EXPERIMENTAL rmnuunon OF MICHIGAN TROU‘F LAKES Ms flow {‘11. Dogma of Ph. 3-. MICHIGAN STATE COLLEGE Heward Alien Tanner 1952A THESIS) 1 1 . ' '1 _ I I 2. < l w -. . l ‘. — This is to certify that the a l 1, thesis entitled ‘ . 75 Experimental rerulization of 7 ' . " Michigan Trout Lakes l ‘ presented by . .' l x , Howard A. Tanner has been accepted towards fulfillment ; 4" of the requirements for I ‘ " ‘ i " Mdegree mm 37' QMW - 1 Major professor Date May 23, 1952. . I \ \ \ \ ‘ l I J , ' r, t l I! ‘ . 4 . V' l , . I 2 . _' , '54 l w l 1L l‘ ' Kl - w 3 fr)» '3: isn‘t»: ,- ~";"r 2 “in i“ - ‘ ? Wflémm—iwfi ‘~.. :31 6‘ l , I: s: A ”W ‘I‘\‘..,‘., 3-" ‘1‘ {‘5‘ ‘ (WIVN'Wflml r; ' “”7"“ ‘1 \EK v Q; '-. 1 fi ? .;V‘-’ ' .V \v :‘$e§*”"“ ‘ e:- ‘.’l;$*:"< :3")! ‘7‘! ‘ h ‘- imd _- 4—12.} h u... : ——o—_....-_uJ-. AL.— L . . ‘ A. ‘ v - ~ . ' ‘A; ' I . . ‘9' mi??? 34¢”; «i I S 5.9”} 2‘2”: R’..-II-.—-h- .. !hr- ' -‘ g” - ‘fii'flrl u: Vim—3 W'Utn‘3*'-’IT'WA 7’77“"F71*T ,‘KF‘JT 5* =51 “*0, WI“. m'o 'vvm 'T'fi“."'f‘fi .211. .L-Jal.l.4a~4.11l\4.£‘t -‘u-..4._l__JI.J.'1.L.1.Ow 0.. u-«l.J-Ll “'l LLLL’U; LtdL'Q Howard Allen Tanner ‘ new...“ AR ASSTRACT Submitted to the School of Graduate Stu ies of Iich M L‘ 5} go :3 State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PIILOSCPEY Department of 200103 Lear 1952 Approved QM*QB$9~Q Howard Allen Tanner ABSTRACT The data presented are from a program directed towards determining the feasibility of applying fertilizer to un- productive trout lakes and by so doing increase fish pro- duction. The program is the latest of several studies in Michigan all of which.have been directed towards examining the advantages and disadvantages of fertilizing several types of natural and artificial waters. Six trout lakes lying in the generally unproductive pine plains of the north central portion of Michigan's lower peninsula were selected as the cite for the present study. Biological sampling and measurement of certain chemical and thermal features of the environment in 1948 established the level of standing crops of several groups of organisms and the characteristics of each of the lakes prior to ferti- lization. Trash.fish.were removed by poison and brown trout planted in each of the lakes at the same number per acre. Fertilizer was applied to four of the six lakes at varied rates during the summers of 1949 and 1950. Data collections were continued during the period of fertilization and the effects have been evaluated by comparisons on the basis of before and during fertilization observations and between fertilized and unfertilized lakes. Analysis of the data using statistical procedures re-. 2&1712CH3 Howard Allen Tanner vealed that the position of the thermocline became shallower following fertilization and that the total hardness decreased in the fertilized lakes. Oxygen depletion occurred during summer and winter stagnation periods and was closely corre- lated with.amounts of fertilizer added. The biological sampling indicated increases in amount of plankton and bottom fauna organisms. Changes in the com- position of the bottom fauna and in the depth distribution were observed and the degree of dependence of the trout on the bottom fauna for food was studied. A complete creel census indicated the yield of trout to anglers and the total net production per acre was calculated from.the creel census and a population estimate of the trout present at the end of the program. The trout were killed with a toxicant and estimates of the population were made on the basis of mark and recovery (Ball, ms.). It was concluded that trout production could be stimu- lated with small applications of fertilizer. The small amounts of nutrient material used did not result in winter- kill conditions. It was also concluded that the eutrophying effect of additional nutrients (Healer, 1947) makes imprac- ticable the fertilization of trout lakes with the exception of small lakes unproductive of trout because of extreme oligotrophic conditions. fl .[rl‘illl I\’ ll\.’lkl\:ll|ll|ll. ll“) >lll,l> Howard Allen Tanner Ball, Robert C. ---- Standing crop of brown trout based on recovery of marked fish following poisonings of six trout lakes. (In preparation.) Hasler, Arthur D. 1947 Eutrophication of lakes by domestic drainage. Ecol. 28 (4): 385-394. '6‘ stem. .. 5TH u .3 r .4 . _._.. NI .. . . .. . q. Frontispiece Section-Four Lake EXPERIMENTAL FERTILIZATION OF MICHIGAN TROUT LAKES by Howard Allen Tanner A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1952 // +1934 11 ACKNOWLEDGMENTS The writer wishes to express his gratitude to Dr. Robert C. Ball, Department of Zoology, Michigan State Col- lege. His generous cOOperation and stimulating interest while directing the study has been a constant source of inspiration. Appreciation is due The Institute for Fisheries Re- Search of the Michigan Department of Conservation for the doctorate fellowship granted the writer in support of the program. The use of the facilities of the Institute and the cooperation of its staff, the director and staff of the Pigeon River Trout Research Station in particular, have been of great assistance. Dr. Don'W. Hayne, Zoology Department, Michigan State College, has been most helpful with the statistical por- tions of the paper. Special thanks is due my wife, Helen, for her assist- ance throughout the program. Mr. Paul Barrett assisted in portions of the field work. ' 3!?203 Howard Allen Tanner iii candidate for the degree of Doctor of Philosophy Final examination, May 23, 1952, 1:00 P. M., Natural Science Building, Room 404 Dissertation: EXperimental Fertilization of Michigan Trout Lakes Doctorate committee: Chairman - Dr. R. 0. Ball, Zoology, M. S. C. Dr. G. W. Prescott, Botany, M.S.C. Mr. W. F. Morofsky, Entomology, Outline of studies . M. S. C. Major subject: Zoology Minor subjects: Botany, Entomology Biographical items Born, Sept. 4, 1923, Kalamazoo, Michigan, English decent Undergraduate studies, Western Michigan College, 1941-1943 Michigan State College, 1946-1947 Graduate studies, Michigan State College, 1947-1952 Master of Science degree, June, 1950 Major subject: Zoology Minor subject: Aquatic Entomology Thesis: The Biological Effects of Fertilizer on a Natural Lake Emperience: Graduate assistant Michigan State College, 1948. Member of U. S. Army 1942-1946. Re- search.fellowship from The Institute for Fisheries Research of the Michigan Conser- vation Department in support of masters and doctorate research problems Publication: Ball, Robert C. and Howard A. Tanner 1951 The effects of fertilizer on a warm-water lake. Michigan State College, Tech. Bul. 225. Member of the Society of the Sigma.Xi TABLE OF CONTENTS ACI‘CI‘T OVI'IIEDGII‘IEI‘I TS . V IT A I O O O O O O O O O O O O O O 0 TABLE OF CONTENTS . . . . . . . . . . LIST OF ABLES . . . . . . . . . . . I INTRODUCTION . . . . . . . . . . . II DESCRIPTI H 0? THE STUDY AR EA . . . . III DESCRIPTION OF TIE LAKES . . . . . . IV IETHODS AIII'D ECTTIPI E‘IIT . . . . . . . General Work Pattern . . . . . . . Field Program . . . . . . . . . . Selection and Preparation of Lakes Chemical and Thermal Analysis of the Water Stocking . . . . . . . . . . . . Application of Fertilizer . . Secchi Readin s . . . . . . Bottom Fauna Collections . . . . Fish Sampling . . . . . . . . . Aquatic Vegetation and Filamentous Al Lake lapping . . . . . Laboratory Procedure . . . . . . Bottom Fauna Samples . . . . . StOIIlaCl'l Analyses o c c c c c e V THE EVALUJTION OF THE ETFECTS OF PERT Physical and Chemical Effects of Fertilization ThGI’IIIOCline Chanfse S o c o c c c 0 Oxygen Depletion . . . . . . . . Changes in Total Hardness . . . . £16 ILIZATION iv Page ii iii 10 21 21 22 22 22 23 27 29 50 51 34 55 58 38 59 4O 4O 40 As .5 57 TABLE OF CONTENTS (Continued) Page Biological Effects of Fertilization . . . . . . 65 Plankton . . . . . 64 Filamentous Algae and Higher Aquatic Plants . 75 Bottom Fauna . . . . . . . . . . . . . . . . 76 Stomach Analyses . . . . . . . . . . . . . . 125 Angling Yield . . . . . . . . . . . . . . . . 126 Fish Production . . . . . . . 147 V.I DISCIJSSIOIJ O O O O O O O O O O O O O O O O O O O 171 V I I SLTSvIILKx—FiY . C O O O O O O O O O O O O O O O O O C O l 77 VVIII - LI{EEP TTffl CITED 0 O O O O O O O O O O O C O Q . 180 IX APPEKDIX . . . . . . . . . . . . . . . . . . . . 186 Table number 1. 2. 3. 4. 5. 9. 10. 11. 12. LIST OF TABLES Title Physical and Chemical Characteristics Prior to Fertilization . . . . . . . . . . . . . . 1948 Brown Trout Plantings . . . . . . . {ates of Fertilizer Applications . . . . . . Area Reductions Indicated by Remapping (1950). Average Depth in Feet of the Thermocline Mid-Points . . . . . . . . . . . . . . . . Summer OXygen Levels, Average Depth in Feet Where oxygen Content Fell Below 4 P.P.M. . . Winter Oxygen Levels, Depth in Feet Where Oxygen Content Fell Below 2 P.P.M. . . . . Rates of Reduction in Total Hardness in the Unfertilized and Fertilized Lakes During the Fertilization Period . . . . . . . . . . Average Secchi Disk Readings (Depth in Feet) Invertebrate Fauna Collected by Bottom Sampling South Twin Lake 1948, 1949 and 1950 Invertebrate Fauna Collected by Bottom Sam- pling Section-Four Lake 1948, 1949 and 1950 Invertebrate Fauna Collected by Bottom Sam- pling Lost Lake 1948, 1949 and 1950 . . . . Invertebrate Fauna Collected by Bottom Sam— pling Hemlock Lake 1948, 1949 and 1950 . . . vi Page 27 29 58 47 49 58 78 79 80 81 Table number 14. 15. 17. 18. 19. .20. 21. 22. 25. 24. LIST OF TABLES (Continued) Title Invertebrate Fauna Collected by Bottom Sam- pling West Lost Lake 1948, 1949 and 1950 . . Invertebrate Fauna Collected by Bottom Sam- pling North Twin Lake 1948, 1949 and 1950 . Calculated Average Volumes of Small Invertebrates . . . . . . . . . . . . . . . Average Volume of Trout Food Organisms Pres- ent During Summers of 1948, 1949 and 1950 . South Twin Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . . . . . Section—Four Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . . Lost Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . . . . . . . Hemlock Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . .1. . . West Lost Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . . . . . North Twin Lake Depth Distribution of Trout Food Organisms by Volume . . . . . . . Relative Importance of Invertebrates as Food of Brown Trout Indicated by Stomach Analyses (1950) . . . . . . . . . . . . . . vii Page 82 85 97 99 106 106 107 107 108 108 124 ill'lulr'll"l|llllllll‘ll|tll{[ ll({.[(|l‘l viii LIST OF TABLES (Continued) Table number Title Page 25. Fishing Effort and Success by Lakes . . . . 128 26. Distribution of Fishing Effort and Success by Months (1949) . . . . . . . . . . 129 27. Distribution of Fishing Effort and Success by Lakes (1950 Season) . . . . . . . . . . . 150 28. Distribution of Fishing Effort and Success by Months (1950) . . . . . . . . . . 151 29. Distribution of Percent of Total Catch by Months . . . . . . . . . . . . . . 152 50. Average Length in Inches of the Brown Trout Removed by Angling Before and After Adjust- ment for Time of Removal . . . . . . . . . . 155 51. Hours of Fishing Effort and Percent of Total Hours (1950) . . . . . . . . . . . . . 142 52. Catch per Hour for Each.Week and Rank of th Lake by Fishing Success (1950 Season). . . . 14 55. Variation Due to Regression ("r") of Rank of Lakes by Quality of Catch to Difference in Effort from the Expected . . . . . . . . 145 54. Record of Fishing Yields and Estimates of Standing Fish Crops Prior to Fertilization of 1949 and 1950 . . . . . . . . . . . . . . 148 LIST OF TABLES (Continued) Table number Title Page 55. The Net Yield in Pounds per Acre of Brown Trout During 1949 . . . . . . . . . . . . . . 152 56. Weight and Composition of the Anglers Catch (1950) . . . . . . . . . . . . . . . . . 156 57. Estimated Numbers and Weights of Fish Pres- ent at the Time of the 1950 Poisoning . . . . 158 58. Weight and Length of Brown Trout from 1950 Poisoning . . . . . . . . . . . . . . . . 159 59. Net Weight of Fish Removed by all kethods (1949 and 1950) . . . . . . . . . . . . . . . 160 A Summary of the Statistical Analyses of Changes in the Position of tho Thermocline (Analysis of Variance) . . . . . . . . . . . Appendix B Summary of Analysis of Variance for Summer Oxygen Depletion . . . . . . . . . . Appendix Ca Difference in Average Length of Trout Caught from the Six Lakes During 1950 Season (Analysis of Covariance) . . . . . . Appendix b Derivation of Adjusted Hean Length of Fish from 1950 Fishermen's Catch . . . . . . . . Appendix D Differences in Percent of Effort from Ex— pected Effort in Percent . . . . . . . . . . Appendix LIST OF FIGURES Figure number Title Page Section-Four Lake . . . . . . . . . . . Frontispiece 1. Map of the Problem Area . . . . . . . . . facing 6 2. Cave-in, Section—Four Lake . . . . . . . . 11 5. North shore Section-Four Lake . . . . . . 15 4. Another View of the Typical Precipitous Incline Surrounding Section-Four Lake . . 17 5. Lost Lake Shown Far Below the Level of the Surrounding Terrain . . . . . . . . . l9 6. The Brown Trout (Salmo trutta fario) . . . 24 7. Section-Four Lake Showing the Tangle of Sunken and Floating Logs Present Near the Shore . . . . . . . . . . . . . . . . 52 8. &.9. Shoreline of South Twin Lake before and After 1946-1947 Fertilization Showing the Algal Mats Which Occurred . . . . . . . . 56 10. North Twin and South Twin Lakes. The Vertical Location of the Thermocline and Volume of Water with Less than 4 p.p.m. of Oxygen . . . . . . . . . . . . . . . . 41 11. West Lost and Lost Lakes. The Vertical Location of the Thermocline and Volume of #3 0] Water with Less than 4 p.p.m. of Oxygen . Figure number 12. l5. 14. 15. 16. 17. 18. 19. 20. 21. LIST OF FIGURES (Continued) Title Section-Four and Hemlock Lakes. The Ver- tical Location of the Thermocline and Volume of Water with Less than 4 p.p.m. of Oxygen . . . . . . . . . . . . . . . Depth of Oxygenated Water Remaining (by percent) During Period of Fertilization The Quotient of the Average Total Hard- ness During Fertilization Divided by Averages of Total Hardness Prior to Fer- tilization . . . . . . . . . . . . . . . North Twin and South Twin Lakes,Secchi Readings . . . . . . . . . . . . . . . . West Lost and Lost Lakes,Secchi Readings Section-Four and Hemlock Lakes,Secchi Readings . . . . . . . . . . . . . . . . South Twin Lake, Composition of Bottom Fauna by Volume . . . . . . . . . . . . Section-Four Lake, Composition of Bottom Fauna by Volume . . . . . . . . . . . . Lost Lake, Composition of Bottom Fauna by Volume . . . . . . . . . . . . . . . Hemlock Lake, Composition of Bottom Fauna by Volume . . . . . . . . . . . . Page faCinn 45 51 60 69 71 75 85 87 91 Figure number 22. 25. 25a. 24. 25. 26. 27. 28. 29. 51. LIST OF FIGURES (Continued) Title West Lost Lake, Composition of Bottom Fauna by Volume . . . . . . . . . . . . North Twin Lake, Composition of Bottom Fauna by Volume . . . . . . . . . Midge Egg Masses (Section-Four Lake) . . The Volume of Trout Food Organisms per Square Foot Present in Each Lake (1948, 1949 and 1950) . . . . . . . . . . . . . South Twin Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . Section—Four Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . Lost Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . Hemlock Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . West Lost Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . North Twin Lake, Depth Distribution of Trout Food Organisms by Volume . . . . . The Corrected Average Length of Brown Trout Caught from Each Lake During 1950 xii Page facing 95 ._....._.__.>2 95 97a 101 109 111 115 115 117 119 154 Figure number 52. 55. 54. xiii LIST OF FIGURES (Continued) Title Page Average lengths (each week) of Brown Trout Removed by Angling Throughout the 1950 Season . . . . . . . . . . . . . . . . . . facing 159 Brown Trout from Section-Four Lake . . . . 155 Production of Brown Trout per Acre (net weight) Related to Amounts of Fertilizer Added . . . . . . . . . . . . . 161 INTRODUCTION The energy from the sun and the allogenetic inorganic (and organic nutrients from the surrounding soils provide the basis for the whole complex of biological activity within the lake basin. To the worker desiring to increase the pro- duction and yield from.a lake, Ricker (1946) suggests two methods of approach. Either nutrient material can be added artificially and from the increased productivity greater production and yield may be possible or a more efficient use of the present nutrient supply may be encouraged as, for example, the removal of competing warm-water species from lakes to be managed as trout lakes. Ricker's suggestion of artificial enrichment to in- crease the yield of fish from a body of water is not a new idea. An increase in the yield of one or more species of fish by the addition of nutrient material is made possible by the general productivity of the lake being stimulated, with larger standing crops being formed at each of the tro- phic levels upon which fish production is based. Pond fertilization has been practiced in the orient since early in recorded history (Drew, 1951). Schaperclaus (1955) and Neess (1949), in reviewing practices of pond culture, noted the use of fertilizers in Central Europe several centuries ago. In spite of its antiquity, pond fer-' tilization in this country is relatively new. The develop- ment of farm ponds and the pioneer work with fertilizers began in the southern states during the 1950s (Hogan, 1955; Neehean, 1955; Swingle and Smith, 1959). Fromthis begin- ning, fertilization has spread throughout this country and Canada and applications have been tried on a variety of aquatic habitats, including natural lakes. In Michigan, an eXperimental fertilization program has been underway since 1946. The work has encompassed many types of water and one phase has been the fertilization of natural lakes. Hasler (1947), in discussing the advisabil- ity of fertilizing natural waters, points out the absence of complete information on the effects of fertilization and points out the need for further experhmental work. When the present study was begun in June of 1948, certain informa- tion from completed experiments served to guide the direc- tion of the program. Heavy applications of fertilizer to a warmrwater lake and to a trout lake had created winterkill conditions in both after two summers of fertilization (Ball, 1948; Ball and Tanner, 1950). Smith (1945), fertilizing trout lakes in Nova Scotia, discovered that enriched food supplies favored coarse fish entering through the outlet more than the trout population. Evaluation of creel census reports has emphasized that fishermen do not harvest warm-water species and that in- creasing these fish.by fertilization could not be economi- cally justified. For these reasons landlocked trout lakes in which the rate of harvest has been shown to be high.