t ._ . .- ... . . . ., . . ,. . ... .2 . _. ‘OHMQ. I V . ._ . J _ ..; .xw.._.~.wr«ww ... . A at? ... ‘ o. p ‘ v .ng n 1 a: . x . wk?” yrw . . , . _A .. . . V_.— .I. ., .A. . . ‘ . t _ . _ A . . L .. . . . _ ... A - fl :.r, . . H .. . _ .¢ \v —1 . 1 “cl. v. 3. f I ~ t’. r u w! A. , -u . . I... ....r - - .‘._ .... ..Ufu‘fllwutuw.g CI..u.-;.nUflllul. H, ‘ . 7.,1. I w.v§Ilp. cu."~'!v|’b'|1hf I . . 2 0.3... 3-3 , ;::1 . A .‘ . fl 1 4 . . . I. ..|- II . t . r v 3 ...!!HI 9 ,N ‘ Ill, IoLuul-Jafi.‘ . . Q c | til: {UL V flat. [5' ' \ \ Wu ‘ r’u‘IJtr’W- ”.1 . . n ‘ ,. . z ,, l ‘ . . , , , .r 4 .4 I l.’ ‘ 9 I ’3, . ‘ l . . .. . .. 341:»: . J1 . ... L L . T. q A...‘~), I nt. .. (45'. $7.: . . I-‘ . 3- I.» L ., l. J? .L..- wi- -.l 1.15% -.-. . 811...- LP “$qu $6 mum? . 31.1.4”. I d3flnéWn...» n. 1.. u... t w c. \l‘ <‘lv‘...ooH ..., .---.«vfivltnx. .v.ox ... INN: “Handmaix .fir»~fi«.. . . p "J”.aVIQHJ’ucRIfl-v lwv’érif:l|lnjain4 ...? dirk. t . a g o. . ‘ . 't' J'.'. This is to certify that the thesis entitled A POpulation Study of the Cottontail Rabbit in Southern Michigan presented by Aelred D. 0815 has been accepted towards fulfillment of the requirements for Ph. D degree in Fisherie§ and Wildlife /.‘/:.‘_C /2 /!-(..’ [Z1 Georg A. Petrides Major professor Date_/~/ / 7: / 7 +st 1. 0-169 ‘ um mm m l1 1| mm Ln! ll! 1111111 1111 W121“ 3 1293 A POPULATION STUDY OF THE COTTONTAIL RABBIT IN SOUTHERN MICHIGéN By Aelred Dean Geis A THESIS Submitted to the College of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 1956 Pt ...8 g‘i" an: 9; 33‘: 1'11? 1933 .O semi to 1, ‘ U‘nu' Q u...;_ an 5.»... D .. J. I 32d in irfi‘fidct: "le 2C 10;? i: .97.: ..- a te- , , ",c"_/ 5/ ACKNOWLEDGMENTS Many people assisted the writer during this study. Grateful acknowledgment is made to Dr. George A. Petrides, Associate Professor, Department of Fisheries and'Wildlife, Michigan State University for his guidance; and to Dr. Don W. Hayne, Zoology Department, Michigan State University, for much help with statistical matters. Mr. Walter Lemmien, forester in charge of the Kellogg Forest; and Dr. Arthur E, Staebler and Mr, Roswell Van Deusen, former and current directors of the Kellogg Bird Sanctuary reSpectively, provided cooperation that was essential to the completion of this study. Many former and current Michigan State students aided in many ways. Special thanks are due Samuel.M. Carney, Donald.Moore and.Rainer Brooke who assisted by checking traps. Financial assistance was obtained from Michigan State University in the form of research assistantships in the Department of Fisheries and Wildlife during the 1951-52, 1952-53 and l95h-55 school years; and in instructorships in the Departments of Fisheries and Wildlife and Zoology during l9S3-Sh and in the Zoology Department during 1955-56. The aid of the Wildlife Management Institute is gratefully acknowledged for providing funds for labor. *%%%%%**5 ii A POPULATION STUDY OF THE COTTONTAIL RABBIT IN SOUTHERN MICHIGAN By Aelred Dean Geis AN ABSTRACT Submitted to the College of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife Year 1956 x a 9 _ Approved /‘7 2f 4 gig»: WA .‘LB ST WT Cottontail rabbit (Sylvilagus floridanus) populations on two 500- acre areas, the Kellogg Bird Sanctuary and Farm,and the Kellogg Forest, were studied between June 1951 and April 1956. The areas are properties of Michigan State University and are located near Battle Creek, Michigan, Major objectives were: 1. Evaluate the accuracy of available census methods. 2. Appraise the effects of factors influencing rabbit abundance, especially hunting. 3. Analyze hunting pressure as it occured on a public area, Live trapping was conducted almost constantly during the summer of 1951 on the Sanctuary study area, Trapping during the fall was conducted on both areas. At the SanCtuary Spring trapping was conducted during 1952— 1956. At the Forest, Spring trapping took place only in 1955 and 1956. Trapped rabbits were marked with ear tags and in some cases also their tails were dyed yellow; Hunting was the chief means by which rabbits were collected. An evaluation of census procedures indicated that the Lincoln Index method was reliable where rabbits were trapped and marked and then the marked fraction of the population determined by some other method than retrapping. Either shooting a sample or observing the tail colors was suitable. The De Lury method, based on the decrease in the hunting yield as the population was reduced, could not be applied at the Forest because the rate of kill did not decrease during the season. At the Sanctuary, the rate of kill decreased during the short and intensive iv hunting season, but the method consistently gave results that were about 50 percent low. Several indexes of rabbit abundance were