium arsenite treatment. The herbicide was applied in the middle of the afternoon and within two hours the D.O. dropped from 9.5 mg/l to 8.0 mg/l. Within two days the D.O. reached a summer low of 4.2 mg/l (Fig. 11). It was almost two weeks after the application before the D.O. in Pond C exceeded pretreatment values. The control pond during this same interval had no D.O. values below 7.5 mg/l, and some values ranged as high as 11.5 mg/l. Pretreatment D.O. values in Pond D and Pond C were nearly equal. It was misreported in an early paper on this same study that the D.O. values in the treated pond reached a minimum of only 6.0 mg/l (Bails and Ball, 1966). A re- checking of twenty-four hour dissolved oxygen recording charts revealed this previous error. The possible reasons for a lower minimum D.O. following the 1964 treatment, rather than the 1963 treatment as previously reported, will be discussed later. Figure 11. 66 Maximum-minimum dissolved oxygen and production/respiration (P/R) ratios in Pond C during 1964 study 67 Euro _ B: a I \ / \ // H In \ \ \ \ \ .I \ \J / \ II \\/. / . iiiii I, / h /1. / l\ . R . 0 A / \/ A / \ 4 A .VI / / h‘u‘l‘ll u \\\v \\ A, m \ \\\I [I’ll a A \ Q, \ J4 d\ N \ . J O 1.111. I P .m ‘n‘|l“|llll‘|ul|ll // I/ .m III." lllll \ 'I \W \ l/ \ a \\ pl, //0 \ H _ _ _ _ u H F _ p a u o a 6 4. o. 6. o. a II 2 2 II cl so 26 22 O . 14 I 10 August Sodium Arsenite Added July Fig." 68 Oxygen Metabolism The pretreatment oxygen production values in Pond C ranged from 2.3 to 9.5 gm-OZ/mz/day. Within one day after treatment, the gross oxygen production in Pond C was re- duced to 1.6 gm-Oz/mz/day from a 4.2 value the day before. The control pond, Pond D, also exhibited a reduction in oxygen production (i.e., from 2.1 to 1.4 gm—OZ/mz/day) the day after the arsenite was added to Pond C. It was very small in comparison, however, and it was attributed to a decrease in solar energy (Fig. 12). Efficiency in Pond C was reduced to a summer low of 0.7 the day following the arsenite application. The pretreatment efficiency percentages in Pond C showed an inverse relationship with solar radiation (i.e., high radiation, low efficiency) and they ranged between .72 and 5.0. Post-treatment efficiency values in Pond C never exceeded 2.0. The efficiency values in Pond D were considerably less than Pond C. They remained relatively constant and inversely related to solar energy throughout the study period. The inverse relationship of solar energy and oxygen metabolism efficiency will be discussed later. The P/R ratios in Pond C ranged from .8 to 2.4 before treatment and dropped to a summer low of .45 the day following the treatment. Post—treatment P/R ratios Figure 12. 69 Available solar energy (La), efficiency (Pg/La), respiration (Rt) and production (Pg) in Pond C (treated) and Pond D (control) during 1964 study 7O s /~ ,. $1200 I \‘/’\\ /\\ /\‘// ~ \\ A s .0. / \ I \ :3 \ I \J 5 400 ll Solar Energy 4 ' [rPondC *5 3 g I- /\I l /‘ \ 33 2 I .\ /\/\ ~ \\/,~"\ ’ \\J/ \‘l/ l EffIcrency Pond D r A fr—PondC /\ I\ / \ 5*-I\\ I I \ IA ‘ l/ I I \ I V \ r~ le l \\ s I \ I \ I l I 1 I A e I \ l \ I \ I ‘ /' N I l I \ 5 3-I \ I \I I/\ I ‘u' l \ I 'I \ I o I \ ‘_ J PondD \I Ifrespiration I? l\ ‘ I PondC(Treated) I\\ 5.1, \1 II II \\ A x g . \ ,I l/l ,\ l/ \ ~ , \ . e?- l \ I ' \ l \ / \ H l \ l, E 37' \ / \ I ‘ / \ I \ / <5. Lv/ I I \\ I \J 5 I . I/ A \I , ,. Po dD (Control) I Production . ‘ IO 14 18 22 26 30 3 7 II A July Sodium Arsenite Added August Fig. 12 71 in Pond C did not exceed 1.65. The R/R ratios in Pond D were quite low all summer and their comparison to Pond C is nearly meaningless except to say that P/R ratios in Pond D showed no significant change during the period immediately following the treatment of Pond C. Summarizing the reaction of oxygen metabolism in Pond C to the arsenite treatment, it appears that while the oxygen production recovered to pretreatment levels quite rapidly, the efficiency and P/R ratios were signifi- cantly reduced for the entire study period. Hydrogen Ion and Alkalinity The hydrogen ion content of the treated pond in 1964 increased after the application of the sodium arse- nite as was predicted from the preliminary study. The pH of Pond C varied between 7.5 and 8.7 before treatment and dropped as low as 6.00 after treatment. The control pond maintained a pH above 8.0 throughout the study period. Again, as predicted by the preliminary study, the total alkalinity increased in Pond C after treatment, while the total alkalinity of the control pond remained relatively constant throughout the study. The total alkalinity in Pond C ranged between 60 and 70 mg/l before treatment and climbed as high as 97 mg/l after treatment. Pond D main- tained a total alkalinity between 60 and 70 mg/l all through the summer. 