(Esch- meyer, 1958) were selected. The fertilization rates have been reduced. Some workers have attacked artificial enrichment of natural lakes as unwise (Hasler, 1947 and 1948; Hayes, 1951). Their objections apply particularly to water areas with pop- ulations of salmonid or corrigonid fishes with their need for cold, well oxygenated water. These objections are, there- fore, of particular importance in evaluating the present study. Hayes (1951) discussed the subject in considerable de- tail and opposed the use of fertilizers on the basis of three major points: (1) the addition of nutrients to a lake speeds the process of eutrophication and extinction; (2) the eutrophication of the lake favors the coarse and less desirable species of fish.and results in the elimination of trout; and (5) there is little or no carry over of the added nutrients necessitating a continual and, hence, uneconomic fertilization program. Hasler (1947) reviews the conse- quences of inadvertent enrichment of lakes in EurOpe with the accompanying disasterous results to the trout and white- fishes and concludes that a continuing addition of nutrients to trout lakes would have shmilar results. It is true that unless these objections can be met fertilization of trout waters must be considered undesirable. However, the pos- sible rewards warrant investigations to determine if the ob- Jections which have been raised can be circumvented in a practical and economical manner. The present study has been directed toward examining the effects of the fertilizer on biological productivity and production and providing additional information concerning the effects of fertilizer on certain of the physical and chemical factors of the environment. While other studies have shown general increases in productivity following ap- plication of fertilizer, the effects on the various trophic levels and on the chemical and physical environmental fac- tors have remained unpredictable. The accurate prediction of fertilization results will become possible only with the complete understanding of the complex trophic-dynamic re- lationships as they exist in our natural waters. ‘With this in mind, the present study was designed to measure not only changes in biological production but also the secondary ef- fects on the chemical and physical features of the environ- ment and, in so doing, provide certain of the information necessary to predict the total effects of fertilization on natural waters. The study has been under the direction of Dr. Robert 0. Ball of Michigan State College and has been supported by The Institute for Fisheries Research of the Michigan Conserva- tion Department with a doctorate fellowship. DESCRIPTION OF THE STUDY AREA A group of six trout lakes lying in the Pigeon River State Forest was selected for use in the present study. These lakes, generally known as the Pigeon River "Pot-Hole" Lakes, were selected because of their small area, close proximity to one another, location on state owned lands, and general similarity. Two of the lakes had been used in a fertilization experiment by Ball (1948). Five of the six are in Otsego County (North Twin, South Twin, West Lost, Lost, and Section-Four Lakes) and Hemlock Lake is just over the line in Cheboygan County. Prior to the second summer of field work the immediate area in which the lakes were located including five miles of the Pigeon River was designated as the Pigeon River Trout Research Area and placed under the administration of The Institute for Fisheries Research. The establishment of the Research.Area made possible a compulsory creel census which aided greatly in the determination of fish production in the lakes. Figure 1 indicates the location of the area and the close proximity of the lakes to one another. The soil and drainage are typical of the so-called pine plains of the north central lower peninsula. These plains are, properly speaking, a central upland, with a very acid sandy soil of low inherent fertility with a glaciated topog- raphy. The area is drained by the Pigeon River which runs .dons moan» on» we soapseoa on» moaned nomad hpdsoo use awash .mosaaasoa .aonpo nose on was nopam doowdm on» on scandaon a“ momma on» no doapamom wcfipeoaosa some hogan can no Adz .H 98mg northward through Mullet Lake to the Cheboygan River and into Lake Huron. The original forest cover was of white and red pine with isolated hardwood stands consisting largely of beech-maple and hemlock and which, in general, occupied a less sandy soil type. Hemlock Lake lies in one of these iso- lated hardwood stands where the soil type is an undulating phase of the Emmet-Nester association and the surface a sandy loam to fine sandy loam with a substrata of reddish sandy clay loam with occasional gravel pockets (55 percent Emmet sandy loam, 25 percent nester sandy loam). This soil is classed as moisture retentive. The soil types in which the other five lakes lie (Roselawn and Rubion sand) are considered to have an extremely rapid drainage and are the types occupied by the original pine forest. These soils have a surface of sand and a substrata of brownsih yellow sand with an occasional gravel pocket. Following the logging of the pine near the close of the last century the region was repeatedly burned by fire. The establishment of adequate forest fire control in the early 19503 has permitted fast growing aspen, popular, fire cherry, and shad'bush to become established. These transient species are in various stages of being replaced by jack, red, and in some areas white pine as a result of natural and artificial seeding. Small areas of white and red pines survived the fires either by chance or because of a protected location. The steep slopes of the lake basins provided a measure of protection for the trees and today some portions of each lake basin are covered by large pines providing erosion control and adding to the beauty of the lakes. 10 DESCRIPTION OF THE LAKES The unusual physical characteristics of these six lakes with their nearly symmetrical outline, the water surface 40 to 60 feet below the surrounding terrain, and the steepness of the slopes which surround each lake, with the exception of Hemlock, have created considerable interest in their origin. In earlier investigations, Eschmeyer (1958) pointed out the similarity of these lakes to the glacial-pit type of lake as described by Scott (1921). In the intervening years, geologists of the state have become more familiar with the region and the lakes and are now in agreement that the lakes are limestone sinks, Bergquist (personal communication). The evidence that supports this contention is that the area is underlain by limestone formations and the presence of many other more typical limestone sinks both with and with- out water in the area. Dramatic proof of the correctness of this conclusion was presented when, in late May of 1950, the west shore of Section-Four Lake collapsed (Figure 2). Soundings of the lake indicated that the collapse of the bank was due to the settling of part of the lake bottom. It was concluded that further erosion of the limestone forma- tion had occurred. The general similarities of the salient physical and chemical characteristics have been tabulated (Table 1). Certain important differences existed and should be noted. 11 .OmmH .aoz ea ooandooo moans sano>so on» wdakomn exam Anomadoapoom no enema poo: one .m oAdem 15 .psomoad Haws no mangosw Hasam..w .mmma on noahm oodam Beam commasm on concaosfi moaooom .mmma ode so dsomm ma pH was poapdo do own was haaoanom xoOHSmm..n .mSoShdodhm coaoofimdoo was moon anaconda odd Hwoam..m .ooaooaossa HH¢..H ohms o.e me-ma s.mn mod Hosmv Hams Hows Hm ms o.m osomneofipoom use eooasafl m.e onuma H.4H sea pooo peso Hoosv as so m.m naooasom w poem . osso n.n no-ma o.sH ssa answv Haws Hows He mm m.n pooh use ones o.HH nauma s.nm mad psoo epsoo ones mm as s.n noon poo; oopasaa o.na omuma H.ma m psoo pooo econ mm as >.w ease spooz opam n.ma onuma s.ma as pooh pooh scam em en m.n sage spoon adopxo soap comm odHHo anooom mmos gamma HwaOppHH Hmonm gpdoo npmoo Amoaomv momma owwao>¢ .SSEa macaw H :Nwz commafim nwpomo> cassava pace noaaone them -ooos Hsooe unsm omaw HHom soppom 1% ZOHB¢NHQHBmmm OB MOHmm mOHBmHmmBo .¢ onzmdm Figure 5 Lost Lake shown far below the level of the surrounding terrain. This lake occupies the deepest depression of any of the group. 21 METHODS AND EQUIPKENT General Work Pattern The summer of 1948 was spent in collection of preferti- lization data and the poisoning of those lakes which had trash fish. Fertilizer was applied to four of the lakes during the summers of 1949 and 1950. The remaining two were kept as controls. Data were collected to facilitate comparison of physi- cal, chemical and biological conditions before and during a period of fertilization. At the end of the summer of 1950 the lakes were again poisoned to obtain population estimates of the trout present. Investigation of the problem of winterkill necessitated chemical examination of the lakes each winter. During the period adjudged to be the most critical, the lakes were visited and complete chemical series obtained. The creel census station established just prior to the 1949 trout season collected a complete record of all trout removed from the lakes by angling. The checking station supplied a record of the daily weather which was of consid- erable value in evaluating the results of the program. The files of The Institute for Fisheries Research also were available to the writer. A winter trip into the lakes during February of 1951 was the last date of collections. [U 10 Field Program Selection and Preparation of Lakes The six lakes selected for the program were examined early in the summer of 1948 to determine whether or not un- desirable populations of warm-water fish were present. Application of a fish toxicant to North Twin, South Twin, West Lost, and Section—Four Lakes was decided upon in order to eliminate large populations of yellow perch (Perca fla- vescens) and common suckers (Catostomus c. commersonnii). In West Lost Lake, the pumpkinseed sunfish (Lepomis gibbosus) was present in addition to perch and suckers. Hemlock and Lost Lakes were apparently populated by only a few brook trout and limited populations of forage fish, so poisoning was considered unnecessary. Known numbers of marked brook trout were released in each lake; in West Lost Lake known numbers of pumpkinseeds were caught, marked, and released. The percent return of these marked fish following poisoning was utilized to esti- mate the standing crop of fish (Ball, Ms.). Chemical and Thermal Analysis of the Water Using a Negretti and Zambra reversing thermometer, tem- perature readings were taken at weekly intervals throughout the three summers of the field investigations. Readings were taken from top to bottom in order to establish the position 25 of the thermocline. After locating the thermocline, a chem- ical analysis of the water was determined from three verti- cal locations: chemical analyses were made at the mid-epi- limnion, the mid-thermocline, and the mid—hypolimnion. All water samples were collected with a modified Kemmerer sam- pler. The rapid, unmodified Winkler method was used for oxygen determinations. Determinations of hardness and free carbon dioxide were made using methods outlined in Standard Methods for Examination 93 Water and Sewage (1946). The pH was determined using a Hellige colorimeter. The objective of using the three vertical stations was to record changes in the chemistry of the three water layers. Additional samples were taken at three foot intervals to determine at what point the levels of dissolved oxygen fell below 4 p.p.m. It was deemed of particular importance to determine if oxygen supplies in the thermocline during the mid-summer stagnation period were depleted below the levels necessary for trout survival. Often the thermocline at this period is the only region of the lake suitable for trout from the standpoint of temperature and oxygen. Stocking After removal of trash fish populations from four of the lakes in 1948, brown trout (Salmo trutta fario, Figure 6) were stocked in all six lakes. The reason for using brown trout was that brook trout (Salvelinus f; fontinalis), the trout 24c 26 most commonly stocked in Michigan lakes which are to be man- aged exclusively as trout lakes, were considered to be too vulnerable to angling. It was highly desirable that fish planted in 1948 be present in fair numbers at the end of the program to permit the evaluation of their growth under con- ditions of fertilization. The brown trout were planted as yearlings and at rates that were considered to be excessive so that a sufficient population might be present to exploit fully any increase in fish food resulting from the added nutrients. Through a misunderstanding, 500 brook trout fin- gerlings per acre were planted in South Twin, North Twin, West Lost, and Lost Lakes in addition to and at the same rate as the brown trout and at about the same time. These trout averaged 5.5 to 4 inches. Numbers, average lengths, and average weights of the brown trout planted in each lake in the fall of 1948 are presented in Table 2. In the fall of 1949, following the closing of the fish- ing season, the lakes were again stocked with brown trout. These trout had the left pectoral fin clipped making them distinguishable from the 1948 plant. Since populations of trout were present in all the lakes, the rate of stocking was cut from 500 per acre, the rate used in 1948, to 250 per acre. Trout planted in 1949 were larger than those planted in 1948. Their average length was 7.0 inches and they aver- aged 56.7 grams in weight as compared to a 5.65 inch and 54.4 gram average for the 1948 plant. 27 TABLE 2 1948 BROWN TROUT PLANTINGS Average Average Total No. fish length weight weight Lakes planted (inches) (grams) (pounds) South Twin 2,150 5.64 54.43 163 North Twin 2,850 " " 217 Lost 2,500 " " 175 West Lost 2,000 " " 152 Section-Four 1,550 A " 125 Hemlock 2,600 A " 178 Totals 15,550 1,028 Application of Fertilizer During 1946 to 1947 Dr. R. 0. Ball, in fertilizing South Twin In 1946 he to produce 600 pounds 1947 for a 5.9 acres. Lake, fertilized five times during each summer. used 400 pounds for each application and, failing more than a moderate increase in phytoplankton, per application were used during the summer of total of 5,000 pounds on a lake with an area of Winterkill (Ball, 1948) occurred during the win- ter of 1947-48 which could be directly attributed to the addition of fertilizer. Taking his results as a starting point, it was decided to fertilize four of the six trout lakes at reduced rates. As originally planned, Section-Four 28 was to receive 80 percent, Lost 60, Hemlock 40, and West Lost 20 percent as much fertilizer as did South Twin Lake in 1946-47. Several events occurred to modify these percen- tages: (l) a remapping of the lakes after they had been fertilized at rates based on the old areas, (2) a halt for fear of winterkill after only three applications produced spectacular blooms each summer, and (5) a change of ferti- lizer formula from one of a 10-6-4 N-P-K to 6-10-4 N-P-K dictated by availability. In general, these changes served to reduce the rates of application below the intended per- centages. The rates at which each lake was fertilized have been calculated on the basis of p.p.m. by weight (Table 5). The fertilizer was placed in a tub on the rear seat of the boat and distributed over the surface of the shoal areas with a small dipper. The four lakes were fertilized at different rates in order to determine if a rate of application low enough to avoid winterkill is enough to stimulate biological activity sufficiently to be reflected in increased growth of trout. The greatest amount of fertilizer was added to the lake with the greatest total hardness, Section-Four Lake, and the softer lakes received fertilizer at lower rates. 29 TABLE 5 RATES OF FERTILIZER APPLICATIONS Total* Actual Planned Lakes Pounds p.p.m. percent percent 1946 1947 South Twin 2,000 5,000 28.5 100. 100 1949 1950 Section-Four 1,700 1,272 15.0 52.6 80 1949 1950 Lost 1,725 1,275 11.4 40.0 60 1949 1950 Hemlock 1,570 1,250 6.4 22.5 40 1949 1950 West Lost 400 500 5.9 15.7 20 1949 1950 ‘ North Twin 0 0 0 0 0 4 - For the two summers. Secchi Readings The phyt0plankton group is one of the basic converters of inorganic and organic nutrients to forms of energy use- able by higher'aquatic organisms. The phytoplankton is most responsive to the stimulus of added nutrients and the measure of this response has been used as an index of changes in bio- logical activity during and following periods of fertiliza- tion (Swingle, 1959). Secchi readings were used to obtain a chronological record of the changes in amounts of plankton 50 present. It has been shown (Van Dusen, 1947; Chu, 1945) that secchi disk readings provide a reasonably accurate es- timate of changes occurring in amounts of plankton present. Plankton samples were collected periodically for species identification. Secchi readings were taken at least once a week during the summers of 1948, 1949, and 1950 and addi- tional readings were taken as other duties permitted. Bottom Fauna Collections Bottom fauna samples were collected by means of a 6 inch by 6 inch Ekman Dredge. In sampling, the dredge was raised to the surface and a sifting screen of 50 wire mesh slipped beneath the dredge. The sample was then sifted to concentrate the invertebrate organisms. The organisms and the material not passing through the screen were placed in quart jars, numbered and removed to shore for examination in white bottom enameled pans. The organisms were removed and preserved in a solution of six parts water, three parts 95 percent alcohol and one part formalin. Each sample was labeled as to collection date and source and preserved in separate bottles. The area of collection was confined to the region from shore to a depth of 50 feet. This area had been revealed by preliminary sampling to support over 95 percent of the in- vertebrate fauna inhabiting the lake bottoms. Samples were 51 collected monthly and totaled 55 to 50 samples from each lake each summer. ‘ The sampling was directed towards the evaluation of changes in quantity, quality, or depth distribution of the bottom fauna. The sampling program made possible compari- sons of fertilized and unfertilized lakes and comparisons of bottom fauna populations of each lake prior to and following fertilization. In each of the six lakes large numbers of sunken logs and other forest debris provided a protected habitat for many invertebrate groups (Figure 7). These groups included particularly Ephemerida, Odonata, Neuroptera and many cray- fish, all important fish food organisms. This type of cover could not be sampled quantitatively with a dredge and esti- mates of this segment of the bottom fauna population were not possible. Collections for identification were made of the invertebrates living under the logs in the shallows. Fish Sampling Records of fish were secured from several sources. Early during the summer of 1948 fish.were collected by hook and line, gill nets, minnow dip nets, and a few records were\ secured from a voluntary creel census. The purpose of this sampling was to determine which species were present in order to decide which lakes required poisoning. Data were taken concerning the fish removed by poison in 1948 from Section- 52 Section-Four Lake showing the tangle of sunken and floating logs present near the shore. Such cover is important as cover for trout and trout food organisms. Four, South and North Twin, and West Lost Lakes. A popula- tion estimate was made on the basis of percentage recovery of marked fish. Prior to the 1949 season, the area was officially organized as a trout research area and through a compulsory creel census a complete record of the fish removed by anglers was secured. The fishing pressure during the 1949 season was light and, to suppliment the records of the creel census, samples were collected by gill nets. During the sea- son of 1950, the heavier fishing pressure increased the num- ber of catch records and the poisoning of all the lakes, which terminated the program in September of 1950, also pro- vided a very large sample from each lake. An estimate of the fish populations present was made for each of the lakes on the basis of percent recovery of known numbers of marked fish planted in the lakes just prior to the 1950 poisoning (Ball, Ms.). The creel census was conducted by the staff of the Pigeon River Trout Research Station. Aquatic Vegetation and Filamentous Algae Collections of aquatic plants were made from.each lake. Changes in the conditions and amounts of aquatic vegetation present after fertilization were observed. Since the formation of odorous and unsightly floating mats of filamentous algae resulted from the fertilization of 55 South Twin Lake (Figures 8 and 9) during the summers of 1946 and 1947, close watch was kept on all the lakes for its ap- pearance during the present fertilization program. Only limited quantities of Spirogyra sp. appeared on one lake (Section-Four) during the present period of fertilization but mats covering large areas were present on South Twin Lake during 1948 and to a lesser extent during 1949, appar- ently from a residual effect from the 1946 and 1947 applica- tions. Lake Mapping The maps on which the program was planned were con- structed by crews from The Institute for Fisheries Research in 1951 and 1952. A remapping of all six lakes carried out in September of 1950 indicated that all of the six lakes were smaller than shown on the older maps. Changes in the lake areas are presented here (Table 4). These reductions in area and volume changed the rate of fertilization which had been established on the basis of the old maps. 56 g .1 A L La in \0 vs \‘\ 3 \\ K- \‘ ‘ O 1‘ c ‘ .4. “\\.A \1 ~ A 4~a Ca «C A: ‘ "‘ y‘ A. .h-‘ILA‘ \ S iii“ 0 .38 re- 4“ "V~“ .3 \s‘ r» '\ \Ik' ~.: -‘V 9‘— \ \‘ unmet. .