compared to Lincoln Index population estimates. The hunting kill made during a constant effort, and the total hunting kill appeared to be reliable population indexes. The rate of kill based on the total kill and effort was unreliable when hunting effort varied from year to year. Trapping data examined in various ways provided rough indexes of abundance at the Sanctuary, but showed very little relationship to population density at the Forest. Age ratios were reliable fall population indexes only during the few years when spring population levels were constant, Fall population fluctuations at the Sanctuary between 1932 and 1955 and at the Forest for l9h0 and for l9h6-l955 were indicated by the hunting kills during fairly constant hunting pressures. No marked change in rabbit productivity has apparently occurred on either area during the periods considered. This is noteworthy since marked changes took place in the vegetation and predator populations on the two study areas. On both study areas, autumn juvenile: adult ratios were inversely proportional to spring adult population levels. At the Sanctuary, the size of Spring populations depended largely on the extent of the previous winter's hunting kill. it the Forest, the hunting kill took fairly constant percentage of the population and the size of the Spring population was apparently related to the previous fall's population level. Wide variation occurred in rates of pOpulation increase between Spring and fall. High rates seemed to be associated with a better survival of early litters and a higher incidence of breeding among juveniles. . 'During the warm months an adult mortality of up to about 60 per- cent occurred. This not only influenced fall population levels but also caused fall age ratios to be exaggerated indexes of breeding success. At the Forest the hunting bag for the 1951 season was h9 percent of the estimated population. During the other years it varied from 56 to 61 percent, At the Sanctuary the hunting kills ranged from 22 to 66 percent of the estimated fall populations depending upon the hunting pressure exerted. The hunting kill influenced the following spring population levels but apparently had little erreet on the following fall populations. Experiments were conducted to measure the crippling loss. Under experimental hunting conditions at the Sanctuary, it was roughly 10 percent of the recovered kill. With public hunting at the Forest it was about 20 percent. . The non-hunting winter mortality at the Sanctuary was estimated at from 29 to Sh percent of the rabbits not killed by hunting. It appeared to be higher during years which had a low hunting kill. The cause of this mortality was unknown. The fall age composition at the Sanctuary over a five year period was 82 percent juveniles, 1h percent one and one-half years of age, and only four percent two and one-half or more years old. Heavy hunting vi kills apparently decreased the rate of survival indicated by the age composition. Intensive live trapping was conducted between June 1951 and March 1952 and a life equation was determined for that period. Because rabbits are a very important game animal the character- istics and effects of heavy public hunting were studied on the Kellogg Forest. Data were gathered when hunters checked out of the area after each hunt. Yearly, monthly, weekly, daily and hourly patterns of hunt- ing effort, kill and rate of kill were determined. Hunting effort was high-at the beginning d? the season, rapidly tapered off, reached a second peak in early December and dwindled until it was uniformly low during the last four weeks of the season. The extent of the kill followed the trends in effort, except that it was relatively higher during late November and Decemba~ when the rate of kill was higher. In general, the rate of kill early in the day was slightly higher than that later. Hunting effort was usually greatest about 11:00 A.M. {The entire kill was made by 19.6 percent of the hunters. Six and eight-tenths percent harvested 65.8 percent of the total kill. If there was a uniform probability of success it would be expected that 7.1 percent of the hunters would take SO.h percent of the total bag, Significant differences were found among hunters in the rate of kill. The distribution of effort among the hunters indicated that 85 percent hunted a total of less than six and one-half hours and 65 percent visited the area only once. vii The influence of various weather factors on hunting effort was examined. Clear October days, cool November days, warm December days, calm days every month and November days with snow cover all seemed to induce an increased hunting effort. Rain fall caused a decrease in effort. Hunters that had had previous experience with the