72 Comparison Between 1963 and 1964 Treatments In both 1963 and 1964, Pond C was treated with sodium arsenite, and although the reaction of the pond metabolism to each treatment was similar, the magnitude and duration of the effects were different. Before dis- cussing these differences, the condition of the pond eco- system before each treatment and timing of the arsenite application in each year will be analyzed. In 1962, Pond C was treated with copper sulfate for a previous study (Sohacki, op. cit.). This treatment removed much of the algae, Chara sp. which, under normal conditions, is common in these hardwater ponds. By the time the sodium arsenite was applied in 1963, the higher aquatic macrophytes, predominately Potamageton natans and P; praelongus, were well established. The abundance of the macrophytes in Pond C in 1963, would not have, under normal circumstances, required removal to obtain full utilization of the pond. The Chara sp. in Pond C, although present in early 1963, had not recovered fully from the previous year's copper sulfate treatment. Periphyton and planktonic algae in Pond C were probably exhibiting normal production before the 1963 sodium arse- nite application. The 1963 treatment removed all submergent aquatic macrophytes with the exception of Chara sp. Although these higher aquatics showed some signs of recovery by September 1A1 01‘ -v v. ’5 n: 73 1963, they had obviously not fully recovered before the arsenite was again applied in June of 1964. In 1963, the sodium arsenite was applied early in the morning. In 1964, the arsenite was added at the peak of photosynthetic activity in the late afternoon. The treatment came two weeks later in the growing season in 1964. In general, the herbicide produced more pronounced effects in 1963. The pH dropped much lower, the total alkalinity increased more, and the gross oxygen production was suppressed much longer in 1963 than in 1964. These differences can be accounted for by the fact that there was a larger quantity of macrophytes susceptible to removal by sodium arsenite before the 1963 treatment. Thus, the entire metabolism of the pond was more dependent on these plants and the ecosystem responded more dramatically to their removal. The minimum D.O. in Pond C was lower after the treatment in 1964 compared to 1963, and this observation is difficult to explain in view of the relative abundance of macrophytes in the two years. Perhaps the answer lies in the unusual weather conditions that persisted in 1964. The days immediately preceding the 1964 treatment were very cloudy and solar radiation was less than 1000 kgm-cal/mz/day. The maximum D.O. in Pond C the day before the arsenite was added in 1964 was only 10.0 mg/l. In the 74 two days following the 1964 treatment, the D.O. was reduced to 4.2 mg/1--a gross D.O. reduction of 5.8 mg/l from the pretreatment high. The overcast weather following the 1964 treatment restricted even the remaining algae from producing suffi- cient oxygen to replace that being consumed by the decaying macrophytes. The gross reduction in D.O. following the 1963 treatment amounted to 6.5 mg/liter in a period of three days from a maximum the day before treatment of 11.5 mg/l, to a low of 5.0 mg/l two days following the application. Thus, even though the minimum D.O. recorded after sodium arsenite treatment was less in 1964 than in 1963, the gross D.O. reduction from a pretreatment high was greater in 1963. The overcast conditions in 1964 perhaps obscured the more pronounced effects of the arsenite treat- ment in 1963 when the minimum D.O. values of Pond C after the two treatments, are compared. Comparison of Results with Other Studies One of the best means of testing the validity of the methods used in this study is through comparing the results with those obtained by numerous other authors in similar attempts to measure aquatic community metabolisms. n n \ .—b n.‘ o i ii! .A u '- 'o- i~| F ‘ 75 Solar Radiation An accurate measure of total solar radiation is essential to daily calculations of metabolism efficiency. Solar radiation figures used in this study were recorded at the study site with an Epply pyroheliometer and they were very similar to those recorded by Crabb (1950) in his studies at East Lansing, Michigan. While the maximum daily total solar radiation values recorded at Lake City were higher than those recorded by Crabb (op. cit.) the mean weekly values were not significantly different. Oxygen Metabolism Oxygen production and respiration, and P/R ratios have been calculated for numerous lentic communities (Copeland, Whitworth 1963; Odum, 1956; Knudson and Dorris, 1963; and Odum, 1957). The range of values reported by these authors are graphically presented in Figure 13. This oxygen production, oxygen respiration graph was first proposed by Odum (1956) to classify biological communities into heterotrophic or autotrophic types. The diagonal line from the lower left hand corner to the upper right hand corner of the graph represents P/R ratios of one or unity. Points to the right of this line have P/R ratios less than one and points to the left have P/R ratios greater than one. Figure 13. 