- o I. x. .mrv., ( 58 TABLE 4 AREA REDUCTIONS INDICATED BY REKAPPING (1950) Area indicated New area Lakes on old maps (1950) Change in area South Twin 4.7 acres 5.9 .8 North Twin 5.4 4.7 .7 West Lost 4.5 5.7 .6 Lost 5.0 5.5 1.5 Hemlock 6.0 5.9 .1 Section-Four 5.5 2.6 .7 Laboratory Procedure Bottom Fauna Samples The laboratory procedure for examination of bottom samples consisted of identification of organisms and deter— mination of numbers and volumes. Identification to species was not attempted. Had such identification been possible, the expenditure of time would not have been justified in this type of investigation. Numbers were determined by actual count and volumes by the displacement method described by Ball (1948). In obtaining the volumes of some of the smaller organisms it was necessary to accumulate large numbers and determine average individual volumes from the total volume and total number. 59 Stomach Analysis Brown trout stomachs were collected from each lake. Examination procedure consisted of identifying food organisms present and obtaining a numerical count. Volume determina- tions were made on the entire stomach content and percent of total volume made up by each taxonomic group estimated. Brook trout stomachs were collected from South Twin Lake and examined for a comparison of feeding habits between brook and brown trout. 40 THE EVALUATION OF THE EFFECTS OF FERTILIZATION Physical and Chemical Effects of Fertilization Thermocline Changes A weekly temperature series determined the thermocline position (Figures 10, 11, and 12). The thermoclines of the fertilized lakes (West Lost, Lost, Hemlock, and Section— Four) appear to have been in shallower positions in 1949 and 1950, the fertilized period, than in 1948, the year prior to fertilization. The thermoclines in North Twin and in South Twin Lakes (unfertilized) showed very little change during the same period. The changes in positions of the thermo- clines were analysed statistically with an analysis of vari- ance as described by Snedecor (1946). The mid-points of each thermocline determination were used as the sample value (Table 5). The thermocline position in 1948, prior to fer- tilization, is compared to the position occupied during fertilization, 1949 and 1950 (Appendix p. A). Thermoclines of the three lakes receiving the largest applications of fertilizer (Hemlock, Lost, and Section-Four) changed to a shallower position during 1949 and 1950. No real change oc- curred in the average position of the thermoclines in North or South Twin, the unfertilized lakes, or in West Lost Lake, the lake receiving the least amount of fertilizer. 41 .aowhno no .E.m.m w can» mama ape: hopes no easaop use eaaao lashes» an» no soapmeoa Hmoapnob ans 583 55 £58 95 53. 5.82 . 3 35E E! 5303. _ 00m. 52, $235 E 582 _ >42. _ m22. Kma— 53034 _ >435 _m22, 0+2 gm. ........ ........ ........ ........ 1 ..~\\\ m K < I. Z _ \S ._.. \\ \ .H\\\\\\\\\ .x\\ 33530 I No uosnlmvv § mzjoozmmxs a Ewm_ 5303. H 52. :23 Km: 5302 _ >35 :22, Km: 530344 52, _m231 0mm. 32 mg. Mde 7:2; IFEOZ 45 .nomhxo no .E.m.m ¢ can» need and: none! no magma» use onfiao nosuonu on» no coapaooa Hmoapaop one 583 53 use 33 to; .3 353 WEST LOST LAKE I950 I AUGUST [SE PT JUNE[ JULY |949 JUNE [ JULY ] AUGUST [SEPT LOST LAKE |948 JUNE[ JULY [AUGUST [SEPT W <4F’.P.M.OFO2 m THERMOCLINE - OVERLAP -.;.;.-.;. . .... .......... 20 3O ('15) H1030 I950 JUNE] JULY [ AUGUST [SEPT - |949 JULY [ AUGUST [SEPT JUNE [ ..... ..... ..... ..... ....... ........ |948 AUGUST [SEPT 1 JUNE [ JULY (13)H1d30 45 .nomhxo mo .E.m.m w can» need nun; hope: no oEfiao» use czado Iosuonp an» no coupwooa Hwoapnob one .moxeq Moodaom cam Adomncoapoom .mH ohsmfim ............. n.n.u.u.u.u.u...u.u.u.u.u.u.u.. ... . ..n.. ................................ ................................... ....................................................................... ..................................................5.”.................u.u.u.. ............................................................................ .....................................................u.u ....................................................... ............................. ......................... .............................. ............... ....n.u.u.u.u.n.n.n.u.u.u.u.u. ................. ...........u.u.n.u.u.n.n.u.u.u..... ...................................... .......................................................................... .............................u.n... .......................................... ............................................................ ..................................... n.n.u.u.n.u.u.u.u. ........................................................... .......................................................................... ............ ............................................... mun"xx"mu"mu””mum”xunxxuumumummm”mmumwwwwwwwww HHHH .u.u.n.u.u.u.u.u.u. ...................................................... .................................................................. . . . ..n.u.u.u.u.u.u.u.n.u.n.u.u.u.n.u.n.u.n. .... ..n.u.u.u.n.u.u.n.u.u.u.u.n.u.n.n.n.n. .... ........................ ....... >435 mzaw Emm Llama? _ 52, U235 _ a 9 L V dej XUOJEwI 9.9 midi n50“. 29%me TABLE 5 AVERAGE DEPTH IN FEET OF THE TI nRMOCLIN‘E KID-POINTS Lake s 1948 1949* 1950* Unfertilized North Twin 25.1 28.0 26.7 South Twin 21.6 24.6 21.2 Fertilized ‘West Lost 25.8 26.0 25.4 Hemlock 22.0 15.2 14.9 Lost 25.6 21.5 22.4 Section-Four 50.1 27.6 21.0 % - Period of fertilization. An explanation for the change in the position of the thermocline lies in the density of the plankton populations. Clarke (1946) presented data to show that the depth of light penetration was dependent on the amounts of plankton pres- ent. Since an increase in plankton occurred in the fertilized waters, the heat no longer penetrated to the depth that it did prior to fertilization but rather was absorbed in the water layers a short depth below the surface. The water of the hypolimnia remained at temperatures only slightly above 4 degrees C. The sheltered position of the lakes permitted the establishment of the thermocline in the shallower posi- 48 tion. The importance of the upward shift of the thermo- cline is that, in lakes where oxygen is depleted in the hypo- limnion during the summer stagnation period, the total volume of oxygenated water available to the trout popula- tions is reduced. In a lake such as Hemlock, where the hypo- limnion is already proportionally very large, this is an important factor. The thermocline in Hemlock Lake was at a level higher than that of the other five lakes prior to fer- tilization due to greater turbidity present. Oxygen Depletion The addition of fertilizer to lakes may reduce the amount of dissolved oxygen. Hasler (1947) has pointed out that inadvertant enrichment resulted in the exhaustion of oxygen from the hypolimnion of EurOpean lakes and caused the extinction of the Whitefish and trout populations. Ball (1948) has reported the death of the fish and invertebrate populations caused by depletion of oxygen and accumulation of decomposition gases under the winter ice following appli- cation of fertilizer. That fertilizer increases the fish food organisms has been adequately demonstrated (Swingle and Smith, 1959; Patriarche and Ball, 1949). However, changes in oxygen levels may reduce fish production and prohibit artificial enrichment of natural waters. During the present study attention was given to the detection of changes in the oxygen content of the lakes. 49 The average depth at which the amount of dissolved oxygen fell below 4 p.p.m. was determined by weekly observations during each summer of the investigation. The amount of oxygenated water present during the summer stagnation period was reduced in the fertilized lakes while little or no re- duction occurred in the unfertilized lakes (Table 6). TABLE 6 SUMNER OXYGEN LEVELS AVERAGE DEPTH IN FEET WHERE OXYGEN CONTENT FELL BELOW 4 P.P.M. Lakes 1948 1949* 1950* Unfertilized North Twin 55.5 57.1 55.5 South Twin 27.0 50.7 28.7 Fertilized West Lost 57.1 29.0 29.0 Hemlock 27.1 20.0 18.4 Lost 49.5 55.5 50.5 Section-Four 59.9 54.7 55.2 % - Period of fertilization. An analysis of variance was applied to the data from each lake to determine if the differences in amounts of oxygenated water present in the prefertilization period and during the fertilization period were significantly different. 50 The amount of oxygenated water present during the period of fertilization in the fertilized lakes was significantly less than before fertilization. There was no change in the un- fertilized lakes during the same period. From the analyses (Appendix p. B) it was concluded that the oxygen reduction was due to fertilization. In general, the oxygen reduction occurred only in the hypolimnion. Graphing depth at which the oxygen content fell below 4 p.p.m. chronologically (Figures 10, 11, and 12) indicates that the line of oxygen depletion after fertilization follows quite closely the bot- tom of the thermocline. Kusnetzow (1959) found that the respiration of bacteria is the primary agent of oxygen re- duction in the hypolimnion. The stimulus of added nutrients increases the biological production and increased amounts of dead organic matter sink into the hypolimnion. The respir- ation of the bacteria growing upon it results in a more rapid oxygen depletion during the summer stagnation period. The average amount of oxygenated water present during the two summers of fertilization has been compared with the average amount present during the summer prior to fertili- zation (Figure 15). The prefertilization oxygen levels have been assigned the value of 100 percent. The percent of oxygenated water remaining during the period of fertiliza- tion is indicated by the graph. The more fertilizer added the greater the reduction of oxygenated water. The amount of oxygen depletion in Section—Four Lake was slight during 51 6.9coenea ooa o» Haswo mao>ea cmmhxo soapwuaawpneuohmv .ao«pwuaadpnom no weapon madame Apnea Isoa hpv meanHwEon nope: vopwaowhHo mo Apnea .na onswam E ............................................................... ............................................................................................................................... ............................................................................................................................... ............................................................................................................................... .............................................................................................................................. .. oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo IOO—u— BO—t— 60-w- lNBDBBd 55 the first summer of fertilization in spite of the fact that it was receiving the most fertilizer. The very large volume of the hypolimnion in relation to the surface and littoral areas accounts for this lag in the oxygen being depleted in the hypolimnion. In South Twin Lake a slight reverse trend can be ob- served. South Twin Lake was fertilized during the summers of 1946-47 and in 1948 there was a considerable volume of water in the thermocline and below in which there was less than 4 p.p.m. of oxygen. These data are an indication of carry over of the effects of the fertilizer added the pre- vious summers. There was oxygen in excess of 4 p.p.m. throughout the entire lake volume during the summers of 1949 and 1950. Several important physical and chemical differences existed between Hemlock Lake and the other lakes of the group. The thermocline occupied a very shallow position prior to fertilization and, as indicated previously, moved to an even shallower position during fertilization due to the increased turbidity. Prior to application of fertilizer, the proportionately large hypolimnion was completely devoid of oxygen during most of the summer stagnation period. The volume of the lake available to the trout during this period was not large originally and was reduced further when fertilization resulted in the repositioning of the thermocline. Thus, while the oxygen was not depleted from 54 the zone of the thermocline by the effect of fertilizer, the water volume available for the trout was reduced. This may have contributed to the low production of trout from.Hemlock Lake and should be considered when lakes are selected for fertilization in the future. There is another quite different evaluation of the re- moval of oxygen from the hypolimnion. Einsele (1958) has in- dicated that, in general, when iron is present in the hypo- limnion it combines with phosphorus as ferric phosphate in the presence of oxygen. The precipitate formed is insoluble and results in the removal of phosphorus from the nutrition of the lake. If, however, when the iron and phosphorus com- bine oxygen is absent the product is ferrous phosphate which is soluble and it can be presumed that some of it will be reused and will not be lost to the lake. Mortimer (1941) has indicated that, generally, the processes of reduction re- turn the nutrient material in forms more useable by plants and bacteria than do the processes of oxidation. The process of eutrophication as described by Lindeman (1942) is ex- tremely slow while the lake is oligotrophic. However, once the lake assumes the characteristics of an eutrophic lake which include lack of oxygen in the hypolimnion during the summer stagnation period the process of continued eutrophi— cation and extinction becomes very rapid and, as Mortimer points out, the differences of the products of oxidation and reduction processes very likely account for this sudden 55 quickening of the rate at which a lake becomes extinct. Changes in the fish populations of a lake due to oxygen depletion during the summer stagnation period are, in general, long term ones and the final effects may take sev- eral years to evaluate. However, winter oxygen depletion and the resulting winterkill is an immediate pr0position and the results plainly discernable. The winterkills re- sulting from previous fertilization attempts in Michigan were factors that prompted the present study. It was hoped that using several rates of fertilization would permit the determination of what was a safe rate of fertilization and what was not. Unfortunately, the lakes were very inacces- sable during the critical period of late winter, however, one trip to the lakes during the period adjudged to be the most critical was made each of the three winters. The results of the oxygen determinations at the time when snow and ice had been of the longest duration and the sun had not yet removed the snow cover from the ice are presented in Table 7. The level of dissolved oxygen is ex- pressed as the depth in feet at which oxygen supplies fell below 2 p.p.m. Some oxygen depletion occurred in all the fertilized lakes and in Section-Four Lake, the lake receiv- ing the most fertilizer, the amount of oxygenated water re- maining was only four feet in depth on February 26, 1951. Considering that this was only one reading and conditions may have deteriorated before spring, winterkill conditions 56 may have occurred later and remained undetected because of the absence of fish. It must be concluded that the rate at which fertilizer was applied to Section-Four Lake was ex- cessive. TABLE 7 WINTER OXYGEN LEVELS DEPTH IN FEET WHERE OXYGEN CONTENT FELL BELOW 2 P.P.M. Lakes 1949 1950* 1951* Unfertilized North Twin ---- 27 27 South Twin 1 12 8 25 Fertilized West Lost 27 18 22 Hemlock 22 11 18 Lost 55 50 24 Section-Four --—- 56 4 * - Period of fertilization. For the other lakes, winterkill conditions apparently were not approached. The small amount of fertilizer applied to West Lost Lake did not create conditions approaching a winterkill and could be used elsewhere on similar lakes with- out endangering fish life. The data also indicate that 57 oxygen reduction due to fertilization tends to increase as fertilization is continued (Table 7). This is also discern- able in the data for the summer oxygen levels (Table 6). South Twin Lake which was fertilized heavily in 1946 and 1947 and which winterkilled during the winter of 1947-48, was observed carefully in order to detect any residual effect of fertilization on winter oxygen levels. The oxygen deple- tion during the winters of 1949 and 1950 was severe while in 1951 the volume of oxygenated water in South Twin Lake was comparable to the amount present in North Twin Lake, the un- fertilized lake (Table 7). It was concluded that the oxygen depletion during the winters of 1949 and 1950 was a residual effect of fertilization. Changes in total hardness Included in the weekly water chemistries was the deter- mination of the methyl orange hardness. In comparing the average methyl orange readings of prefertilization dates to the readings during fertilization (Table 8) it is apparent that there was a reduction of hardness in each of the six lakes during the three years of the study. The rate of reduction appeared to be greater for the fertilized lakes than in North Twin and South Twin, the unfertilized lakes. 