76 Range of oxygen metabolism values from several ponds: (1) Lake City Ponds; (2) Four Oklahoma Farm Ponds (Copeland and Whitworth, 1963); (3) Several Northern Ponds (Odum, 1956); (4) Oklahoma Ponds (Knudson, and T. C. Dorris, 1963); and (5) Marine Turtle Grass Communities (Odum, 1957). Shaded areas for (1) and (2) encompass the wide range of values reported in those studies am (I, / M9’/ (my ['1 (uhu‘lrun 77 P/R Ratio > 1 Autotmphic 40—- t—I COO (J1 Production — Gm 02/M2/ Day F/ 94 1 I 1 I 1 1 J— 1 a .9 l 3 5 7 9 Respiration - GmOz/MZ/an Fig.13 10 aiudoomatan 1 > ones tl/d .‘ -- -' ‘ .44 78 This type of graphic presentation not only allows for a visual comparison of production and respiration values from several different communities, it can also be used to classify community types and predict the ecological succession in terms of oxygen production and respiration values. The values recorded in Figure 13 for the Lake City ponds represents 95% of the total range of values. Ex- tremely low production and respiration values in Pond C, occurring after the herbicide was added, were deleted so that the scale of this graphic comparison would be meaning- ful. The values recorded for this study are in the right magnitude when compared to the results of these other authors (Ibid.). The range of values recorded in Figure 13 represent only the productive summer periods for all the communities. The annual herbicide treatments in the Lake City ponds are at least partially responsible for the fact that the Lake City pond values are somewhat lower and to the left of the values recorded for other ponds in Figure 13. The range of oxygen production and respiration values in Figure 13, most similar in magnitude to the Lake City ponds, were reported by Odum (1956). These values calculated and presented by Odum, were extracted from information collected by Juday, Blair, and Wilda (1943), on Little John Pond, Wisconsin; and by Riley (1940) on 79 Linsley Pond, Connecticut. It is interesting to note that both these ponds are in nearly the same latitude as the Lake City ponds and all three sites would, therefore, be exposed to nearly equal solar radiation. The oxygen metabolism values reported by Copeland and Whitworth (1963) exhibit wide variation both to the right and to the left of the P/R unity line in Figure 13. One of the four ponds studied by Copeland and Whitworth was also treated with a herbicide, which perhaps accounts for the wide range of values found in both the Oklahoma ponds and the Lake City ponds. The four Oklahoma ponds were all regularly subjected to fertilization through run- off from land under varied forms of intensive agriculture, making them considerably more productive than the Lake City ponds. This is clearly illustrated in Figure 13. The oxygen metabolism values of a Marine Turtle Grass Community (Odum, 1957), were included in Figure 13 to compare the magnitude of one of the most productive, naturally occurring lentic aquatic communities with various small ponds. Efficiencies Figure 4 indicates the various types of ecological efficiencies that can be calculated for the transfer of energy from solar radiation through the first trophic 1evel--the producers. Generally, these efficiencies are .- pvt ’V ‘ . i 80 one of four types: (1) Pg/Li; (2) Pg/La; (3) Pn/Li; or (4) Pn/La. Each of these ratios represent a different type of energy transfer efficiency. Table III compares the calculated efficiencies in these four categories from various aquatic communities with the ratios obtained from the Lake City ponds. With the exception of Lindeman's figures on the Cedar Bog Lake, all the Pg (gross production) efficiencies presented in Table III were calculated on the basis of gross oxygen production values obtained from diurnal oxygen curves. For these studies the range of gross oxygen pro- duction values reported by each author were also included in Table III. The upper range of Pg/Li, Pg/La and gross oxygen production values recorded for the Lake City ponds are generally below those reported by the several other authors in Table III. However, for each of these same three para- meters the Lake City ponds exhibited the lowest values recorded in these various aquatic communities. There are probably two primary reasons why the lowest efficiency and oxygen production values would be observed in the Lake City ponds. First, the sodium arsenite application significantly reduced oxygen production and efficiency levels in the I treated pond far below normal summer levels. Secondly, since both efficiency and oxygen production were calculated on a daily basis the lowest values occurring would be 81 .osmmH .sseo 151 can N$3 .ameom ego aseo Hoe “mmmH .mHnnoa can nommsa Amy “msmH .cnosom rev “NemH .aoEoeaHH Amy 133 .soess Ame “NmmH .cooaHo HHS .omcmu mHMOflGGH menu “5000 mosam> 03p so£3rrmmmmusoouom mm commoumxo mowocoHOHmmms m.o Imo.o mum mwnm mocom aufio mxmq . mm.o No.0 o.m H.o I . mm 0 mm a m OH.v om.m m.