58 TABLE 8 RATES OF REDUCTION IN TOTAL HARDNESS IN THE UNEERTILIZED AND FERTILIZED LAKES DURING THE FERTILIZATION PERIOD Average total Average hardness total Standard 1949 and hardness error of Lakes 1950 1948 Quotient quotient Unfertilized 4 North TVlin 2905 0/0 31.7 = 0950 0016 South Twin 67.8 ./. 74.0 a .916 .055 Fertilized West Lost 120.5 ./. 157.1 a .875 .010 LOSt 12602 0/. 17705 = 0715 0010 Section-Four 155.0 ./. 191.9 a .798 .006 % - Divided by. To determine if there was a real difference in the rate of reduction in hardness between the fertilized lakes and un- fertilized lakes during the period of fertilization the fol- lowing statistical procedure was applied. The average hard- ness of each lake for the prefertilization period and for the period of fertilization were calculated, together with the standard error of their means. The value for the aver- age total hardness of the period of fertilization was divided by the figure for the prefertilization period. 59 Since all the lakes were softer during the period of ferti- lization than previously, the quotients were all less than one and represented actually the percent of the prefertili- zation hardness present during the period of fertilization (Table 8). The standard error of the quotient (Baten, 1958) was used to determine confidence limits of the quotients. The quotients and plus and minus twice the standard error of each were plotted (Figure 14). This graphic method of sta- tistical presentation (Dice and Leraas, 1956) enables one by observation, with some exceptions, to determine if a signif- icant difference exists. Plus and minus twice the standard error is enclosed in a rectangle for each quotient. When the rectangles of the quotients overlap numerically there can be no significant difference in the values, when no over- lapping occurs there exists a significant difference between the values of the quotients. Exceptions occur when over- lapping is slight or overlapping is very closely approached. In these instances "t" tests must be employed (walker, 1945). The quotients for West Lost and South Twin Lakes were tested with a "t" test and it was determined that there was no significant difference. Hemlock, Lost, and Section- Four, the lakes receiving the heavier applications of fer— tilizer, were reduced in hardness at a rate which was greater (highly significant) than for the two unfertilized lakes, North and South Twin (Figure 14). West Lost Lake was fertilized very lightly and did not differ signifi- 6O Figure 14. The quotients obtained by dividing the average total hardness during fertilization by average total hard- ness prior to fertilization for each lake. Graphed with 3 twice the stan- dard error of the quotient. Fer- tilization rates indicated by graph at base of chart. .80 UNFERTILIZED FERTILIZED N.TWIN I 5, ——[ ‘_‘l_'_W_IWN _ __[ W.LOST _" T B SEC.FOUR E HEM LOCK LOST E TOTAL FERTILIZER APPLIED RPM. .lu 62 cantly from South Twin Lake, one of the unfertilized lakes, but was different from North Twin Lake. Also shown are the amounts of fertilizer added to each lake which indicates that the rate of hardness reduction was a function of the rate of fertilization. The increased biological activity which occurred fol- lowing the application of fertilizer was apparently the cause of the decrease in total hardness. The increase in the phytoplankton populations present resulted in an increase in the demand for carbon for the synthesis of carbohydrates during photosynthesis. The plants use the carbon present in the form of carbon dioxide and are capable of using the half- bound carbon present in the calcium bicarbonates. Further, the sudden reduction of the carbon dioxide present in the water has been observed to reduce the solubility of carbon- ates in the water and bring about their precipitation in crystalline form (Huber-Pestalozzi, 1958; Hasler, 1947 quotes Hinder, 1922). These factors singly or in combina- tion would result in a decrease in total hardness. The importance of this reduction in total hardness should not be overlooked. When calcium carbonates and bicarbonates exceed a level of about 177 p.p.m. of methyl orange hardness, phosphorus combines with the calcium and is precipitated out in an insoluable compound and thus may be removed perman- ently from the nutrient cycle of the lake (Hubalt, 1945). Barrett (1952) suggests that the critical hardness level is 65 lower. Two of the four fertilized lakes, Section-Four and Lost, had an average total hardness that exceeded 177 p.p.m. (Table 8) prior to the application of fertilizer. In both, the action of the fertilizer reduced the total hardness below the critical level. These results may offer a possible ap- proach to the problem of increasing the productivity of marl lakes, many of which are very unproductive at the present time. Biological Effects of Fertilization The factors necessary to measure production are an es- timate of the standing crop and the rate of turn over. The standing crop can be determ ned by a sampling program but rate of turn over is difficult to measure. Brujewica (1959), in a study to evaluate the dynamics of production in the CaSpean Sea, describes methods of estimating rate of turn over. He determined the yearly P/B coefficient for the or- ganism of each of the trophic levels of productivity by es- timating production (P) and the biomass (B) obtaining the coefficient P/B as an estimate of the rate of turn over. Since a high level of production requires either a large standing cr0p or rapid rate of turn over, or both, increases of the standing crop will be reflected in relative increases in production when there is little or no change in the rate of turn over (Clarke, 1946). In the present study the standing crops of organisms at several trophic levels have been measured periodically. Changes in the standing crop have been interpreted as reflecting a similar change in production. 64 Considerable confusion exists in the terminology used to describe various aspects of the dynamics of productivity and, as Clarke (1946) suggests, it is necessary to define there terms. Biological productivity as used here (Ivlev, 1945) is the capacity of a body of water to produce a given quantity of organic matter in some particular form. Production is the output or increase of the biomass of any group of organisms per unit of time. Biomass is synonomous with standing crop which has been defined by Clarke (1946) as the amount of or- ganisms existing in the area at the time of observation. Yield is the amount removed by man per unit of time. Plankton Throughout the program emphasis has been placed on mea— surement of quantitative changes in various levels of the food pyramid. Qualitative examinations of plankton were not undertaken and frequent secchi disk readings were the only measure of changes in the planktonic populations. Several workers have evaluated the use of the secchi disk as a means of estimating changes in plankton levels (Van Duesen, 1947; Chu, 1945). It is believed that sufficient understanding of the relationships between light absorption and plankton concentration will permit general conclusion as to the ef- fect of fertilization on the amounts of plankton present. The secchi disk measures the rate of light absorption. Equal fractions of light are absorbed by successive layers of equal thickn ss of the light absorbing substance. If, 65 for example, 20 percent of the light falling on the surface of the lake is absorbed in the first foot of water, then 20 percent of the remaining light is absorbed in the next foot of water. A secchi disk reading is approximately the point at which 85 percent of the light falling on the surface has been absorbed (Clarke, 1946). Changes in the transparency of the water change the readings in a geometric proportion and not a straight line or numerical proportion. The concept of a direct relationship between amounts of plankton present and rate of light absorption has been inves- tigated by many workers under a variety of conditions. Clarke (1958) found no correlation between plankton densities and changes in rates of light absorption. However, the same author (Clarke, 1946) showed that there was a relationship between depth of secchi readings and amounts of plankton present. Utterbank (1946), working in Puget Sound, found a considerable degree of correlation between levels of plankton populations and rates of light absorption in some areas of the Sound and not in others. He pointed out as factors re- ducing the degree of correlation, tide, surface disturbance, wind velocity and direction and amount of light falling on the surface of the water. Depending on the degree of interference of other un- measurable sources of variation, a correlation exists be- tween rates of light absorption and quantities of plankton present which can be measured by secchi disk readings. The relationship is a geometric one, a factor that must be kept in mind when evaluating changes in secchi disk readings. 66 For example, if the rate of absorption (amount of plankton present) is increased five percent to 15 percent then the reading would be approximately ll feet, a total change of seven feet. however, if the rate of absorption should be 50 percent with a secchi reading of 5 feet and a five percent increase in the rate of absorption (amount of plankton pres- ent) to 55 percent occurs then the change in the reading is only one foot. If the changes in rates of absorption were due entirely to plankton increases there was an increase of 5 percent in each instance. However, there was seven times as much change in one secchi reading over the other. Because of the geometric nature of the relationship secchi disk readings cannot be used to compare the quantities of plankton present in bodies of water that differ in trans- parency due to other factors. Secchi disk readings should not be used to indicate changes in the amounts of plankton present in the same lake if turbidity is affected periodi- cally by other factors. In the present study, the factors effecting the transparency of water other than plankton are alike for each of the bodies of water being compared. No other factors affecting transparency occur sporadically dur- ing the period of comparison and secchi disk readings may be used in evaluating changes in quantity of plankton present. iany workers have indicated that increases in plankton will occur following the application of fertilizers (Swingle and Smith, 1959; Hogan, 1955; Keehean, 1956; Howell, 1942). In southern waters it is desirable to maintain a bloom of 67 plankton throughout the growing period. However, during the summers of 1946 and 1947 Ball (1948), in fertilizing South Twin Lake, One of the group used in the present study, failed to achieve more than a moderate increase in plankton and yet winterkill conditions occurred during the winter of 1948-49. There was, therefore, some question if the reduced rates of application proposed for the present study would achieve any detectable increase in the amounts of plankton present. Prior to fertilization frequent secchi readings served to establish the relative amounts of plankton present in each lake. The average secchi readings for each lake were calculated (Table 9) and the readings for each lake are charted chronologically (Figures 15, 16, and 17). Gener- ally, the amount of plankton present was low. The low level of the plankton population in Section-Four Lake is emphasized by readings of greater than 40 feet. The secchi readings indicated slightly more plankton present in the other five lakes but all are considered to have been very low in plankton production prior to fertilization. The first fertilizer was applied in mid-June of 1949 and two more applications followed at three week intervals. There were greatly increased amounts of plankton present in each of the fertilized lakes. Fearing repetitions of the 1948 winterkill on South Twin Lake, fertilization was halted. 68 TABLE 9 AVERAGE SECCHI DISK READINGS (DEPTH IN FEET) Lakes 1948 1949* 1950* Unfertilized North Twin 15.1 55.1 28.9 South Twin 12.7 21.6 17.1 Fertilized west Lost 25.7 15.5 15.8 Hemlock 11.1 4.1 4.1 Lost 17.0 9.9 7.5 Section-Four 55.7 15.7 12.2' % - Period of fertilization. During the remainder of the summer there was little reduc- tion in the amount of plankton present in the fertilized lakes with the exception of Section-Four Lake. The exces- sive hardness of this lake may have caused the rapid de- cline of the plankton population by combining with the nutri- ents, particularly phosphorus, to form in insoluble carbon- ate compounds (Barrett, 1952). Applications of fertilizer in 1950 were begun on June 17 and three applications were made during the summer. The standing crOp of plankton (Table 9) the second year (1950) did not greatly exceed those of 1949 in the fertilized lakes. 69 obvad and mfiad wcandv doudduphou as! oxen case npfiom and Uonaaauhoh non amt date nvhoz .ommH find mvmd .mwmd awnfiuaoh «nooou ho chooom .mH unfiwdm 00w. m¢m_ 8m. om e 3 d I l. H \...:I ll: .1 o. ( Emm— 53034 _ 52 323 Fm: 5303.— Saw _mz:n Hum 5:034 _ haw _ mzsw com. www- gm. wx<1_ 2.2:. IFDOm 13 1min 1 G 3 III .0 1 fl (ix #ol. Emm— kmaoaa. _ 52. 323 Km: 5:93.— 52. :23 Emfl GaeD<_ has Lmzsn wxaj 2.2; IFGOZ ’71 .omoa ecu mama mufinse eomfiaapuoc neon one: moses smog use smog use; .ommH Una mwod .mmmfl uwsfivwoh Eooom .Ho Uhooom .0.” osmam 8] 1.9. +1 m om m 1 H 1] \4d om O J 1] o. Ewfl 5:92 _ 52. :22. Emm_ 53054 4 52. _m22. mm.— »maog _ 52. _ wzaq one. 39 39 my?) kmOJ O 3 d l H A.» u 1.9 Emfl 5:53 _ 52. 5233 5mm.— ..maei _ 53 :5? Emfl 5:34— 52. _m22, one. 23. 9.9 $2: SS 5%. '75 .ommH ecu meme msaase emuaadpmmm npon one; 33H Mooaaom was 83H yachtsoauoom .ommH one mmma .mema «museums deacon so egoomm .ma onsmam >I‘l/III\ |.\|III\/I./I\l ramm— 5303 _ 53 323 Ken _ 5.303% 53 _mz2. 000. ova. Kmfl 5:03 _ 5% 7522. 0?? (‘13) HidBG wI5, 0 >5 //// Q17 A? Q 0 Q9 Ix A\\\\\\\\\\\\\\\\\\\\\\\\\\\\\§ 1N3383d 87 .oEfiao» aspen he assumed ha msaamsdm Beau sandy Happen no noanamoasoo Lemma . mama eouaaapnmmv mama asomucoapoom .mH masmsm III). :1] )I ll m :4 <[ .J (r :3 O u. 2 9 r— o w en 8 8 1N3383d 89 .oESaop aspen no smocked hp madamsdn seam dqsmu_soppon ho soauauomsoo Emma .. mama 683355 .23 5.3 .8 6.3m; LOST LAKE 11411 0 o v 30*- 20—- lOr—- l . l 1 O O O m h o 90_— fl L~_ lNBDBSd 91 .ossao> aspen no psooaom hp wcaflqamm 869M «gnaw Happen mo soapamomaoo Aommd I mwmd wouaadpnomv 0369 Mooafimm .Hm Ohfiwfim HEMLOCK LAKE A\\\\\\\\‘ “’5 W k\\\\\\\\>§ . . ' ‘=‘=‘="‘==°='=°=-= =.=.=.=.=.=.=.=.=.=.=.= LNBDUSd 999 6:9" 0/ ’1' Qz 4’» 95 I) I] III’IIII ‘1 I I“ .oEbaop Have» we unconoa hp managed» seam «sash Heaven Ho coapfiuoqaoo Emma . mama 66325.5 623 £3 63: .mm 93me KE LA LOST ST E W {0:5} 5‘9ka A 0 °’ 2 i 3.. _ m \\\\\\\\\\\\\\\\ \\\\\\\\ 93,9 \\ . \\\\\\\\\\\\\\\\\\\\\\\ ”E. ..\\\\ d NBDHB 1 0&2) a”? 95 .oBSao» Have» we assayed hp maaamsdn Beau «sash soupop no soupfimomsoo 863226.28 353 55 5.32 .mm .3me NORTH TWIN LAKE \\\\\\§\\: lNBDHQd 97 volume of the bottom fauna was made up of midges. The feed- ing habits of the midges as "ooze browsers" made them one of the first groups to reflect the increase in amount of food present in the form of dead plankton. In July of 1949 midge egg masses covered every object from the lake surface to a depth of 18 inches (Figure 25a) and is indicative of the large increase in midges. During the determinations of the volume of the indivi- dual organisms it was necessary to calculate the average volume of some of the smaller invertebrates. Presented below (Table 16) are the average volumes of several sizes of midges and of the small scud Hyallela and the small mayfly Caenis. TABLE 16 CALCULATED AVERAGE VOLUMES OF SHALL INVERTEBRATES Number of Nillimeters Volume in cc. organisms Nidges 0 - 4 .0005 200 5 - 6 .0014 150 7 - 8 .0052 200 9 - 10 .0046 200 ll - 12 .0070 100 15 - 14 .0120 5O 15 - 16 .0178 45 Hyallela .0055 250 Caenis .0050 140 97a a \A ‘ 4 I v“. 7“ -A 5'38 0 ~ v-r‘ - . "' ‘ -.--A U ee: \_. . ‘ 1P1“ .-h~s' .9 98 In South Twin Lake there were extensive beds of potomo- getons and 92333 in the littoral zone. In contrast, the other five lakes had very little rooted aquatic vegetation. There were, then, important differences in the availability to the fish of the food organisms between lakes. The fac- tors of availability could not be measured and make difficult the evaluation of differences in potential fish production be- tween lakes on the basis of amounts of fish food present. In comparing the food levels of these lakes with the re- sults of others (Ball, 1948; Eggleton, 1951; Adamstone and Harkness, 1925; Deevey, 1941) the extreme paucity of trout food organisms is indicated. In earlier studies Eschmeyer (1957) indicated that these lakes provided interesting trout fishing prior to the establishment of perch populations in the lakes. Others have shown that an increase in the levels of bottom fauna populations can be expected following re- moval of all or part of warm water fish populations present (Ball and Hayne, 1952; Wilkins, 1952). Thus, care must be taken in attributing increases in the bottom fauna during the present study entirely to the effects of fertilization. The study was designed so that the sampling of control lakes and the two lakes not poisoned would provide a means of determining the increases in fish food organisms due to fer- tilization and that which was a result of the removal of the predation by the fish populations present prior to poisoning. The volumes of the trout food organisms (Table 17) 99 TABLE 17 AVERAGE VOLUME OF TROUT FOOD ORGANISMS PRESENT DURING SUNNERS OF 1948, 1949, AND 1950 Average volume in cc. per sq. foot Record of Lakes 1948 1949 1950 poisonings & fertilizations North Twin .16 .15 .16 Poisoned* Unfertilized South Twin .49 1.47 1.54 Poisoned* Fertilized . 1946-1947 Section-Four .05 1.10 1.54 Poisoned” Fertilized fly 1949-1950 Lost .07 .57 .29 Not poisoned”“ Fertilized -. 