~H m.m osona mo com o.mm I o.m mm.o m.m w.H bmmsflumm Hm>HHm o.m~ o.m mEmooonofiz Emouum . I . H.o o>H on o NH v m n.m mu .m Hm om.a om.o o.m H.N H OH.4 om.~ m.~H m.m «soonpm Ho.0HMHpn< mo.o H.o mmxmu mom “some mm.o hm.o mmuoocoz oxmq No.0 Hosom mu0mosswz Asmn\m2\~oisov mH\cm HH\:m oH\mm HH\mm I coauosooum No mmonw «mmflocoflowmmm >DAGSEEOU owumsvm mpsom muflo oxmq o» nonmmfiou moauflcsfisoo oapmsqd mDOHum> Eoum mosam> cofluosoonm cmmwxo mmouw cam mogmflowmmm mo omsmm HHH mqmflfi ARA '— .viU - ..- a. . tu.‘ ' . an O .... db-. on H II 'I 1 u.- .. “4.". ”In 1 in. c (II (I) . I 82 recorded in the Lake City ponds while the actual lowest values may have been missed in the sampling procedure used by these authors. None of the other authors cited in Table III~ utilized automatic dissolved oxygen recording equipment and, thus, they were restricted to a few twenty-four hour sampling periods in their studies. Kevern (1962) made the most frequent diurnal oxygen determinations and he sampled on a weekly basis. No net production values were calculated for the Lake City ponds. Attempts to accurately measure the standing crop of Chara SE: in the Lake City study were unsuccessful, and periphyton production was not converted by bomb calorimetry to obtain net periphyton on a gram- calorie basis. However, using Table III as a guide, the net production efficiencies of the Lake City ponds can be estimated. In three studies recorded in Table III (Lindeman, 1942; Kevern, 1962; and Odum, 1957a) both gross and net production efficiencies were calculated. In each case, the net production efficiencies were about one third the corresponding gross production efficiencies. E. P. Odum (op. cit.) cites several studies that indicated that fifty percent or more of the gross photosynthetic production may be used up by the plants themselves. 83 It is perhaps reasonable to assume, therefore, that net production efficiencies for the Lake City ponds, during the summer period, were from thirty to fifty per cent less than the gross production efficiencies recorded in Table III. The maximum Pn/Li for the Lake City ponds would then fall between .3 and .51 and the maximum Pn/La between 1.7 and 2.5. The inverse curvilinear relationship between effi- ciency and solar energy has been observed by many authors. Recently, Duffer and Dorris (op. cit.) were able to demon- strate this relationship by calculating efficiencies and recording solar energy on an hourly basis. They concluded that, "In general, efficiency calculated from total solar radiation indicates the same relationship as that calculated from total radiation each hour (higher efficiency with low light intensity)." E. P. Odum (op. cit.) theorized that low efficiency is necessary for maximum power output of a biological system, and that more rapid growth per unit time has a greater survival value than maximum efficiency in the use of available energy. Figure 14 illustrates the relationship between solar energy (La) and efficiencies (Pg/La) during the two summer study periods at Lake City. The period immediately following the sodium arsenite application in each year was excluded from this curve so that it would more closely represent the relationship between solar energy and 84 Figure 14. The inverse, curvilinear relationship between solar energy and efficiency in the Lake City ponds Kgm—callM’l Day (La) 85 140% IOOOF 600l- .4 2.0 36 Percent Efficiency (Pg/La) Fig.14 v u 3 l c“ 86 efficiencies under normal conditions. The variance of the points plotted in Figure 14 is fairly high; however, the inverse curvilinear relationship between solar energy and efficiency is, nevertheless, clearly illustrated. Periphyton As was mentioned earlier, a number of periphyton samples were dried and weighed to obtain an estimate of the relationship between dry weight and optical density of the chlorophyll content. A portion of the dried samples were then placed in a muffle furnace to obtain an estimate of the ash-free dry weight (organic weight). Only the vertical substrates were used to obtain the relationship between dry weight and optical density. King (op. cit.) and others have found that substrates placed horizontally with respect to the surface usually have significant quantities of heterotrophic and inert material which tends to distort the true relationship be- tween the chlorOphyll content and dry weight of the auto- trophic material. Figure 15 illustrates the dry weigh- Optical density relationship on vertical substrates from both study ponds. The linear relationship between these two measures of organic material is clear, however, if all optical density measurements were to be converted to dry weight measures of periphyton two separate curves would have to 87 The linear relationship between optical density of chlorophyll content and dry weight of periphyton in Pond D and Pond C in 1964 Figure 15. 88 '/ °/ / / . / / 160A // * / . * g // ‘ / / . // . H OIZOI 3. / "I / " / .2? / . E / ‘3 / — / § / * E / 80+ / Pond D (Control) - as // Pond C (Treated) f- . / / / / «x- *- / a: // i ' 40E. 7 ' C .03 .13 Dry Weight .- Grams Fig. 15 .18 be calculate: chlorophyll . than in Pond entirely Cle herbicide tr 51:10:} C-f ti“. 1L Qrg; .-\ fits‘l 9h 35 1. FIE-S 89 be calculated. Figure 15 indicates that in general more chlorophyll was observed for the same dry weight in Pond D than in Pond C. The reason for this difference is not entirely clear, however, it may be a result of the numerous herbicide treatments applied to Pond C. The species compo- sition of the periphyton community in Pond C may have been selectively altered by sodium arsenite. King (op. cit.) found that different zones of the Red Cedar River had different periphyton communities and that dry weight-optical density conversion factors had to be calculated independently for each zone. The ash-free dry or organic weights obtained from the Lake City periphyton samples were extremely variable. Organic weights ranged between 80 and 47 per cent of the dry weights. The organic weights showed little or no relationship to either the dry weights or the optical densities. The only plausible explanation seems to be an inherent error in the procedure used to obtain organic weight. The dry weight of each individual sample was always less than .1 grams and usually less than .05 grams. Since each sample had to be handled several times to ob- tain organic weight, even small errors in either handling or weighing were magnified and became significant because of the small weight of each individual sample. SUMMARY Automatic analyzing equipment was utilized to measure the oxygen metabolism of two small ponds, one of which was treated with sodium arsenite. Other measures of primary productivity were also made to determine the effects of the herbicide treatment. A system was developed to automatically monitor and record dissolved oxygen and pH in the two study ponds inside the laboratory. A computer program was designed to calculate the gross oxygen production, community respira- tion, and the production to respiration ratios from the hourly dissolved oxygen readings. The pH in the treated pond (Pond C) in the pre- liminary study (1963) drOpped from a maximum on the day of treatment of 8.65 to a low of 7.1 one week after the herbicide was added. Adjacent control ponds had a range of pH 8.4 to 9.8 during the same period. It was approxi- mately one month after treatment before the pH returned to normal levels in Pond C. Pretreatment total alkalinity in Pond C in 1963, ranged below 70 mg/l and a maximum of 106 mg/l was recorded two days after the addition of the herbicide. Adjacent 90 91 ponds ranged below 70 mg/l total alkalinity throughout the study period. The total alkalinity appeared to return to normal levels in the treated pond in three weeks. Oxygen production, community respiration and pro- duction to respiration ratios were all significantly reduced in Pond C following the 1963 sodium arsenite treat- ment. Both oxygen production and respiration reached a summer low of 1.0 gm-OZ/mZ/day just after the treatment. The lowest P/R ratio (0.4) was recorded in Pond C on the day of treatment. Dissolved oxygen in Pond C drOpped from 8.5 mg/l to a summer low of 4.8 mg/l within 72 hours after the arsenite was added. The dissolved oxygen remained relatively constant in the adjacent control ponds during the study period. Pond C was again treated with sodium arsenite in 1964, and the changes in both pH and alkalinity were similar to those observed in the 1963 preliminary study. The effects of the sodium arsenite on the produc- tion of periphyton was also measured in 1964. The peri- phyton production in Pond C, as measured by the artificial substrate method, increased significantly after treatment when compared to the control pond (Pond D). However, the periphyton colonization of new substrates appeared to be inhibited by the arsenite application. Samples of the standing crop of phytoplankton in Pond C and Pond D in 1964, indicated that phytoplankton 92 production was temporarily restricted by the herbicide. Spirogyra sp. blooms occurred in the treated pond about ten days after the sodium arsenite was added and lasted about a week. The pretreatment oxygen production values in Pond C ranged from 2.3 to 9.5 gm-Oz/mz/day. Within one day after treatment, the gross oxygen production in Pond C was reduced to 1.6 gm-Oz/mz/day. Efficiency in Pond C was reduced to a summer low of 0.7 the day following the treatment. While oxygen production and respiration re- covered to pretreatment levels quite rapidly, the effi- ciency and P/R ratios were significantly reduced for the entire study period. Results of the Lake City study compare very favorably with the results of various other authors using similar methods, particularly with respect to oxygen metabolism and efficiency calculations. The inverse curvilinear relationship between light energy and primary efficiency could be illustrated with the Lake City data. And, the straight line relationship between chlorophyll content and dry weight of periphyton samples was confirmed by the Lake City data. LITERATURE CITED Ambuhl, H. 1955 Die praktische Anwendung der elek trochei— schen Sauerstoff bestimmung is Wasser. Zeit. Fur Hydrol., 17:123-155. American Public Health Association, American Water Works Association, and Water Pollution Control Federation. 