1949-1950 Hemlock .04 .71 .24 Not poisonedww Fertilized fl 1949-1950 West Lost .07 .19 .44 Poisoned” Fertilized 1949-1950 * - August 1 and 2, 1948 complete kill achieved. ** - Minnow populations present throughout period of the study included fatheads, bluntnose, northern redbellied dace and golden shiners. together with the record of the poisonings and applications of fertilizer are shown for each of the lakes. From these data the apparent increase of trout food organisms in five of the lakes and the effect of the competition of minnow pop- ulations in trout lakes may be evaluated. Four of the lakes, West Lost, Hemlock, Lost, and Section-Four, were fertilized during 1949 and 1950. During the first year of fertiliza- tion, the increase in the volumes of food organisms present was correlated very closely with the amount of fertilizer lOO added (Figure 24). During the second year of fertilization, 1950, the food organisms of the two lakes in which trout alone were present, Section-Four and West Lost, continued to increase, however, the amounts of food organisms present in the two lakes in which populations of minnows were present decreased sharply. The interpretation applied to these data is that the bottom fauna organisms were able to utilize larger food supplies to achieve important increases in one season. However, the larger supplies of food organisms available increased the survival of the minnows spawned in the spring of 1950 and the increased predation of the min- nows depressed the pepulation of fish food organisms. Ad- ditional evidence that this interpretation is correct may be found in Figures 27 and 28 showing the depth distribution of the fish food organisms for the two lakes with minnow populations, Lost and Hemlock. The 1950 decrease in amount of food organisms occurred chiefly in the depth zone of from 0 to 5 feet, where the minnow populations were concen- trated. If the above interpretation of the data is correct, then it is apparent that minnows in trout lakes are unde— sirable. While small numbers of minnows were utilized by the trout for food, the decrease in the efficiency in the transfer of energy when another trophic link is added has been pointed out by Lindeman (1942) and Clarke (1946). Supporting evidence that the presence of minnows in Lost 101 .UoquHUGH one moves nouaaapaom no madness Hence .aomma ecu mama .memav oxma some :a anemone poem museum non mandammao doom pdonp no ofiflao» one .wm oaswam S.TWIN SECTION FOUR x\\\\\\\\\\\\\\\\ 0.. 0...... :...OI.-. o I o c o... a HEMLOCK mmW W. LOST N.TWIN \\\\\\\\\ 8 \N 24 20 9 ‘3 (D V FERTILIZER IN 9. RM 105 and Hemlock Lakes was undesirable may be seen in the sec- tion of fish production. The fish from Lost and Hemlock Lakes caught by fishermen were smaller than the fish caught in the other lakes (Figure 51). The production of pounds of trout per acre for Lost and Hemlock Lakes was lower than for any of the other lakes. In comparing the standing crop of the bottom fauna for each of the lakes during the three years of the study, first consideration is given to the volume of trout food organisms present. The volume of the trout food organisms present in the bottom fauna for each lake for the years 1948, 1949, and 1950 have been correlated with the amounts of fertilizer added in p.p.m. (Figure 24). During 1948, the summer prior to the present fertilization program, South Twin Lake had the largest standing crop of trout food organisms by volume and was followed by North Twin Lake with the other four lakes of the group supporting lower populations of food organisms. South Twin Lake had been fertilized during the summers of 1946 and 1947 and the greater volume of trout food organisms present there is thought to be largely the result of the added nutrients. The 1948 sampling indicated that the volume of trout food organisms in North and South Twin Lakes, .19 and .45 ccs. per square foot respectively, was considerably in excess of the amounts present in the other four lakes of the group, all of which had less than .07 cc. per square foot. 104 The volume of food organisms present increased during the period of fertilization in the fertilized lakes and did not increase in North Twin Lake where no fertilizer was added. In South Twin Lake, fertilized in 1946 and 1947, important increases also occurred. The effect on the bottom fauna of the removal of pre- dation from warm.water fishes may be observed in North and South Twin Lakes from which slow growing populations of perch were removed but which were not fertilized during the present study. There was no change in the volume of trout food organisms in North Twin Lake following the removal of the perch while in South Twin Lake the volume of the trout food organisms was approximately trebled following the removal of the perch population (Table 17 and Figure 24). The eXplanation for this seemingly contradictory evidence is believed due to the presence of extensive beds of sub- mergent aquatic plants in South Twin Lake. Beds of potomo- getons and th;g_were present around the entire perimeter of the lake from near shore to a depth of 15 feet and be- yond. It was in the bottom areas occupied by these plant beds that the increases of food organisms occurred. It is concluded that the predation of trout on the food organisms of the bottom fauna groups was efficient enough to replace to a large degree the former predation of the perch in North Twin, Section-Four, and West Lost Lakes but was inef- fective in South Twin Lake because of the presence of dense 105 beds of submerged aquatics. If this assumption is correct, the increase in the food organisms in the fertilized lakes was largely due to the applications of fertilizer. Two important food items of the brown trout in North Twin Lake were backswimmers (Notonecta undulata) and water boatmen (Arctocorixia alternata) were not collected by the Ekman sampler at a rate commensurate with their obvious abundance. The analysis of the stomach content of the brown trout from.North.Twin Lake (Table 24) also revealed that zooplankton (mostly Daphnia) was an important source of food for the trout. The trout in North Twin Lake grew at a faster rate than fish from any of the other lakes and the production of trout in pounds per acre was second only to South Twin Lake. These facts suggest that sampling the bot- tom fauna did not provide a true picture of the amount of trout food present in North Twin Lake and partially explains how the production and growth.rate of the trout present could be so high while the volume of food organisms indi- cated by the bottom sampling program.remained low (Figure 24). Changes in the vertical distribution of the bottom fauna (mollusks omitted) were recorded during the three years of the study (Tables 18, 19, 20, 21, 22, and 25 and Figures 25, 26, 27, 28, 29 and 50). Certain patterns of distribution are present in each lake and the effect of fer- tilization on these is indicative of the general increase in the productivity_of the lake. 106 TABLE 18 SOUTH TWIN LAKE DEPTH DISTRIBUTION OF TROUT FOOD ORGANISMS BY VOLUME 1948 1949 1950 Depth Av. volume V Av. volume Av. volume (feet) per sample No.” per sample No. per sample No. 0 - 5 .34 7 1.16 8 .94 9 6 - 10 .07 6 .57 10 .88 ll 11 - 15 .10 6 .18 9 .28 8 16 - 2O .22 9 .04 8 .14 8 21 - 25 .01 6 .09 7 .05 9 26 - 50 T 6 .... .. .03 8 * - Number of samples TABLE 19 SECTION-FOUR LAKE DEPTH DISTRIBUTION OF TROUT FOOD ORGANISMS BY VOLUME 1948 1949 1950 Depth Av. volume " Av. volume Av. volume (feet) per sample No.“ per sample No. per sample No. 0 - 5 T 6 .53 9 .72 8 6 - 10 .01 6 .65 10 .25 11 ll - 15 .05 5 .55 13 .27 8 16 - 2O .01 6 .10 9 .48 8 21 - 25 T 5 .04 6 .44 8 26 - 30 T 4 .01 6 .15 8 _ fi-- Number of samples 107 TABLE 20 LOST LAKE DEPTH DISTRIBUTION OF TROUT FOOD ORGANISMS BY VOLUME 1948 1949 1950 Depth Av. volume m Av. volume Av. volume (feet) per sample No.“ per sample No. per sample No. 0 - 5 T 7 .39 7 .05 8 6 - 10 .06 9 .02 9 .03 9 ll - 15 .02 7 .09 8 .11 7 l6 - 20 .01 8 .08 12 .16 6 21 - 25 T 7 .15 7 .07 5 26 - 50 T 7 .15 l .02 4 * - Number of samples. TABLE 21 HEMLOCK LAKE DEPTH DISTRIBUTION OF TROUT FOOD ORGANISMS BY VOLUME g=========== 1948 1949 1950 Depth Av. volume V Av. volume Av. volume (feet) per sample No.” per sample No. per sample No. 0 e 5 .02 12 .89 11 .28 7 6 - 10 .05 11 .09 18 .04 8 ll — 15 .OO 7 .08 15 .04 6 16 - 20 .OO 4 .01 6 .OO 6 21 - 25 .OO 2 .OO 5 T 6 26 - 50 .OO 4 .OO 1 T 7 * - Number of samples. DEPTH DISTRIBUTION TABLE 22 WEST LOST LAKE OF TROUT FOOD ORGANISMS BY VOLUME 108 1948 1949 1950 Depth Av. volume a Av. volume Av. volume (feet) per sample No.” per sample No. per sample No. 0 - 5 T 9 .15 8 .20 8 6 - 10 .01 8 .05 11 .08 10 ll — 15 T 7 .01 9 .09 8 l6 - 2O .01 6 .05 14 .25 8 21 - 25 T 6 .06 8 .06 8 26 - 50 T 6 T 10 .02 8 % - Number of samples. TABLE 25 NORTH TWIN LAKE DEPTH DISTRIBUTION OF TROUT FOOD ORGANISMS BY VOLUME 1948 1949 1950 Depth Av. volume A Av. volume Av. volume (feet) per sample No.” per sample No. per sample No. 0 - 5 .15 7 .16 8 .07 9 6 - 10 .07 6 .06 ll .12 11 ll - 15 .05 7 T 10 .02 8 16 - 20 .01 6 T 10 T 7 21 - 25 .Ol 10 .OO 2 T 8 26 - 50 .OO 5 .OO 1 .02 9 * - Number of samples. 109 anue 25 South Twin Lake (Fertilized 1946 - 1947) Depth distribution of trout food organisms by volume during the summers of 1948, 1949 and 1950. 111 Figure 26 Section—Four Lake (Fertilized 1949 - 1950) Depth distribution of trout food organisms by volume during the summers of 1948, 1949 and 1950. SECTION FOUR LAKE (,o O TIIIIIIII FT. IN IIIIFIIITIIIIIIIIIIIIIITU LEGEND |95O l949 --------- - '943 ..................... ,.. ..................... 1 -------------- 1 ................. .l_ . I---“~, 0-5 6-IO ll -l5 IG-ZO ZI-Z5 26-30 DEPTH IN FEET 115 Figure 27 Lost Lake (Fertilized 1949 - 1950) Depth distribution of trout food organisms by volume during the summers of 1948, 1949 and 1950. LOST LAKE 3.0 I 1 VIII] III I T I LEGEND I950 l949 ---------.- '948 .................... 0.. j ..... “m - O A o-s silo II-l5 l6;20 2T-25 ze-eo DEPTH IN FEET 115 Figure 28 Hemlock Lake (Fertilized 1949 - 1950) Depth distribution of trout food organisms by volume during the summers of 1948, 1949 and 1950. 11'7 Figure 29 West Lost Lake (Fertilized 1949 - 1950) Depth distribution of trout food organisms by volume during the summers of 1948, 1949 and 1950. WEST LOST LAKE 3.o__ H .. LL :- LEGEND _ |95O 0 ~— I949 --------- - W20- |948 ........................ —-——+—- o; - U — 1.o_-__ N- P h .. fl .................. 0-5 6-l0 ll-I5 IG-ZO 2l-25 26-30 DEPTH IN FEET 119 Figure 50 North Twin Lake (Unfertilized) Depth distribution of trout food organisms by volume durine; the sumners of 1948, 1949 and 1950. NORTH TWIN LAKE I33 TITj‘lTTTIIIIIIIIIIIITIIIIIIIIITIITIIIUI J LEGEND l950 I949 ---------- |948 .................... .. 0-5 GLIO Il-I5 DEPTH IN I6-20 21-25 26‘130 FEET 121 Distribution of food organisms in all six lakes in 1948, with the exception of South Twin Lake, was similar to one of the distribution curves described by Deevey (1941). This curve is characterized by having the highest level of bottom fauna populations in the shallow area and a steady rate of decrease in the numbers present as the depth increases. Deevey states that this distribution is characteristic in general of several unproductive Connecticut lakes and points out that European investigators have found the same pattern in unproductive lakes in Europe. The pattern of distribu- tion in South Twin Lake in 1948 differed from the general pattern of the other five lakes in the group. In South Twin Lake the distribution of the bottom.fauna (Figure 25) in- dicated the greatest population of bottom organisms in the shallow water with a general decline as the depth increased until the region of the sub-littoral zone was reached, there the populations increased and a second decrease followed as the depth increased. This type of distribution is somewhat intermediate to distributions indicated by Deevey as charac- teristic of a more productive type of lake which he termed "mesotrophic Chironomus" lakes. Borutsky (1959) calculated the heavy mortality which occurs in midges at the different stages of metamorphosis. The larval midge mortality in depths beyond the littoral zone is nearly complete in unpro- ductive lakes because of the paucity of food. The increased volume of the bottom fauna in the sub—littoral zone was 122 largely from species of "plumosis type" Chironomus. The in- crease in these midges indicated better survival and was a direct result of greater amounts of food present in the form of dead plankton from the blooms which followed fertiliza- tion (Curry, personal communication). It may be concluded that South Twin Lake, fertilized during the summers of 1946 and 1947, had a pattern of distribution for the bottom fauna as a result of fertilization which was characteristic of an eutrophic type of lake. In June and July of 1949 and 1950, West Lost, Hemlock, Lost, and Section—Four Lakes received applications of ferti- lizer. The distribution pattern of the bottom fauna in these four lakes changed from the pattern characteristic of unproductive lakes to a pattern similar to that of South Twin Lake in 1948 and to that of more productive types of lakes indicated by Deevey (1941) during fertilization (Figures 26, 27, 28, and 29). The increase in the volume and food organisms in the sub-littoral zone was again due to increased numbers of midges of the plumosis type. The: exception was Hemlock Lake which had a distribution pattern very similar to that described by Deevey as typical of cer— tain lakes with sub-marginal productivity. During 1949 and 1950 the distribution of bottom fauna in South Twin Lake changed from a pattern similar to Deevey's mesotrophic Chironomus to the pattern characteristic of less productive lakes. It was concluded that this change was due to the subsiding of the effects of fertilizer added in 1946 and 1947. North Twin Lake, the unfertilized control lake, con- tinued to display the distribution pattern of an unproduc- tive oligotrophic lake during the three years of the study. The conclusion may be drawn that the application of fer- tilizer brought about a change in the depth distribution pat- tern of the bottom fauna from a pattern typical of unproduc- tive oligotrophic lakes to a pattern indicated by others to be typical of lakes with a higher level of productivity. The data from South Twin Lake indicated that the pattern 'induced by fertilization remained for the first summer fol- lowing fertilization but disappeared during the second and third summers after fertilization and the pattern typical of less productive lakes replaced it. Stomach Analyses Certain information is provided by the food analyses of brown trout stomachs collected during the summer of 1950. Collections were made at the time of poisoning and during the Opening week of trout season. The data presented (Table 24) include numbers of sto- machs examined, total volume of food present, and average volume of food in each stomach. The percent of the trout stomachs in which each food group appeared and the percent of the total volume contributed by each group were calculated. RELATIVE IMPORTANCE OF 1 4——: TABLE 24 124 INVERTEBRATES AS FOOD OF BROWN TROUT INDICATED BY STOMACH ANALYSES (1950) North South West Section- Lakes Twin Twin Lost Four Lost Hemlock Number of stomachs 25 44‘ 55 26 25 21 Total volume (cc.) 21.0 50.4 41.2 27.5 2.1 5.70 Average vol. per stomach .60 .69 1.25 1.05 .09 .17 Chironomidae (pupa) A 100 74 71 59 87 95 B 44 8 7 5 54 51 Hyallela A . . 5 ... 58 24 ... B .0. 2 000 19 4 .0. Crayfish A ... 57 28 8 9 4 B ... 82 22 15 71 7 Notonectidae A 72 ... 52 ... . . ... B 26 ... 51 ... ... . . Odonata A 8 l5 5 27 4 24 B 6 l l 22 5 21 Cladocerans A 60 9 ... ... ... ... B 15 T ... ... .. ... Ephemerida A 8 4 ... l9 ... ... B T T ... 2 ... ... Corixidae A ... ... ... 4 55 52 B e'ee '00. 000 T 11 15 Terrestrial A 40 .. 42 ... ... l9 insects* B 6 ... 5 ... ... 2 % - Winged ants. A - Percent of stomachs with organisms present. B - Percent of total volume. 125 Bottom fauna provided much of the food for the brown trout. There was, however, considerable variation in impor- tance of the food organisms between lakes. Midge pupae were the most frequent article of food but were not volumetri- cally the most important. While midge pupae were present in nearly all the brown trout stomachs no midge larvae were found. Midges were apparently eaten only during the time of emergence. Crayfish formed the bulk of the food volume in South Twin and Lost Lakes. Notonectidae were important as food in West Lost and North Twin Lakes and corixids were present in many of the trout stomachs from Hemlock and Lost Lakes. Scuds were a common food only in Section-Four Lake and daphnia were present in over half of the fish from North Twin Lake. When the fish had been feeding on daphnia they were the only food present, indicating that the trout actively sought this food. North Twin Lake was the only one of the group where large numbers of daphnia were present consistently. They were present in the lower thermocline and were apparently readily available to the trout. With the exception of midges, the major food items of the trout in North Twin Lake were not adequately represented in the bottom fauna samples. The effectiveness of an Ekman dredge in capturing crayfish is also open to question because of their avoiding action. The average total volume of food present in the sto- machs of fish taken from Lost and Hemlock Lakes was much 126 less than for the other four lakes. Both.lakes were fer- tilized but contained populations of minnows. The growth rate for the trout from these lakes and production in pounds per acre were also lower. Minnows were never an important food item. Foods of brown and brook trout from South Twin.Iake were compared. Midge larvae and snails, common foods of the brook trout, were not present in the brown trout stomachs. Terrestrial insects were found more frequently in the brown trout. Brook trout fed on bottom food organisms not selected by the brown trout, which fed more from.the surface. While much information can be drawn from the limited stomach.sampling program, conclusions drawn should not be weighed too heavily without additional data. Angling Yield During the fishing seasons of 1949 and 1950 four of the lakes were fertilized and in this period the yield of trout to the anglers greatly increased. Trash.fish were removed from four of the lakes in 1948 and in September of the same year heavy concentrations (500 per acre) of yearling brown trout were planted in all of the lakes. The limited records indicate that, prior to fertiliza- tion, poisoning, and plantings of brown trout, only Hemlock and Lost Lakes provided trout fishing. Total take was esti- 127 mated as 50 brook trout from Lost Lake and 25 brook trout from Hemlock Lake in 1948. The average length of these fish was less than eight inches. Very little fishing effort was expended on the lakes after the Opening week end. During the winter of 1948-49 The Institute for Fisheries Research, under a special act of the Michigan legislature, was given control of the six lakes utilized for the present study and five miles of the Pigeon River. The area was designated as the Pigeon River Trout Research Area and is dedicated to trout research. Since its organization, the research station has been headed by Dr. Edwin C00per. During the 1949 and 1950 trout seasons, fishing was by free permit only. A limit of five, seven-inch fish per day was placed in effect on all lakes. Length and weight of all fish removed by fishermen from the lakes as well as complete information on the numbers of hours fished, baits used, and time of day fished were recorded. The collection of these data and the cooperation of the staff at the Pigeon River station was of the utmost value to the writer. During the 1949 season there was an increase in the num- ber of fishermen utilizing the area. The increased use of the area reflected the publicity given the area as a trout research station. The use and fish yield from the lakes increased during the 1949 fishing season (Table 25). The total of 825 hours of fishing represents consider- able use of the lakes. The major portion of effort was 128 TABLE 25 FISHING EFFORT AND SUCCESS BY LAKES (1949) No. of Per cent No. of Catch fishing success- hours Fish per Lakes trips ful fished caught hour South Twin 142 45.0 410.5 200 .487 Lost 85 50.6 224.5 76 .548 Hemlock 50 45.5 104.0 59 .575 Section-Four 19 47.4 51.0 28 .548 West Lost 9 0.0 27.5 -- .000 North Twin __4_ _0_.