1960 Standard methods for the examination of water and waste water, 11th ed. Boyd Printing Inc., Albany, N.Y., 626 pp. Armstrong, N. E. 1963 Productivity measurements by pH, M2 UTEX PRBYPH. University of Texas Comp. Center (Mimeo), 12 pp. Bails, J. D. and R. C. Ball 1966 Response of pond metabolism to sodium arsenite. Papers Mich. Acad. Sci., LI:193-208. Ball, R. C. and F. F. Hooper 1965 Use of As-74 tagged sodium arsenite in a study of effects of a herbicide on pond ecology. Mich. Dept. Cons., Res. and Dev. Rpt. No. 43, 30 pp. Bartsch, Alfred F. 1959 Algae in relation to oxidation processes in natural waters. Water Supply and Water Pollution Research, R. A. Taft, San. Eng. Cen., U.S. Pub. Health Ser., 1:57-70. Briggs, R., G. O. Dyke and G. Knowles 1959 Electrical recorder for dissolved oxygen. The Water and Waste Treatment Journ., Jan.-Feb., 1 pp. Butler, John L. 1964 Interaction of effects by environmental factors on primary productivity in ponds and microecosystems. Ph.D. Thesis, Okla. State Univ., Stillwater, Okla., 88 pp. 93 94 Clark, L. C., R. Wolf, D. Granger, and A. Taylor 1959 Continuous recording a blood oxygen ten- sions by polarography. Journ. of Applied Physiology: 6:189-193. Copeland, B. J. 1963 Oxygen relationships in oil refinery effluent holding ponds. Ph.D. Thesis, Okla. State Univ., Stillwater, Okla., 110 pp. , J. L. Butler and W. L. Shelton 1961 Photosynthetic productivity in a small pond. Proc. Okla. Acad. Sci., 42:22-26. and W. R. Whitworth 1963 Oxygen metabolism of four Oklahoma farm ponds. Publ. Inst. of Marine Sci., 9:157-161. and T. C. Dorris 1964 Community metabolism in ecosystems receiving oil refinery effluents. Limnol. and Oceanog., 9:131-147. and W. R. Duffer 1964 Use of a clear plastic dome to measure gas diffusion rates in natural waters. Limnol. and Oceanog., 9:494-499. Crabb, G. A. Jr. 1950 Solar radiation investigations in Michigan. Mich. State Univ., Agricultural Exper. Stat., Tech. Bull. 222, 153 pp. Dineen, C. F. 1953 An ecological study of a Minnesota pond. Amer. Midl. Nat., 50:349-376. Duffer and Dorris 1966 Primary productivity in a southern great plains stream. Limnol. and Oceanog., 11:143-151. Gameson, A. L. H. and D. Griffith 1959 Six months' oxygen records for a polluted stream. The Water and Waste Treatment J. 4 pages. Jan.-Feb. 95 Grzenda, A. R. and M. L. Brehmer. 1960 A quantitative method for measurement of stream periphyton. Limnol. and Oceanog., 5:191-194. Juday, C. 1940 The annual energy budget of an inland lake. Ecology 21:438-450. Juday, C., J. M. Blair and E. F. Wilda. 1943 The photosynthetic activities of the aquatic plants of Little John Lake, Vilas County, Wisconsin. Amer. Midl. Nat., 30:426-446. Kevern, N. R. 1962 Primary productivity and energy relation- ship in artificial streams. Ph.D. Thesis, Michigan State Univ. Lib., 132 pp. King, D. L. 1964 An ecological and pollution-related study of a warm water stream. Ph.D. Thesis, Michigan State Univ. Lib., 154 pp. Knight, A., R. C. Ball and F. F. Hooper. 1962 Some estimates of primary production rates on Michigan ponds. Mich. Acad. of Sci., XLVII:219-231. Knudson, V. and T. C. Dorris. 1963 Effects of environment on plankton species diversity in central Oklahoma farm ponds. Proc. Okla. Acad. Sci., 1963, 52-55. Li, Jerome C. 1957 Introduction to statistical inference. Edwards Brothers, Inc., Ann Arbor, Michigan, 568 pp. Lindeman, R. L. 1942 The trophic-dynamic aspect of ecology. Ecology. 23:399-418. Lynn, W. R. and D. A. Okum. 1955. Experience with solid platinum electrodes in the determination of dissolved oxygen. I. North Carolina studies. Sewage and Industrial Wastes, 27:4-9. 9 -fl. .- | ..-AV1 v-‘v ¢ofinn ("H-3. "L 96 Macklin, M. 0., D. J. Baumgartner and M. B. Ettinger. 1957 Performance test of continuous recording dissolved oxygen analyzer. Sew. and Ind. Wastes. Mackenthum, K. M., et a1. 1964 U.S. Department of Health, Education and Welfare, Pub. Health Serv. Publ. 1167, 1:176. Minter, K. W. and B. J. Copeland. 1962 Oxygen relationships in Lake Wooster, Kansas during wintertime conditions. Kan. Acad. of Sci., 65:932-939. , B. J. Copeland and Troy C. Dorris. 