-0_ =2 _—_: £3392 Totals 289 58.5% 825.0 545 .419* 4 - Average. still early season fishing. The lowered rate of success after mid-Hay is indicated when the fishing success and effort is compared by months (Table 26). Nearly all of the fishing in August and September was confined to South Twin Lake. A decided improvement in catch per hour occurred after the July low and the best fishing of the season in terms of catch per hour was in September. A comparison between the 1948 and 1949 fish pepulations of these lakes indicates the progress made toward their re- habilitation. The removal of the trash fish population in 129 four of the lakes and a measurable increase in the food organisms of the five fertilized lakes provided a situation basically favorable for the brown trout which were intro- duced. Growth and survival was best in South Twin Lake and Section-Four Lake and corresponded to the highest rates of fertilization. West Lost and North Twin Lakes showed poor survival and moderate growth rates. Hemlock lake seemed to have moderate survival and growth. Lost Lake returns in- dicated reasonably good survival but very poor growth. TABLE 26 DISTRIBUTION OF FISHING EFFORT AND SUCCESS BY MONTHS . (1949)- ====;_ __;====================================== Hours of Fish Catch Month effort caught per hour Apr11* 80.5 50 .521 May 458.5 205 .446 June 90.0 54 .577 July 71.5 4 .056 August 51.5 14 .270 September** 59.5 58 .658 Total 821.5 % - Season opened on last day of April. *4 - Season closed on September 11. 150 The six lakes of the present study achieved widespread popularity during the 1950 season. The lakes were visited by fishermen from nine states and 48 Michigan counties. Excellent trout fishing was provided by most of the six lakes. Limit catches (5 fish) were the rule during the fore part of the season and a high quality of fishing was to be had on one or more of the six lakes throughout the season. Valuable information was drawn from the creel census records concerning the food habits of the trout in the lakes, the effectiveness of hook and line fishing in removing brown trout to the angler in comparison to the other species of trout. The data establishes the amount of use made of the lakes in 1950 and the fishing success (Table 27). TABLE 27 DISTRIBUTION OF FISHING EFFORT AND SUCCESS BY LAKES (1950 SEASON) No. of Percent No. of Catch fishing success- hours Fish per Lakes trips ful fished caught hour South Twin 590 47.4 1,065 548 .52 Section-Four 567 58.9 1,000 688 .69 Lost 221 55.4 592.5 470 .79 iemlock 115 55.1 506.5 251 .75 West Lost 167 61.7 596 581 .96 North Twin 257 55:9. 569.5 222 :22: Totals 1,495 51.5* 5,927.5 2,540 .85* * - Average 151 In 1,495 fishing trips to the lakes during which a total of 5,955.5 hours of effort was expended on the lakes, 51.5 percent of the trips were successfull and 2,540 trout were caught at the rate of .65 fish per hour. The catch- per-hour figure is excellent by Michigan standards. The creel census records of 1949 and 1950 compared to the small use made of the lakes in 1948 illustrate the large increase in the recreational value of the lakes. Not only was the increased effort spent on the lakes more profitable in terms of fish per hour of effort but also the effort and success was more equally distributed throughout the fishing season (Table 28). TABLE 28 DISTRIBUTION OF FISHING EFFORT AND SUCCESS BY MONTHS (1950) Hours of Fish Catch Month effort caught per hour AprllT 210.5 251 1.021 May 1,712.5 1,296 .757 June 262.5 117 .446 July 481.0 180 .574 August 851.5 524 .615 September%* 415.0 220 .550 m: . Total 5,952.0 % - Season included only last two days of April. 4% - Season ended September 11. 152 The average size of the trout taken by angling from each lake during the 1950 season was compared using the statistical procedure of covariance (Snedecor, 1946). Since the times of capture varied between lakes (Table 29), it was desired to eliminate the variation in average length of the trout due to differences in time of capture. Covariance in this instance allows the worker to evaluate the effect of time and compare the variation due to differences between lakes after the effect of time has been subtracted. TABLE 29 THE DISTRIBUTION OF PERCENT OF TOTAL CATCH BY MONTHS II * Lakes May June 1 July August September* Hemlock 95.5 4.0 .5 0.0 0.0 Lost 87.7 12.1 0.0 .2 0.0 Section-Four 65.6 6.7 25.1 5.1 1.5 South Twin 55.4 4.9 1.5 54.4 6.0 North Twin 52.8 0.0 2.4 52.4 52.4 . West Lost 10.5 0.0 .5 60.6 28.6 4 - Includes last two days of April. *4 - Season ended September 11, 1950. After applying covariance (Appendix p. C), in order to determine the adjusted mean length of the trout from each 155 lake (Table 50), the method of graphic presentation (Dice and Leraas, 1956) for the mean with plus and minus twice the standard error indicated was used (Figure 51). From these comparisons it was determined that the trout from North and South Twin had a greater average length than trout from any of the other lakes. There was no real differ- ence in the size of the trout between the Twin lakes. The fish from Section-Four Lake were smaller than those from North and South Twin, larger than the fish from Lost Lake and were not different in length from the fish from West Lost and Hem- lock Lakes. No difference could be shown between average length of the fish from Vest Lost, Hemlock, and Lost Lakes. TABLE 50 VERfGE LENGTH N INCHES OF THE BROWN TRO EHOVED BY ANGLING BEFORE AND AFTER ADJUST- HENT FOR TIKE 0F REHOVAL Lakes Unadjusted Adjusted South Twin 9.66 9.65 Section-Four 9.22 9.18 Lost 8.26 8.59 Hemlock 8.59 8.95 West Lost 9.41 9.01 North Twin 10.15 9.96 154 Figure 51 The corrected average length of crown trout caught from each lake “‘T‘:V‘~ 10 Ifi GS“ ‘8‘: V V. INCHES N.TWIN I000 —— SIWIN I I SEGA ‘I'_‘ W. LOST f HEMLOCK — — 1° 0 I LOST . 800 —— 156 The value of the lakes in terms of fishing hours and the fish in the angler's creel underwent a striking increase during the period of the present program. In 1948 two to three hundred hours of early season fishing produced approx- imately 75 brook trout averaging less than eight inches. In 1949 some 825 hours of effort produced 545 trout averaging about eight and one-half inches and in 1950 nearly 4,000 hours of fishing were spent on the lakes and 2,540 brown trout averaging 9.4 inches were creeled. The effort was no longer concentrated in the first month of the season butwas spread throughout the entire season with much of the very best fishing occurring in August and September. Considerable information has been gained concerning the catchability of brown trout in lakes. Fishing quality re- mained high throughout the summer. The heavy fishing pres- sure took a high proportion of the fish present but did not remove nearly all the fish present as so frequently occurs when brook trout lakes are subjected to comparable pressure. Because more trout avoided capture for longer periods, the fishing public had the opportunity of catching fish of larger average size. Further, the brown trout fed on or near the surface and could be taken on a dry fly throughout the season, thus, offering sport superior to the taking of fish from deep water which is the only method of taking brook trout after the early season fishing. It has been suggested that the 1950 season was an ex- 137 ceptionally cool one and the presence of brown trout in the upper waters throughout the season was an exception. Fur— ther trials should provide the answer. -It is suggested, however, that brown trout comfortably occupy warmer water than brook trout and will therefore spend a greater propor- tion of any season in the surface waters than will brook trout. The credit for these changes in recreational value must be attributed to three factors. Trash fish were removed from four of the lakes; the planting schedule was changed from 500 fingerling brook trout per acre to the same number of yearling brown trout; and fertilizer was applied to four of the lakes (a fifth, South Twin, was fertilized in 1946 and 1947). All three changes undoubtedly contributed to in- creasing the yield of fish. These were important factors in total production as well as in yield to the angler and fur- ther discussion is withheld until the discussion of trout production. In the preparation of the data from the creel census program certain interesting observations concerning the nature of the data from this source developed and have been evaluated. The average length of the trout caught each week was calculated. During the first five or six weeks of the 1950 season the average length of the trout captured decreased. This fact was observed on the four lakes (Figure 32) where 138 weekly catches were sufficiently large (15 to 255 fish). In seeking an eXplanation of the decrease in the aver- age length of the fish, selective action of fishing was in- vestigated. It was suspected that, because of the ease in which the five fish limit could be taken on some lakes of the group during this period, that the fishermen were returning the smaller legal fish to the water and retaining only the larger. Continuing this line of reasoning, as the fish be- came more difficult to take a decrease in average length re- sulted. To determine if this was actually what was happen- ing, the limit catches were compared to the catches of only one or two fish, reasoning that the man taking five fish would be more selective. There was no difference in the average of the two catch groups and both showed the decrease noted in the general averages. It was decided that the selection of the larger fish on the part of the fishermen was inherent in hook and line fishing. That the larger, faster growing portion of fish population is more vulnerable to fishing has been one ex- planation of Lee's phenomena (Kile, 1956). COOper (Ms.), working on the Pigeon River in the same area as the present study, recently found that the fishermen in the stream selected the faster growing fish present. It is necessary to explain why the factor of selectivity should produce this type of curve of the average lengths (Figure 52). The correct explanation is thought to be that little growth oc- 159 .dopmoacna Oman cud Aommd .mm .mm nongopmomv wnwcoawom nouww oopoboooa anon» chomp no mnpmdoa ommnobd .comwom ommd on» pnonwfionsp waaawnd hp deposon pack» abomn no Axum: mommy nnpwdoa omwno>¢ .mn onawam kamm km303< >433 MZDU >ogom mo oaflp pm pgmfioa : ts .mwsflpamam mead one mwma 509% muflpwcfimfino nopwo mo QOHp90domm I * n.NH n.bm H.mH H.mm mm. mo. mmm >.w QHEB flpmoz H.ma m.mm w.mm m.woa OS. on. Hmm >.m umoq pmoa o.m >.¢n m.> m.mw mm. ma. Ham m.m Moowaom m.© o.mm ¢.Hm m.¢b as. am. 05¢ m.n pmoq m.mm H.0m o.w© ¢.wma om. ow. mmm m.m AzomanHpoom m.>m o.mw m.mw m.hba aw. Hm. mwm ©.m GHEB Sagom :.omow pod mafipqmam chow pom modded pnoonom uqoonoa gmwm moho¢ momma ...).w pemfioa poz pm pewfio; pemfioa proe Hence smemfl *mema Aommav moeeo mmmqeza Mme mo ZOHBHmomsoo oza emema mm mqmde 157 The survival of the 1948 plantings is indicated in the proportion of the catch contributed by the two plantings (Table 56). The 1948 plant was most important to the 1950 catch in South Twin Lake and Section-Four Lake indicating the best survival in these two lakes which had received the most fertilizer. The poor net yield from Lost and Hemlock Lakes is an indication of the very poor growth in these lakes. Following the 1950 fishing season the lakes were poi- soned. An estimate of the trout population was based on the rate of recovery of 100 marked brown trout, planted just prior to the application of the toxicant, of the same age and size of those already in the lakes (Ball, Ms.). While the differences in the rate of recovery between lakes is large the estimates of the fish population present are be- lieved to be accurate. Differences between lakes in physi- cal features, the amounts of turbidity, and the amounts of rooted aquatic vegetation offer explanations of why the fish were recovered at quite different rates. In.North Twin Lake where the water was clearest, the slope one of the least precipitous and little aquatic vegetation present, the re- covery rate was 81 percent of the marked fish (Table 57). The large portion of deep water and steep sloping shoal areas in Section-four and Lost Lakes are believed to have been factors reducing the rate of recovery (21 and 59 per— cent respectively). In South Twin Lake the heavy beds of 158 ®.no m.mm ¢.om mow mam Hm. CHER Qphoz 0.0H m.ma m.mm wmfi mm mm. pmog pmoa v.¢ H.h m.HH mo . Sm Hm. MooHEo: H.0H m.mm o.mw moa mm mm. pmom m.©H ¢.m b.mm so ma Hm. hflohncoapoom b.vw o.¢m >.mw own mma Hm. QHBB flpdom omow mom popgmam chow nod psmflos .ooy poxmmads ano>ooom momma pflmfios flmflw mo psmfloa Umpmfiflpmm mo pgmflma ooxgmfi use pnmuom seemsflpmm pcoowmm Ammzmom 2H mamemav wszomHom omoa mum mo mmHe mam as azmmmmm mmHm mo memch; oza mmmmmpz ommamHemm hm mqmfia 159 potomogetons undoubtedly accounted for the disappearance of many trout. West Lost, similar to North Twin Lake, had only a slightly lower rate of recovery. Fish samples taken at the time of poisoning have been used to calculate the difference in size and weight of the trout in each lake (Table 58). The small size of the trout from Lost and hemlock Lakes are again indicated. It will be noted that in Lost Lake the 1948 fish which were an inch less in average length at the time of planting than the average length of trout planted in 1949 grew so little that they re- mained smaller even though a year older. TABLE 58 WEIGHT AND LENGTH OF BROWN TROUT FROM 1950 POISONING Fish from 1949 planting Fish from l948 planting Average Average total Average total Average length weight length weight Lakes (inches) (grams) (inches) (grams) South Twin 9.67 - 54 160.4 - 54 10.44 I O1 [.4 201.7 - 51 Sec.-Four 9.89 - 17 149.5 - 17 10.28 - 18 185.6 - 18 Lost 8.55 - 55 94.1 - 55 8.28 - 54 90.5 - 54 Hemlock 8.28 - 54 89.0 - 54 9.41 - 22 152.1 - 22‘ West Lost 9.55 - 54 155.0 - 54 9.95 - 14 155.4 - 14 North Twin 10.80 - 50 215.0 - 50 11.20 — 12 255.0 - 12 160 The net weight of the trout removed from each lake by angling, netting, and the 1950 poisoning have been added to- gether (Table 59). Net production of trout per acre has been related to the amount of fertilizer added (Figure 54). TABLE 59 ET WEIGHT OF FISH REIOVED BY ALL maraops (1949 and 1950) *2. Net weight Net weight Total per acre, per acre, net weight Net weight per acre, 1949 ang- 1950, ang- 1950 poi- per Lakes ling, netting ling soning acre South Twin* 9.5 27.8 54.7 102.1 Section-Four 5.5 55.2 16.5 55.0 Lost .9 6.5 19.1 26.5 Hemlock .9 2.0 4.4 7.5 West Lost .8 19.1 19.6 59.5 North Twint* .5 12.5 55.9 75.5 % - Fertilized 1946 and 1947. 4% - Unfertilized. .3 The effect of fertilization on the production of trout is the most important consideration of the present study. Trout production is, however, the end result of complex processes of energy transfer and is affected by many fac- tors. The interpretation of the effect of these factors on production is difficult. To facilitate understanding of the 161 .thoa .om .mm nopaopoomv we“ naoufioq Mo asap aw paomonm no oopmsapuo Hawaok on» and wzaawnd hp oopoaou 9509p no pnwaot one .codcw nonaaapnom no nuanced on oopwHon Apnmfior pony cued pom 9509p stomp no aoanosoonm . .3 93mg ° k 24 20 - FISHERMEN’S CATCH m POISONING l I 0 co 383V 83d 'SQ'I IOU—*— 80-F- P. RM. IN FERTILIZER effects of fertilization, each lake will be discussed separately. South Twin Lake was estimated to have had a standing crop of approximately 29 pounds of perch to the acre in 1954 (Table 54). A total of 28 p.p.m. of fertilizer was added during the summers of 1946 and 1947. A winterkill occurred during the winter of 1947—48. In August of 1948 the lake was poisoned and the standing crop of perch present was es- timated at 250 pounds to the acre on the basis of marking and recovery. The increase is interpreted as largely due to the effects of fertilization although the estimate of Eschmeyer may have been slightly low. Trout planted in 1948 grew well, better than any of the trout in the other lakes prior to applications of fertilizer in 1949. During the 1950 season, fishermen removed 548 fish from the lake or 28 pounds per acre of not yield. The net weight of the fish removed and the net weight of the fish population present in September 1950 total 102.1 pounds of fish to the acre. The bottom fauna studies indicated that the volume of the trout food organisms per square foot in South Twin Lake was greater than in any of the other lakes (Figure 24). Fertilization is concluded to have resulted in a ten- fold increase in the standing crop of perch. Dense beds of potomogetons appeared following fertilization and probably were a direct result of it. The high production levels of this lake in comparison with the others of the group are 164 believed to have been partially the effect of fertilization but both South Twin and North Twin Lakes must be considered to have been basically more productive than the other four. Section-Four Lake had a standing crop of 25 pounds of perch to the acre in 1954 (Table 54). The estimate of 97 pounds of perch per acre in 1948 is almost certainly too high. During 1949 and 1950, 15 p.p.m. of fertilizer was ap- plied, the largest applications of the present program. Fishermen removed 688 trout weighing 166 pounds during the 1950 season. The total surface area of Section-Four Lake is only 2.6 acres. Prior to fertilization the necessary basic elements of the food pyramid were almost completely lacking. Section- Four Lake could have supported only very small numbers of trout and yield would have necessarily remained very low. Following fertilization the volume of the trout food organ- isms per square foot increased forty-fold (Figure 24) making possible the excellent trout production observed. The popu- lation estimate at the end of the 1950 season indicated that fishermen were able to harvest most of the trout present at the beginning of the season. Had larger numbers survived until the final poisoning their additional growth increment might have raised the production figure to a still higher figure. Lost Lake produced a few large brook trout in 1951 which had been planted in 1927. Regular planting began in 165 1955 and a complete creel census in 1955 and 1956 indicated that fishermen harvested 50 pounds of brook trout per acre in 1955 and 12 pounds in 1956 (Table 54). These are total weights. The net weights as calculated in the present study are not known. In 1948 the yield was 5 pounds of brook trout per acre, all taken in early season fishing. A total of 11.4 p.p.m. of fertilizer was applied during 1949 and 1950. Important increases occurred in the volume of trout food organisms present. Net yield to fishermen during 1950 was only 6.5 pounds per acre. At the time of the poi- soning the trout population was estimated as 19.1 pounds_per acre. Growth of the trout in Lost Lake was the poorest of any of the lakes (Table 58). The increase in amount of food present after fertili- zation was not reflected in increased production or rapid growth. It is believed that the pepulations of several species of forage fish present competed for the food and were not in turn food items of the trout. Because of the error in the old map of the lake (Table 4) Lost Lake was stocked heaviest. The increase in crowding and resulting greater need of food for maintaining body weight is an additional reason for the poor growth of the trout in this lake. Hemlock Lake was planted in 1927 with a few brook trout and from this early planting six trout taken in gill nets in 1951 had a total weight of 20 pounds which is very excep- 166 tional for Michigan. However, the yield in terms of pounds per acre was never high. The creel census of 1955 and 1956 indicated a harvest of 8 and 4 pounds per acre respectively. A total of 6.4 p.p.m. of fertilizer was applied to Hemlock Lake during 1949 and 1950. The increase noted in the amounts of trout food present did not result in a high production of trout per acre. The net weight of fish harvested by anglers and the weight estimated as present at the time of poisoning totaled only 7.5 pounds. The trout in Hemlock Lake ranked fifth in growth rate with Lost Lake sixth. Contributing fac- tors to the low productivity of Hemlock Lake are believed to be the competition of the minnows, the small littoral zone, and the large and biologically unproductive hypolimnion. The pelagic area in Hemlock (largest lake of the group) was also unproductive. West Lost Lake had, in 1948, a standing crop of perch and sunfish estimated as 67 pounds per acre (Table 54). The yield of trout to the anglers during 1955 and 1956 was in- dicated by the general creel census to have been 22 and 21 pounds of trout to the acre. These data, together with the estimated trout food available, establish that, prior to fertilization, West Lost Lake ranked behind North and South Twin and ahead of the other three lakes of the group in potential fish production. West Lost Lake was fertilized very lightly (total of 5.9 p.p.m.) during 1949 and 1950. The removal of the warm-water fish from this relatively pro- 167 ductive lake accounted for some of the increase in food organisms and contributed to the production of 59.5 pounds of trout to the acre. However, the added nutrients pro- vided considerable stimulus to the lower levels of the food pyramid. The increase in the amount of plankton present following fertilization is indicated by the record of secchi disk readings (Figure 16). That increased amounts of organic matter were produced during the period of fertilization is indicated by the increased rate of oxygen depletion in the hypolimnion. North Twin Lake served as a control lake in the pres- ent study. The net weight of trout produced was second only to South Twin Lake and the rate of growth exceeded that of all other lakes in the group (Figure 54 and Table 58). The poisoning of 1948 indicated the standing crop of perch and golden shiners of 118 pounds per acre, second by weight to South Twin Lake (Table 54). The production of bottom fauna before fertilization was higher than all lakes except South Twin. Barrett (1952) determined that the amounts of phosphorus present in North Twin Lake (without artificial enrichment) usually equalled or exceeded the amounts present in the waters and bottom muds of the rest of the lakes in the group. These data establish that the natural productivity of North Twin Lake was high in rela— tion to other lakes in the group without added nutrient mate- rial. Perhaps the most important single factor in the ex- 168 cellent production and growth of the trout in North Twin Lake was the unexplained poor survival of the 1948 trout planting. The original planting rate of 500 yearling brown trout per acre in each lake was considered excessive but was used intentionally to insure the use of any increase in fish production capacity following fertilization. Crowding the lakes with trout placed a burden on the supplies of available food merely to maintain the weight of the fish planted. Because of the poor survival of this planting in North Twin Lake the food supplies were not taxed as heavily. The planting of trout in 1949 at the rate of 250 per acre meant that North Twin was understocked while most of the other lakes, with the exception of West Lost Lake where sur— vival of the first plant was also low, were overstocked. For these reasons the growth of the trout in North Twin Lake was greater than for any of the other lakes and the produc— tion of pounds per acre was exceeded by only South Twin Lake. The production of trout in each lake and the effects of fertilizer in the production have been reviewed. The conclusion has been reached that the effects of fertilizer on trout production can be more accurately determined by comparisons on a before and after basis than by comparing fertilized and unfertilized areas. The differences in fac- tors which affect the potential production capacity of the lakes are so large even in a group as homogeneous in char- acter as the group utilized in the present study that evalu- 169 ation of the effects of fertilization are difficult to deter- mine by comparisons between lakes. North Twin, South Twin, and West Lost Lakes were bas- ically far more productive of fish than Hemlock, Lost, and Section-Four Lakes. The more productive lakes occupied shallower basins and were less alkaline than the three deep marl or partially marl lakes of the group. Veatch (l952) points out that in proceding from the time of formation through the process of aging and to extinction lakes in Michigan generally proceed from very alkaline condition to a less alkaline condition. From this it would also be con- cluded that the North and South Twin and West Lost Lakes were closer to an eutrophic and more productive condition than the remaining three lakes. The excellent production of trout from North Twin Lake (unfertilized) has served to demonstrate that it and South Twin and West Lost Lakes as well, should provide excellent trout growth and a high yield of brown trout without fer- tilization. Lost Lake, hemlock Lake, and Section-Four Lakes are so low in basic biotic potential that trout yield will remain low without artificial enrichment. The small number of trout the lakes are capable of supporting will not pro- vide interesting fishing in the face of the concentration of fishing effort that the trout waters in this state re- ceive. 170 The question of the ethics of artificial enrichment of natural lakes is now pertinent. Hasler (1947), in outlin- ing the objections to fertilization of natural lakes, stresses that enrichment from any source will hasten the eutrophication and extinction of the lakes. Since the basic concept of conservation is the use of natural resources, water included, so that the most benefit will go to the most people. Any action of the present generation which hastens the extinction of lakes is not good conservation. However, in the situation which.exists in the lake type represented by the three unproductive marl lakes where the capacity to provide interesting fishing is lacking or nearly so, the use of fertilizers in effect provides a new source of fish- ing and is, therefore, ethical and does not violate the con- cepts of conservation. The situation is similar to the treatment of bog lakes by Hasler (1951) with.lime to in- crease fish production. ” That fertilization will.make productive this type of marl lake is demonstrated by the large increases in volume of fish.food organisms present and the resulting high.yield of brown trout to the fishermen from Section-Four Lake during the period of fertilization. 171 DISCUSSION The objective of the present study has been to evaluate the effects of the addition of inorganic fertilizer to a group of Michigan trout lakes. Previous studies have indi- cated that fertilizing natural lakes to increase fish.pro- duction may be limited by undesirable changes in the environ- ment or by economic considerations. The investigation‘has been undertaken to determine if reduced rates of fertiliza- tion can avoid or reduce the undesirable effects. Creel census studies have indicated a low harvest of warm-water species and strongly suggest that the expense of fertilizing to increase a crop not adequately harvested can- not be justified. The high rate of harvest from trout lakes has suggested that fertilization there might be econmmically justified if undesirable effects could be avoided by modi- fied techniques. For these reasons trout lakes appear to be the only type of water justifying the expense of fertiliza- tion. The winterkills which followed heavy application of fertilizer to lakes in the same part of the state (Ball, 1948 and Ball and Tanner, 1951) are the most serious re- sults of fertilization to be avoided. The effect of the re- duced rates of application on the winter and summer oxygen supplies were carefully measured. Winterkill conditions were not produced by the low rates of application but some oxygen 172 reduction occurred and was related directly to the amount of fertilizer added. The highest rate of application very nearly resulted in winterkill conditions during the winter following the second summer of fertilization. The effects of fertilization are definitely accumulative from.year to year and applications should not be repeated until the effects of the first application can no longer be detected. Jduging from the one lake in the group which had been fertilized dur- ing 1946 and 1947 the duration of the increased biological activity and reduction of oxygen may be three or more years. Summer oxygen reduction occurred in the four fertilized lakes. By the end of the second summer little oxygenated water remained below the thermocline in the fertilized lakes. The oxygen reduction was not severe in the thermocline and it is doubtful if oxygen was a limiting factor during the summer stagnation period. The repositioning of the thermocline in a shallower poSition as a direct result of fertilization combined with the removal of oxygen from the hypolimnion reduced the amount of water available to the trout during the mid-summer period. The decrease in total hardness which apparently occurred as a result of the increased amounts of plankton present may be a possible solution to the management of some of the ex- cessively hard waters of the state. Increased amounts of plankton following application of fertilizer preceded general increases in biological activity. 1'73 ' Important increases in the bottom fauna occurred. The midges showed the greatest increase. The feeding habits of these "ooze browsers" made them quick to respond to in- creased amounts of plankton. The changes in the depth.dis- tribution of trout food organisms resulted from greater amounts of plankton. The organisms dwelling in the sub- littoral zone are dependent on the dead plankton from above for food. Prior to fertilization food from this source was limited. The least amount of fertilizer used resulted in important gains in the amount of trout food available but ine increase in food organisms was directly correlated with the amount of fertilizer added. Aside from.the effects of fertilization, interesting facts were gained concerning the brown trout in lakes. Con- trary to general opinion, the fishermen were able to capture a large percentage of brown trout present. Brown trout were not as vulnerable to fishing as brook trout but were taken more consistantly throughout the season. Brook trout, after the first month of the season, are usually taken only in deep water. During the 1950 season brown trout were present in the surface waters and were the source of much fine fish- ing throughout the season. It may have been that 1950 was an exceptionally cool season and further experiments should determine the answer. Because the brown trout attained a larger size than brook trout in the face of heavy fishing pres- 174 sure, they provided greater sport for the fishermen. It was apparent that the creel census station served to direct the fishermen to the better fishing and the possibility of in- creasing the return of hatchery fish should be considered. Further investigation of the use of brown trout for stocking lakes should include the use of fingerling fish instead of yearlings. Fish production was stimulated by the increase in food available following fertilization. Several factors, the re- moval of trash fish, the presence of minnows in two of the lakes, differences in availability of food organisms, and differences in potential productivity of the lakes, made it difficult to state with accuracy how much of the increase was due to fertilization. The presence of minnows in two of the lakes was shown to reduce the production of trout and to restrict trout growth. South Twin Lake which.was fertilized in 1946 and 1947 was observed throughout the study in order to evaluate the residual effects of fertilization. It was apparent that the high fish production for this lake during the second and third years following fertilization must be attributed to the effects of the fertilizer. The levels of the bottom fauna remained high.but the plankton levels showed definite signs of decreasing and the rates of summer and winter oxygen de- pletion were nearly back to the prefertilization condition by 1950. The large beds of potomogetons which developed 175 after the application of fertilizer showed no signs of de- cline as late as the summer of 1951. It must be concluded that, while other workers have indicated that the nutrient materials disappear from the economy of the lake by the year following fertilization (Einsele, 1938; Barrett, 1952) the biological effects do not subside as rapidly. It appears that the extent of the eutrophication occurring as a result of fertilization is considerable. Because of the eutrophy- ing effect and hastening of the extinction of the lake, fer- tilization cannot be justified in most instances. Hewever, it has been pointed out that there are lakes in this state that are suited for trout except that the amount of trout food present will not support enough.fish to provide inter- esting fishing when the lake is fished heavily. In these instances where fertilization creates a new source of fish- ing it should be considered as a practical and ethical ap- proach to management of these unproductive lakes. Two lakes, Section-Four and Lost, in the present study are lakes in this classification. The excellent fishing and high produc- tion in Section-Four Lake is proof that fertilization can create fishing in lakes of very low productivity. The excellent production of trout from the unfertilized control lake indicated quite clearly that removal of com- peting warm-water species and adequate stocking will provide excellent fishing from the three more productive lakes of the group. 176 The lakes of the present study were lakes of unusual ap- pearance and formation but the conditions effecting the pro- duction of fish are not unique and the results of the present study will apply on generally similar lakes. From the information now available from this and pre- ceding investigations, it appears that the only situation in Michigan where the use of fertilizer would be practical would be on small landlocked trout lakes subject to heavy fishing and with an extremely low fish.production. If a lake is low in fish and fish.food production and so typi- cally oligotrophic as to have oxygen present throughout most of the volume of the hypolimnion during the mid-summer stag- nation period then it would probably justify fertilization. Fertilizer should be applied at not more than a total of 10 p.p.me per year and should not be repeated until the stimu- lus of the added nutrients has dissipated. 177 SIEELABBT Inorganic fertilizer was applied to four of a group of six trout lakes and a fifth lake retained as a control. The sixth lake had been fertilized during the two sum- mers preceding the present investigation and observa- tions were continued to evaluate the residual effects. Low rates of application of fertilizer were used and the lake with the greatest total hardness received the most (a total of 15 p.p.m. during two summers) and the softer water lakes received lesser amounts. The low- est rate of application was 3.9 p.p.m. total for the two summers. The thermocline in three of the fertilized lakes of the group shifted to a shallower position during the sumr mers of 1949 and 1950. This change in the position of the thermocline did not occur in the two lakes remain- ing unfertilized during the present study nor in the lake fertilized at the lowest rate. Three of the fertilized lakes decreased in total hard- ness at a greater rate than did the two lakes which were not fertilized during 1949 and 1950. Fertilization resulted in the oxygen being depleted from the hypolimnion and limited reduction of oxygen in the 178 thermocline occurred. In the one lake of the group fertilized for two summers prior to the present inves- tigation the oxygen was depleted from the hypolimnion during the first summer following fertilization but not during the next two summers. The volume of oxygenated water remaining under the ice during the critical period of late winter was reduced in the fertilized lakes. The lake fertilized at the highest rate very nearly ap- proached winterkill conditions. The lakes treated at lower rates had some oxrgen reduction but winterkill was avoided by a wide margin. The reduction of oxygen both.in summer stagnation periods and in late winter was more severe after the second season of fertiliza- tion. In the one lake fertilized prior to the present study the amount of oxygenated water present beneath the ice increased the third winter following fertiliza- tion. Increases in plankton occurred in all the fertilized lakes and these increases were correlated with the amount of fertilizer added. Filamentous algae did not become a problem in any of the lakes. In the one lake where rooted aquatics were common the plants were largely eliminated when the turbidity of the water increased following fertilization. "1 Important increases in the bottom iauna were observed in 179 all the fertilized lakes. N increase was apparent in 1 the unfertilized lake. The amount of increase was closely correlated with amounts of fertilizer added. The depth distribution of trout food organisms changed dur- ing the period of fertilization with the volume of food organisms in the sub-littoral zone becoming more impor- tant. A similar change was not noted for th unferti- lized lake. In the lake which has been fertilized pre- viously the volume of bottom fauna organisms in the sub- littoral zone became less important each year after the cessation of fertilization. Stomach analysis of the brown trout showed a nearly com- plete dependence on the bottom fauna for food. In the two lakes of the group where forage fish were present they were not important as trout food. The volume of food in the stomachs of trout taken from the lakes with the highest production was much larger than the volume of food present in the stomachs of trout taken from the two least productive lakes. The quality of the fishing and the effort expended on the lakes increased during the period of fertilization. The yield in trout per acre was high in several of the lakes. The brown trout provided fishing of excellent quality during the entire season. There was evidence that angling is selective of the larger faster growing 179 brown trout. Probably considerable direction was given to the fishing effort towards the highest quality fish- q inH by the creel census checking station. (e The greatest trout production was from the lake which had been fertilized previously at a much higher rate of application than was used on any of the four lakes fer- 1 tilized during the present program. The unfertilized control lake had the next highest production of trout and the most rapid growth. The fish production capacity increased in the fertilized lakes. In the two lakes with populations of forage fish present the production was lower and the growth rates poorer than in any of he lakes with only trout present. 