1964 Chlorophyll A and suspended organic matter in oil effluent holding ponds. Limnol. and Oceanog., 9:500-506. Newcombe, C. L. 1950 A quantitative study of attachment materials in Sodon Lake, Michigan. Ecology: 31:205-210. Odum, E. P. 1959 Fundamentals of ecology, 2nd ed. W. B. Saunders Co., 546 pp. Odum, H. T. 1956 Primary production of flowing waters. Limnol. and Oceanog., 1:104-117. 1957 Primary production measurements in eleven Florida springs and a marine turtle-grass community. Limnol. and Oceanog., 2:85-97. 1957a Trophic structure and productivity of Silver Springs, Florida. Ecol. Monographs, 27:55-112. and C. M. Hoskin. 1957 Metabolism of a laboratory stream micro- cosm. Inst. of Marine Sci., 4:115-133. and C. M. Hoskin. 1958 Comparative study on the metabolism of marine waters. Publ. Inst. Marine Sci., 5:16-46. 97 and P. R. Burkholder and J. Rivero. 1959 Measurements of productivity of turtle- grass flats reefs, and the Bahia Forfecente of southern Puerto Rico. Inst. Marine Sci., 6:159-170. and R. F. Wilson. 1962 Further studies on reaeration and metabolism of Texas bays, 1958-60. Inst. Marine Sci., 8:23-55. Ohle, W. 1953 Die chemische und die electrchemische bes- timmung des molekar gelosten Sauer stoffer der Binnengewasser. M. Heil. Int. Assoc. Theret. and Appl. Pamatat, M. M. 1965 A continuous flow apparatus for measuring metabolism of benthic communities. Limnol. and Oceanog., 10:486. Prescott, G. W. 1951 Algae of the western Great Lakes area. Wm. C. Brown Publ., Dubuque, Iowa, 977 pp. Reid, G. K. 1961 Ecology of inland waters and estuaries. Reinhold Publ. Corp., N. Y., N. Y., 375 pp. Riley, G. A. 1940 Limnological studies in Connecticut. III. The Plankton of Linsley Pond. Ecol. Monogr., 10:280-306. Ruttner, Franz. 1953 Fundamentals of limnology. University of Toronto Press, Toronto, 242 pp. Ryther, J. H. 1956 The measurement of primary production. Limnology and Oceanography, 1:72-84. Sneed, K. E. and H. K. Dupree. 1962 An electrical oxygen-temperature meter for fishery biologists. S. S. P. - Fisheries No. 426, Fish and Wildlife Ser., U.S. Dept. Int., Washington, D.C., 13 pp. Strickland, 1960 Sohacki, L. Welch, Welch, Welch, 1965 E. B. 1959 P. S. 1948 P. S. 1952 S. P. 98 D. H. Measuring the production of marine photoplankton. Fish. Res. Ed. Canada, Bull. No. 122, 172 pp. Ecological alterations produced by the treatment of pond ecosystems with copper sulfate and sodium arsenite. Master's Thesis, Michigan State University Lib., 91 pp. The predator-prey relationships between a fish population and its macro-invertebrate food supply. Master's Thesis, Michigan State Univ. Lib., 112 pp. Limnological methods. Blakiston, Philadelphia, Penn., 381 pp. Limnology. McGraw Hill Co., Inc., N. Y., N. Y., 538 pp. APPENDIX 99 d l q IRA in Y “ny0 ' e ob Hi .a.ov& e E s u - a Nil 9..» 9 AIV \. ,s sly 3 Ian -\J 6 art! DJ 1 l l . .s II t or 6 . t I.- Q a l 100 Record of solar energy, oxygen production, respiration and efficiency in Pond C and Pond D in 1963 and 1964. following abbreviations were used in the tables: Total light energy in gm-cal/cmz/day The L = La = Absorbed light in kgm-cal/mZ/day Max. D.O. = Maximum observed dissolved oxygen in mg/l Min. D.O. = Minimum observed dissolved oxygen in mg/l Pg = Gross production in gm-Oz/mZ/day Rt = Community respiration in gm-Oz/mz/day Pg/Rt = Production-respiration ratio Pg/La = Efficiency in per cent Pond C Max. Min. Date L La D.O. D.O. Pg Rt Pg/Rt Pg/La 7- 7-63 507 1015 8.4 5.7 .1 2.0 .05 .04 8 727 1454 5.8 5.3 1.1 1.3 .81 .30 9 -- -- 5.7 5.0 1.4 1.7 .81 -- 10 -- -- 5.9 5.0 1.2 1.1 1.11 -- 11 649 1299 6 0 5.2 1.1 1.7 .62 .33 12 -- -- 7.0 5.4 2.1 1.0 2.05 -- 13 -- -- 7.6 6.1 1.7 1.3 1.29 -- 14 156 313 7.9 6.7 2.4 2.8 .85 3.01 15 346 691 9.0 7.0 2.5 1.3 1.87 1.44 16 573 1145 9.8 8.2 3.0 2.7 1.11 1.04 17 322 643 8.8 8.2 l 3 2.2 .58 .81 Max . Min . Date L La 0.0. 0.0. Pg Rt Pg/Rt Pg/La 7-18-63 606 1211 8.7 7.3 3.1 4.3 .71 1.01 19 474 948 7.9 6.2 3.7 4.0 .90 1.54 20 550 1101 8.8 6.4 1.9 1.7 1.25 .64 21 545 1191 9.1 6.6 3.2 2.1 1.50 1.09 22 504 1008 9.6 6.2 6.1 5.8 1.05 2.42 23 621 1241 10.1 7.0 5.3 4.1 1.28 1.70 24 600 1200 10.8 7.6 7.0 7.2 .97 2.33 725 626 1252 10.8 9.2 4.1 4.3 .94 1.31 26 567 1134 11.0 8.0 5.7 6.8 .83 2.01 27 474 948 9.2 8.2 2.0 2.2 .91 .85 28 381 762 8.5 7.5 1.7 2.3 .72 .88 29 636 1272 9.2 7.1 3.6 3.3 1.08 1.13 30 608 1216 10.5 7.7 5.3 4.8 1.11 1.76 31 173 346 8.2 7.8 1.7 3.6 .46 1.92 8- 1-63 558 1116 8.9 6.7 3.3 2.6 1.24 1.17 2 235 470 8.4 7.5 1.6 2.3 .71 1.39 3 402 804 8.2 7.0 1.7 2.5 .69 .86 643 1286 8.3 6.6 2.7 2.7 1.01 .85 5 i634 1268 9.3 7.0 3.4 5.4 .64 1.09 6 450 900 -- -- -- -- -- -- 7 415 830 8.5 6.7 3.1 3.0 1.06 1.07 8 560 1120 10.5 7.1 5.3 5.3 1.02 1.02 9 517 1034 9.3 7.7 3.7 4.4 .83 .83 Max. Min. Date L La D.O. D.O. Pg Rt Pg/Rt Pg/La 8-10-63 577 1154 9.4 7.3 3.9 5.3 .74 .74 7- 9-64 634 1268 8.6 6.9 2.9 2.5 1.16 .90 10 711 1423 10.2 6.9 6.4 5.5 1.16 1.80 11 341 681 9.8 7.8 4.8 5.2 .91 2.81 12 317 633 8.5 8.8 2.6 4.0 .65 1.64 13 97 194 8.2 7.1 2.6 3.1 .84 5.34 14 338 677 10.1 6.9 2.6 1.6 1.65 1.56 15 698 1396 11.4 8.6 3.0 1.2 2.48 .85 16 692 1384 10.6 8.6 3.0 2.9 1.04 .88 17 616 1233 10.4 8.6 5.4 6.6 .88 1.77 18 509 1018 9.6 8.4 3.7 4.7 .76 1.40 19 622 1244 9.8 7.8 4.7 5.1 .93 1.52 20 440 881 8.4 6.0 1.6 4.0 .41 .74 21 508 1016 7.0 4.0 2.8 2.3 1.23 1.10 22 594 1188 7.5 5.4 5.0 5.1 .97 1.67 23 554 1107 7.2 5.7 3.2 4.3 .75 1.17 24 555 1109 7.9 5.5 3.1 1.7 1.79 1.12 25 452 904 7.9 6.4 4.1 5.5 .85 2.05 26 566 1132 8.3 6.1 4.6 4.5 1.02 1.64 27 479 959 7.9 6.4 4.6 5.1 .91 1.93 28 508 1015 8.1 6.4 2.9 2.5 1.14 1.14 29 632 1264 8.1 6.3 3.5 3.7 .93 1.10 30 -- -- 8.9 6.8 3.6 2.9 1.22 -- 31 -- -- 8.7 7.4 2.2 2.5 .88 -- 103 Max. Min. Date L La D.O. D.O. Pg Rt Pg/Rt Pg/La 8- 1-64 -- -- 8.5 7.3 1.1 .9 1.23 -- 2 564 1128 10.4 7.2 4.5 3.1 1.44 1.60 3 559 1117 11.1 7.3 5.8 5.9 .99 2.09 4 638 1277 10.4 8.3 5.3 6.8 .77 1.66 5 635 1270 10.6 7.7 4.8 4.6 1.05 1.43 6 563 1127 10.1 8.1 3.7 3.8 .95 1.30 7 588 1176 9.1 8.0 -- 4.1 -- .58 8 325 650 8.3 6.9 2.5 3.7 .67 1.53 9 599 1199 9.7 7.1~ 3.6 3.0 1.21 1.21 10 395 791 10.0 7.9 3.6 3.7 .97 1.82 104 Pond D Date L La Pg Rt Pg/La 7-14-64 338 677 1.2 1.7 .73 15 698 1396 1.3 1.5 .39 16 692 1384 2.0 2.0 .57 17 616 1233 1.9 1.1 .31 18 509 1018 2.9 3.0 1.10 19 622 1244 2.2 1.0 .75 20 440 881 1.5 3.0 .72 21 508 1016 2.3 2.0 1.02 22 594 1188 2.9 2.8 .98 23 554 1107 2.6 2.7 .94 24 555 1109 2.7 2.8 .97 25 Y 452 904 2.5 2.7 1.10 26 566 1132 2.8 2.5 .98 27 479 959 2.5 2.6 1.04 28 508 1015 2.4 2.0 .94 105 Fortran computer program developed to calculate average respiration, gross oxygen production, and total community respiration for each day from observed hourly dissolved oxygen readings. PROGRAM BAILS DIMENSION DAYIZSIoADJDAYIZSI PRINY 11 12 READ loDATEoNleZoIDAYIIIoI81025I IFIEOFpGOI999oQ98 SU"1.000 DO 2 I'ZILI POINTl'DAVIII’DAVIIfilI IFIPDINTlISoSoZ 3 SUNl'SUH1*POIN71 2 CONTINUE L2!N2*1 SUH280.0 DO 4 I'L2125 POINTZ'DAYIII'DAYII-i) IFIPDINTZ’SISOQ 5 SUNZISUH2*POINT2 4 CONTINUE AREA'SUH1*SUH2 P0!NTS:(L1-1)¢¢25-N2) AVENITElIAREA/POINYS AVENITE'ABSFIAVENITElI DO 6 I3N13N2 POINT3'DAYIII'DAYIIH13 AOJOAYIIIIPOINYS‘AVENITE 6 CONTINUE LJaN2~1 $UH330,0 DO 7 I3N19L3 IFIADJDAYIIII74805 D IFIADJDAYII*1II7O909 9 TRAPElOoS'IADJDAYIII*ADJDAYII*1II SUHD'SUHSOTRAPE 7 CONTINUE VOYALIAVENIIE‘246 PRINY'ionOATEnAVENITEaSUHiaTOTAL GO TO 12 1 FORMAT,(LDnZIZc17F462/1210DF4023 IO FORMAT (1X. A801OX071D' 5021X071005021X071005’ 11 FORMAT (lflia2X000A750012X49LVERAOE. OXYGEN-USEO AT Nleflftalnxa'PPM lOXYGENIOAYIpIOXo'TOTAL OXYGEN USED IN 24 HOURSt/ZX) 999 20.NTINUE ND l()6 Sample Data Cards flint UUCUUbJ-UUJIHLJEILJ330545E52U__E____UJJ%ULJCLUJ.UUJ_bUJaLLtUL__jE II lDate Imul I2I3I «III-IoIIoInIIzIIIzlaI”4 O Pond D.0.l>Hrs Noon I|al|aolloI||I|||10||oI||00||III|III|ou||aa||oollualloal|oo||aulloo||ollluallall Iaaosorlonuuuuunnunuannuaunnaannnunanxuounuuoauunauusuamnuamuuuuuuuuamnnnunnnnnm I|I||IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 12|2222222|2222222222zzzzzzzzzzzzzzlzzz222|2222222222222222222222222222222I22222 aaaaaaalaaaaaaaa33333333331aaaaaaaaaaalaaaaaa3|:3333a3333333333aaaazaaaaaaazaaaa IIIIIIII44444444IIIIIIIIIIIIIIIIIIIII44444444444IIIII444444444444444444444444444 ssssssssssssssssslssslssslssslllslssslssslssslssslssslssslssslssslssssssssssssss sassslssslssslsssssssssssslssssscsssssssssssssassssssslsesssssssssssslssclcsslss IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII777IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII auaaanaauaasaasauIIIIIIIIIIIIIIIIIIIIII88301303303140.3410aaaalaaaaaaocanasaaaaa 999999999999999999999999999999999999999999 999 9999999999999999 999999.9999999999999 I234at7|anununnunununnnunnnmnmnnnurxnxaouQua Hnuoaunauaunaau Huuuaunaunnnnnnnnn nm 30" 89C .-..- 5'1 I54“. I_-. I_UL‘; :5:I_II_II'5. EIjIIJ. I_ :. _II _'I; I_ILIIJE ‘EIIQIUr. : E'ILIE-T‘E‘IBE-E-E l l 5 I 6 I 7 I I I 9 | Iol III111I —Same- DODHIS Midnight IIIIIIIIIIIIIIOIIIIIIIIIIIIIIIIIIIIIIIIOIIOOIIOIIIIIIIIIIOIIIOIIIIIIIIIIIOIIIIIO 123450toanuuuuuununnnnnnaanaamnunusanaanuunuoununaunausanaaananuuanumnnnnnnnnnnm I|I||IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII zzlzzzzzzzlzzzzzz2222222222222222222222|222|222222222222222222222222222222222222 aaaaaaalaaaaaaaaasasa3333333a:33333333333a33333333333333333333333333333333333333 IIIIIIIIIIIIIIIII444444444444I44444444444444444444444444444444444444444444444444 ssssssssssssssssssssssslssssssssssslssssssssssssssssssssssssssssssssssssssssssss scstslssclcsslssslssssssssssslssslssslsssllstsssssscsssssssssssssssssssssessssss IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII777117777177771171177171771171777777717 IIIIIIIIIIIIIIIIInassaaaaaaaaaInsulatecansauna:IIIIIIIIIIIIaouaaasacaccccscsaaaa 999999999999 M9999'99999999999'9999999999999999999999999999999999999999999 9999 121050709muIzIIIouununnnnnnnanuannunaaxnanauauuuuuuuuuauuaanaaauunuuauuunnnun» Iznnn DOOI 39C MICHIGAN STATE UNIVERSITY LIB II II II II | I IIIIIIIIIES 3 1293 9306’ 5615