183 Adamstone, F. B. and W. J. K. Earkness 1925 The bottom organisms of Lake Nipigon. Univ Toronto Stud., Biol. Ser.; Pub. Ont. Fish Res. Lab. 15: 123—170 American Public Health Association 1946 Standard Methods for the Examination of Water and Sewage. Am. Pub. Health Assoc., New York. 9th. ed., 286 pp. Ball, Robert C. 1948 Relationship between available fish food, feeding habits of fish and total fish production in a Kichigan lake. Tech. Bull. 206, Michigan State College Agr. Exp. Sta. 1949 Experimental use of fertilizers in production of fish food organisms and fish. Tech. Bull. 210, Michigan State College Agr. Exp. Sta. ---- Standing crop of brown trout based on recovery of marked fish following poisonings of six trout lakes. (In preparation.) Ball, Robert C. and Don W. Hayne 1952 Effects of the removal of the fish population on the fish food organisms of a lake. Ecol. 55 (1): 41-48. Ball, Robert C. and Howard A. Tanner 1951 The biological effects of fertilizer on a warm- water lake. Tech. Bull. 225, Michigan State College Agr. Exp. Sta. Barrett, P. H. 1952 Effects of alkalinity on adsorption and regenera- tion of phosphorus in natural lakes. Thesis, Michigan State College. Unpublished. Baten, W. D. 1938 Elementary Mathematical Statistics. John Wiley and Sons, Inc., New York. 529 pp. Borutsky, E. V. 1959 Dynamics of the total benthic biomass in the pro- fundal of Lake Beloie. Proc. Kossino Limnol. Sta. Hydrometeorol. Ser. U.S.S.R. 22: 75-74. 181 Brujewica, S. W. 1959 Distribution and dynamics of living matter in the Caspian Sea. Comptes Rendus (Doklady) de l'Acad- emic des Sciences de l'U.R.S.S. 1959 Vol. XXV, No. 2. Chu, S. P. 1945 The influence of the mineral composition of the medium as the growth of planktonic algae. Part II The influence of the concentration of inorganic nitrogen and phosphate phosphorus. J. Ecol. Clarke, George L. 1946 Dynamics of production in a marine area. Ecol. Monogr. 16: 521-555. October, 1946. Comstock, John H. 1947 An Introduction to Entomology. Comstock Pub. Co., New York. 9th. ed., revised. Cooper, Edwin L. ---- Growth of brook trout (Salvelinus fontinalis) and brown trout (Salmo trutta) in the Pigeon River, Otsego Co., Michigan. (In preparation.) Deevey, Edward 5., Jr. 1941 The quantity and composition of the bottom fauna of thirty-six Connecticut and New York lakes. Ecol. Honogr. 11: 415-455. October, 1941. Dice, Lee R. and Harold J. Leraas 1956 A graphic method for comparing several sets of measurement. Contribution Lab. of Vert. Gen. Univ. of Mich. 5: 1-5, 1 fig. Drews, Robin A. 1951 The cultivation of food fish in China and Japan: A study disclosing the contrasting national patterns for rearing fish consistant with differ— ing cultural histories of China and Japan. Thesis. Univ. of Mich. Unpublished. Eggleton, E. E. 1951 A limnological study of the profundal bottom fauna of certain fresh-water lakes. Ecol. Monogr. 1: 251-552. 192 Einsele, Wilhelm 1958 Uber chemische und killoid-chemische Vorgange in Eisen-Phosphor-Systemen unter limnochemischen und limnogeologishcen Gesichtspunkten. Arch. Hydrobiol. 55: 561-587. 1941 Die Umsetzung von zugefuhrtem, organischen Phosphat im eutrophen See und ihre Ruckwirkung auf seinen Gesamthaushalt. Zeit. f. Fischerei 59 (5): 407-488. Eschmeyer, William R. 1957 Some characteristics of a population of stunded perch. Pap. mich. Acad. Sci., Arts, and Letters, 22 (1956): 615-628. 1958 Experimental management of a group of small Kichi- gan lakes. Trans. Am. Fish Soc. 67 (1957): 120-129 0 1950 Fish and Fishing in TVA Impoundments. Tenn. Dept. of Cons. Hasler, Arthur D. 1947 Eutrophication of lakes by domestic drainage. Ecol. 28 (4): 585—594. Hasler, A. D., O. M. Brynildson and wm. T. Helm 1951 Improving conditions for fish in brown-water bog lakes by alkalization. J. Wildl. Mgt. 15 (4): 547-552. October, 1951. Hasler, Arthur D. and Wilhelm Einsele 1948 Fertilization for increasing the productivity of natural inland waters. Trans. 15th N. A. Wildl. Conf. 1948. pp. 527-555. Hayes, F. R. 1951 On the theory of adding nutrients to lakes with the object of increasing trout production. Canadian Fish Culturist 10: 52-57. Hile, Ralph 1956 Age and growth of the cisco Leucichthys artedi (LeSueur) in the lakes of the northeastern High- lands, Wisconsin. Bull. U. S. Bur. Fish. 19: 211-517. Hogan, Joe 1955 EXperiments with commercial fertilizers in rearing largemouth black bass fingerlings. Trans. Am. Fish. Soc. 65 (1955): 110—119. 185 Howell, Henry H. 1942 Bottom organisms in fertilized and unfertilized fish ponds in Alabama. Trans. Am. Fish.Soc. 71 (1941): 165-179. Hubalt, E. 1945 Les GrandsIacs subalpins de Savoi sont-ils alcalitrophes? Arch. Hydrobiol. 4O (1): 240-249. Huber-Pestalozzi, G. 1958 Die Binnengewasser Band XVI, Das Phytoplanktor des Susswassers. Stutgart. Ivlev, V. S. 1945 The biological productivity of waters. Uspekhi Soveremennoi Biologii (Advances in Kodern Biology). 19 (1): 98-120. {usnetzow, S. J. 1959 Determination of the intensity of oxygen-absorp- tion in the lake water, caused by bacterial pro- cesses (Summary). Proc. Kossine Limnol. Sta. Hydrometeorol. Serv. of U.S.S.R. 22: 75-74. Lindeman, Raymond L. 1942 The trophic-dynamic aspect of ecology. Ecol. 25 (4): 599-418. Meehean, 0. Lloyd 1955 The role of fertilizers in pondfish culture. Trans. Am. Fish. Soc. 65 (1952): 105-109. 1956 Some factors controlling largemouth.bass produc- tion. U. S. Bur. Fish., Prog. Fish-Cult., 16: 1'6 0 Minder, Leo 1922 Uber biogene Enthalkung im Zurichsee. Verh. Insternat. Vereinig. Limnol. 1922: 20-52. Mortimer, C. H. 1941 The exchange of dissolved substances between mud and water in lakes. I and II. J. Ecol. 29 (2): 280-529. Neess, John C. 1949 Development and status of pond fertilization in Central Europe. Trans. Am. Fish. Soc. 76 (1946): 555-558. Patriarche, Mercer H. and Robert C. Ball 1949 An analysis of the bottom fauna production in fer- tilized and unfertilized ponds and its utiliza- tion by young-of-the-year fish. Tech. Bull. 207, Hichigan State College Agr. Exp. Sta. Rawson, D. S. 1947 Deterioration of recently established trout popu- lations in lakes of the Canadian Rockies. Canadian Fish Culturist. 2. 14. Ricker, William E. 1946 Production and utilization of fish population. Ecol. Honogr. 16: 575-591. October, 1946. Schaperclaus, Wilhelm 1955 Textbook of Pond Culture. Book Publishing House, Paul Parey, Berlin. Translated by Frederick Hund. Scott, I. D. 1921 Inland lakes of.Hichigan. Mich. G801. and Biol. Survey, Publ. 50, Geol. Ser. 25, xxi + 585. Smith, M. W. 1945 Preliminary observations upon the fertilization of Crecy Lake, New Brunswick. Trans. .1. Fish. 800., 1945, 75: 165-175. Snedecor, G. W. 1946 Statistical methods. Iowa State College Press, Ames Iowa. xvi + 485 pp. Surber, E. W. 1945 The effects of various fertilizers on plant growths and their probable influence on the production of smallmouth blackbass in hard water ponds. Trans. Am. Fish. Soc. 75 (1945): 577-595. Swingle, H. S. and E. V. Smith 1959 Fertilizers for increasing the natural food for fishes in ponds. Trans. Am. Fish. Soc. 68 (1958): 126-155. Utterback, Clinton C. 1946 The penetration and scattering of solar and sky radiation in natural water bodies of the Pacific Northwest. A Symposium on Hydrobiology. Wisc. Univ. Press. pp. 45-58. ‘ Van Deusen, R. D. 1947 Quantitative and qualitative evaluation of plank- ton from fertilized and non-fertilized hatchery ponds, with an appraisal of methods used. Thesis, Michigan State College. Unpublished. Veatch, J. O. 1952 Some relationships between water plants and water soils. Tran. Mich. Acad. of Sci., Arts and Letters. 17: 409-415. Vialker , 310 L: o 1945 Elementary Statistical Kethods. Henry Holt and Co., Nev} York, No Yo, 568 pp. Ward, H. W. and G. C. Whipple 1918 Fresh-water Biology. John Wiley and Sons, Inc. New York. 1111 pp. Wilkins, L. Price 1952 Relationships of a fish population to the inver- tebrate fauna in two small ponds. Thesis, Richi- gan State College. Unpublished. 186 APPEfl‘i'DDI SUKHARY F THE STATISTICAL ANALYSES OF CHAHGES IN TH“ POSITION OF THE THERHOCLIHE (ANALYSIS OF VARIANCE) Unfertilized Fertilized South. North West Sec. Source Twin Twin Lost Hemlock Lost Four Total Y6 3.1" S — -- _ 55.x. _ 3‘ 1948 vs 1949 and 1950 - - — fifi * % among 1949 and 1950 - - - - — % Months 6:— »::- a. '21-:- at .. Error term - Significant (5 per cent level). i - Highly significant (1 per cent level). zit :3: THERMOCLINE CHANGES IN SOUTH TWIN LAKE (ANALYSIS OF VARIANCE) Source D.F. s.s. M.S. "F" Total 11 441.1 Years _ ‘ 2 76.1 58.05 2.15- 1949 vs 1949 and 1950 (1) 'among 1949 and 1950 (1) Months 3 257.9 95.97 4.92“ Error 6 107.1 17.95 - No significance. * Significant (5 percent level). THERMOCLINE CHANGES IN NORTH TWIN LAKE (ANALYSIS OF VARIANCE) Source D.F. 3.8. M.S. "F" Total 11 717.5 Years 2 24.4 12.2 .44- 1948 vs 1949 and 1950 (1) among 1949 I and 1950 (1) Months 3 549.9 185.5 6.60* Error 6 167.6 27.9 - No significance. 4 Significant (5 percent level). SUTTIIARY OF AITALYS IS OF VARIANCE 110R SITEI. IER OXYGEN DEPLETION Unfertilized Fertilized South North Source Twin Twin Sec. Hemlock Lost Four Total Between periods (prefert. vs fertilization) - - Within (error) \‘.‘ f \ \b 1" I J; I“ \ o a s ' - No significance. 69% Highly significant (1 percent level). .pqooauaamao Aegean . 9* Ia ma.bm mo nomad 4 _ noxoa gonna .aaoo . 4 aoaam _ g boa. Ho.m no noaao . Ho.nm «.msa o.oao.m we manna: «aaoo _ Hogan . atema.sm one.» mm.ma m nosed . os.on «.mm m.ose m ooaoq .9900 H ” Ho.¢m m.oom w.ame.m om Hence . . . «a am «a - A . a: gag .m.s .m.m .a.n oonsom H .m.m .m.n ooaoom wozeHmesoo ac mHmau¢z< .aoooav.mmuaq me may mo moan scam oZquz< am seamen spams mo maezmn moemm>¢ fl and. my AH emgma 0AA». "MO ehdg I aHOeHoHo on.” u Shop noaao no mm» coma e neoa no .ao> .moxwa Nan Had you 0H5» w u a taco no asap no owmaopc Hwaocow aux A x xvn Hm. .. |.|. .9909» no npmcoa omeaopd 9.» A.odndp concaadb .Anpom I and .I :00 on» :u anon noaao scan exoobv onapnoo no 036» omcaopd u.m comes an m one an m opozv mu m u n .anoapobaouno hdxooa no nonanz - 2 NM m unowUQH can enonao Unwcndpu on» wands .an onowaa ea eoppoaa can mean omen mode you npwnoa sees copuanoc one I # mm Hoooa «Ha. mmao. mmm.m rod. n Ham.a + mna.oa 40¢.HH Ha ease nuaoz mmo. Hmoo. nmo.m moo. a 0H0. + me.m nnm.a ma data mason and. HSHO. oem.m nan. + nwo.¢ a nam.m . oom.m m . aoOHsem mwo. whoo. mma.m 040. n bmv. + mmm.o HHH.0H ma Anomnqoapoom mud. mmao. omn.m mun. + mm>.n I mmm.m mbb.m a puqq and. HSHO. woo.m 00¢. a mmm.¢ + mow.m mmH.¢H m noon new; .95 a 5.9 ...a in em: owm . m m z .3 coacand>_ oopmmnnd. moado Coma 20mmvmmHm mo mamzmq zHmmn .HH nopaopmom open wsamo .mm Haamd opwc mqano m1.” m.m I m.wm m.m I m.mm o.mm m.noa worm. atom o.m I H.0m m.m I m.a> o.¢b m.nbm worm. ma m.Hm I b.m¢ m.¢o I o.omd >.oo m.m¢m worm. ma o.mH I H.m¢ o.md I m.om m.mn o.omH worm. SH m.¢ I n.an o.HH I m.mm m.nv 0.05m worm. 0H H.> I 0.0m h.mH I o.¢¢ b.mm 0.0mm menu. ma H.>m I 0.0 H.¢H I 0.0 H.¢H o.mm webm. ea «.9 I o.am o.H I o.w a.a m.wn worm. ma n.0H I o.oa n.0m I o.>a n.n¢ o.omH worm. NH o.wa I H.m m.m I o.m m.¢H 0.0m worm. AH m.>H I a.o m.¢n I 0.0m m.¢m m.Hom worm. 0H o.nw I H.o> o.nm I m.>n m.¢a m.nm poem. 0 o.mH I b.o¢ m.m I o.am «.ma o.m¢ worm. m m.m I n.Hm >.¢ I m.>a «.mm o.mm worm. b n.n I m.nm m.m I o.mH m.am 0.0m norm. 6 m.¢ I n.mm o.mN I n.mma H.mma o.mwm worm. m m.mH I n.¢H 0.0m I o.nn m.mw m.Hnm worm. w «.w I m.dn 0.0H I m.mo n.mm 0.0mm norm. 9 o.> I >.¢n 0.0» I m.HmH o.mHH o.bn¢ nova. m o.a I H.mm H.@ I m.nna w.mma m.¢bw worm. aH usooaoa uaouno Ho mason cu paouue anouuo Meet new paonmo Hoodoo oopoemuo anooaon concaomuan donned no case: momma Add annoy me on» no scan pace defined concomum no anon uaooaom exec; Iaom dd Inc Hopes cocoaomnan g ommd .MHZH ZHBB meom .Bzmommm QMBommNM 20mm Emommm wZHmmHm 2H “azmcmmm-ZHv mozmmmthQ mo chadflbondo .3” 90982300 09.0.0 mfimodo I 0* .00 Hunmd e000 waaaono I t 0.0a I 0.0 0.0a I 0.0 0.00 0.00H 0000. #000 0.0a I 0.0a 6.00 I 0.00 0.00 0.000 0000. 0a 0.H0 I H.0 0.00 I 0.0a v.00 0.000 0000. 0a 0.HH I 0.00 0.0a I 0.0a 0.00 0.00H 0000. SH H.d0 I 0.0 0.00 I 0.0a 0.00 0.000 0000. 0a H.0H I 0.0 0.00 I 0.0a H.00 0.000 0000. 0H 0.0a I 0.00 0.0 I 0.00 0.0a 0.00 0000. 0H 0.0 I 0.0a 0.0 I 0.0 0.0 0.00 0000. 0a 0.00 I 0.00 0.H0 I 0.00H v.00 0.00H 0000. 0a 0.00 I 0.00 0.00 I 0.00 0.0a 0.00 0000. HH 0.00 I 0.00 0.00 I 0.0HH 0.H0 0.H00 0000. oa 0.0a I 0.HH 0.0 I 0.0 0.0a 0.00 0000. 0 0.0 I 0.00 0.0 I 0.0 0.HH 0.00 0000. 0 0.00 I 0.0 0.00 I 0.0 0.00 0.00 0000. b 0.00 I 0.00 0.00 I 0.00 0.00 0.00 0000. 0 0. I 0.00 0.0 I 0.00H 0.00H 0.000 0000. 0 b. I 0.00 0.H I 0.00 0.00 0.H00 0000. 0 0.0 I 0.00 0.0 I 0.00 0.00 0.000 0000. 0 0.H I 0.00 0.0 I 0.0HH 0.HHH 0.000 0000. 0 0.0 I 0.H0 0.00 I 0.00H 0.00H 0.000 0000. 0H paooaen naonno no unuon ca anonno phonno Moot Hon phonno nouaon concomxe paooaon cocoaonnan Manned no mason named Add Hauop no on» no scan psoo dawned oopooaxm no anon unconom axe»: Inca :0 Inc Hauoa oonoaonnnm 003 .053 080-2355 .2805 003058 22E $an 020103 E 3200000 23 062000.030 no 2033033 .HH sonaoaaom opoo manooso I II .00 Hasad oooe waaooao I I H.0H I 0.0 0.0H I 0.0 0.0H 0.00H 000H. .0I00 0.HH I 0.0 0.00 I 0.0 0.H0 0.000 000H. 0H H.0H I 0.0 0.00 I 0.0 0.00 0.000 000H. 0H H.0H I 0.0 H.0H I 0.0 H.0H 0.00H 000H. 0H H.0 I 0.0 0.0H I 0.00 H.H0 0.000 000H. 0H 0.0 I 0.0 b.0H I 0.0H 0.00 0.000 000H. 0H H.0 I 0.00 0.0 I 0.0H 0.0 0.00 000H. 0H 0.0 I 0.0 0.0 I 0.0 0.0 0.00 000H. 0H 0.0 I 0.0 H.0H I 0.0 H.00 0.00H 000H. 0H H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. HH 0.0 I 0.HH 0.5 I 0.00 0.00 0.H00 000H. 0H 0.0 I p.0H 0.H I 0.0H H.0 0.00 000H. 0 0.0H I 0.00 0.0 I 0.0H 0.0 0.00 000H. 0 0.H0 I 0.00 H.00 I 0.00 0.0H 0.00 000H. 0 0.0H I 0.H 0.0H I 0.H H.0H 0.00 000H. 0 0.0H I H.00 H.00 I 0.00H 0.00 0.000 000H. 0 0.0H I 0.0 0.00 I 0.0H 0.00 0.H00 000H. 0 0.0 I H.0H 0.0 I 0.00 0.00 0.000 000H. 0 H.> I 0.0 0.00 I 0.00. 0.00 0.000 000H. 0 0.00 I 0.00 0.00 I 0.0bH 0.H> 0.000 000H. 0H unoosom psonno no asses :H psonno anonno Moe! son psonne acumen 00000080 paoosea oesonennnn Hdspod no mason ueMaH HHd Hanan no on» no scan paoo Heapod dopoemnm do anon anoosem exec; Inon CH Ino H0909 eonenenan omoH .muaq smog .szmommm nmsommxm sous seesaw ezsmmss 2H Aszmommm zsv mozmmmsssn so zossaqpogao .HH sonaoaaom opus wanooHo I II .00 HanoI opoe maneoao I 0.0 0.0 H.0 0.0 H.0 0.00H 0000. $000 0.0 0.H 0.0H 0.0 0.H0 0.000 0000. 0H 0.0 0.0 0.0H 0.0 0.0H 0.000 0000. 0H 0.0 0.0 0.0 0.0 0.0 0.00H 0000. 0H 0.0 H.H 0.0H. 0.0 0.H0 0.000 0000. 0H 0.0 0.0 0.0 0.0H 0.0H 0.000 0000. 0H 0.0 I 0.0 H.0 I 0.0 H.0 0.00 0000. 0H 0.0 I 0.0 0.0 I 0.0 0.0 0.00 0000. 0H 0.0 I H.0H 0.0 I 0.H0 0.0H 0.00H 0000. 0H 0.0 I 0.0 0.0 I 0.0 0.0 0.00 0000. HH H.0 I 0.0H 0.0H I 0.00 0.0H 0.H00 0000. 0H 0.0 I 0.0 0.0 I 0.0 0.0 0.00 0000. 0 0.0 I 0.0 0.0 I 0.0 0.0 0.00 0000. 0 0.0 I 0.0H 0.0 I 0.0H 0.0 0.00 0000. 0 0.0 I 0.0 0.0 I 0.0 0.0 0.00 0000. 0 0.0 I 0.0H 0.00 I 0.00 0.00 0.000 0000. 0 H.00 I 0.00 0.00 I 0.00 H.0H 0.H00 0000. 0 0. I 0.0 0. I 0.0H 0.0H 0.000 0000. 0 0.0 I H.0H 0.0 I 0.00 H.00 0.000 0000. 0 0.0 I 0.H 0.00 I 0.0 0.00 0.000 0000. #H paoosom _psonne no mason GH asonne anonno Moot son asonne condom concomNo accosom ooaosonnnn H05004 no mason qudH HHd Hanan no on» no Bonn pace Hufipo< concomxm do anon unconom axon; Inca nH Ino Haaoa oonosonnnn 000H amudfl MUOASHM gazmommm nmaommxfi 30mm amommm oszmHm 2H Aezmommm ZHV mozmmmmmHQ mo ZOHadflbondo .HH nonempmom 0000 hHomoHo I 0* .00 HHnmd 0000 wnnaomo I * 0.00 I 0.00 0.00 I 0.00 0.0H 0.00H 000H. #000 0.0H I 0.00 0.00 I 0.00 0.00 0.000 000H. 0H 0.0H I H.00 0.00 I 0.00 0.00 0.000 000H. 0H 0.0 I 0.0H 0.0 I 0.0H H.0H 0.00H 000H. 0H 0.00 I 0.00 0.00 I 0.H0 0.00 0.000 000H. 0H 0.00 I 0.00 0.H0 I 0.00 0.00 0.000 000H. 0H 0.0 I 0.0 0.0 I 0.0 0.0 0.00 000H. 0H H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0H H.0H I 0.0 H.0H I 0.0 H.0H 0.00H 000H. 0H H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. HH 0.0 I 0.0 0.0H I 0.0 0.00 0.H00 000H. 0H H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0 H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0 H.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0 H.0H I 0.0 H.0 I 0.0 H.0 0.00 000H. 0 0.0 I 0. 0.00 I 0.0 0.00 0.000 000H. 0 0.0 I H.H 0.00 I 0.0 0.00 0.H00 000H. 0 H.0H I 0.0 H.00 I 0.0 H.00 0.000 000H. 0 0.H I 0.0 H.0 I 0.00 H.00 0.000 000H. 0 0.0 I 0. 0.00 I 0.0 0.00 0.000 000H. 0H pooosoq anonno no mason SH psonno psonno Moo: non psonno deacon vopooano pnoosom eoqosonnnm Hafipod no onion moon HHo prop no on» no scan pnoo Hdsuod 00000000 no anon unconom .0003 I900 :H Ino Hopoe oonosonnnn 000s .0000 smog sow: .szmommm oesommxm sons smosmm oszmHs 2H Aszmommm zsv mozmmmssno so zossaqooqeo .HH 009800000 0000 msHaoHo I 0* .00 H0004 oooo monsooo I I 0.0H I 0.00 0.0H I 0.00 0.0H 0.00H 000H. ##00 0.0H I H.Hn 0.00 I 0.00 0.00 0.000 000H. 0H 0.0H I I0.00 0.00 I 0.00 0.00 0.000 000H. 0H 0.0 I 0.00 H.HH I 0.00 0.0H 0.00H 000H. 0H 0.0 I 0.00 0.0H I 0.00 0.00 0.000 000H. 0H 0.0 I 0.00 0.00 I 0.00 0.00 0.000 000H. 0H H.HN I 0.00 0.HH I 0.0H 0.0 0.00 000H. 0H 0.00 I 0.00 0.0H I 0.0H 0.0 0.00 000H. 0H 0.0 I 0.0 0.0H I 0.HH 0.00 0.00H O00H. NH 0.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. HH 0.0H I 0.0 0.00 I 0.0 0.00 0.Hom 000H. OH 0.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0 0.0H I 0.0 0.0 I 0.0 0.0 0.00 000H. 0 H.0H I 0.0 0.0 I 0.0 0.HH 0.00 000H. 0 0.0H I 0.0 0.HH I 0.0 0.HH 0.00 000H. 0 0.0 I 0.0H 0.00 I 0.00 0.H0 0.000 000H. 0 H.0 I 0.H0 0.0H I 0.00 0.00 0.H00 000H. 0 NIH I 0.0H 0.0 I 0.00 0.H0 0.000 000H. 0 0.0 I 0.HH 0.HH I 0.00 0.00 0.000 000H. 0 0.0H I 0.0 0.00 I 0.0H 0.00 0.000 000H. 0H FQQOHOQ HHOHHO HO flHdeH Gd PROHHO pHOMQO MOO. HO“ ”ROHHO HHOIdOn copoemuo paoosom oonononan Hdspod ,no 00500 nowa HHo Hdpop no on» no Bonn unoo Hddnod 00000000 no anon 0000000 .0003 I000 0H Ino Hmpoe oonosennnn é omoa .0000 zsss 0smoz .sz0o000 o0so0000 soms s0o000 ozsmmHm zH Aszmo000 zsv 0020000000 0o 2oss¢noo0¢o