area, who used dogs and/or who hunted on days with snow cover were much more successful than hunters without these benefits. Single hunters had a relatively higher rate of kill than did groups. Increased hunting success prevailed when there was snow cover, on cool days in October and on warm days in December. The rate of kill at the Forest did not decrease late in the season due to an increase in effort by experienced hunters, to an increased prOportions of hunters that used dogs and a greater number of days with 3 now cover .' (The probable effects of hunting seasons of various lengths were calculated, Lengthening the both ends of the current October 20 to January 31 season apparently would have little effect on hunting effort, kill or success. Great changes in vegetation and faunas have occurred since Allen (Ecol. Mon., 8:3h?~h36. 1938) studied the Sanctuary area during l93h- 1937. These are discussed in detail. Relative rabbit winter food preferences and availability of the various woody plant species on the area were listed. viii TABLE OF CONTENTS INTRODUCTIONOOOOOOOO......OOOOOO...0.000....00000.000.000.000... Description of the Study Areas............................... Kellogg Bird SaDCtuaI'y and FamOOOOOOOO..OOOOOIOOOIOOOOOOO KelloggForeStO......OOOOOOOOOOOOOOOOO.......OOOOOOOOOOOO. G-ENmf-tL PRmEDIJRESOOOO00.000.000.00.......OOOOOOOOOOOOOOOO...... Trapping Periods and Locations............................ Types of TrapS............................................ Handling MethodS.......................................... Sex and Age Determination................................. ‘Marking MethodS........................................... HarvestingiMethodS........................................ POPIMTION DYNmICSOOOO0.000.000.00000.00.0.0...0.00.00.00.00... Population Measurements...................................... Tagged: Untagged Ratio EStimates (Lincoln Index)......... Hunting ResultS........................................... Indexes of Abundance......................................... Hunting Kill StatistiCS................................... Trapping Data............................................. Age Ratios-000.000000000000000000000000000.0000...00.00000 POPUIATION FLUCTUATIONSOOOO..OOOOOOOOOOOOOCOO......OOOOOOOOOOOOO Kellogg Bird Sanctuary and Farm Study.Area,.,............. Kellogg Forest Study Area................................. Population Increase and Mortality Analysis................... .Annual Population Increase................................ Adult Summer Mortality".................................. Hunting Mortality......................................... Crippling Loss............................................ Non-hunting Winter Mortality.............................. Age Composition of the Sanctuary Rabbit Population........ 1951-52 Life Equation Observations........................ Page h] HUNTING EFFORT AND SUCCESS AT THE KELLOGG FOREST................ 103 I. IntrOduCtionOOOOO0.0.0................OOOOOOOOIOICO0......103 II. MethOdSOOOCOOOC00.0.0000.........OOOOOOOOOO.....ICOCOOOOOO103 ix nf‘ ”‘1': T v"-.. -.L—c-fi“ ‘ O, ' ‘0- ' r4133. .- o”. O "' A, - 2 V .4. 0" I“ .“ U. . c- .1 U. '. "‘ L b, .. "7 Dis- J . -'. Ur.-. ' 2 :0, .4 3 Z and. a; r- r: v. u “A. ‘60—... V. “‘01-’4 T H V“ I :2er O - “a 3. .‘ Z ’. H. a. n -‘ U. y '\ ’- ‘J. .54; E “ . -4 9-? .r’- ~ “~- Lug-3;- --- “ . r0:3r_-- ‘ 9 six" .....""~ ‘ gal-...:J- . Cza’wngh :' “:‘J H h< for: ~ Foxy: Fl ' ‘O'vj ... vv- , " ' - Mag-1r?“ TABLE OF CCWTENTS - Continued Page III. Patterns of Hunting Effort, Kill and Success.....,......... llO A.YearlYOOCOOOOOOOOOOOO0.00.0.0...OOOOOOOOOOOOOOOOOOOOO110 B. Montrfl'YOOOOOOOOOOOOOOOOO............OOOOOOIOOOO.IO... 112 COweeld'yOOOOOOOOOOOOOIOOOO......OOOIOOI.........OOOOOOO11h D. DailyOOOOOCOOOOOOOOOOOOO......OOOOOOOOOOOOOOO...OD...119 E. HourlYOOOOOOOOOOOOOOOOOOOCOOO......OOOOOOOOOOOOC.0...121 IV. Distribution of Effort, Kill and Success Among the Hunters. 12h A. EffortOOOOI00.000.00.000...00.0.0000...00.00.0000...O12h BOBagOOOOOOO......OOOOOOOO0.0.0000.........OOOOOOOOOOOO129 CO succeSSOOOO......OOOOOOOOOO0.0.0.0..........OOOOCOOOO129 V. EffSCt of weather on I“holrlt'j-rlg Effortoooooooooooooooo00000009 135 VI. FaCtors AffeCting Hunting successoooooo00000000000000.0000. 1’43 A. Previous Experienco'With Area........................ 1&3 B. Hunter Occupation.................................... lhb 0. Dog Use.............................................. 1&6 D. Party Size........................................... 150 E. Climatic Factors..................................... 151 VII. Analysis of Factors Affecting'Weekly Hunting Success....... 152 VIII. Probable Effects of Seasons of Different Lengths,,,,,...,.. 167 MI$EmeS OBsmvj“TIONSOOOOO00......O......OOOOOOOOOOOOOOCOOO 171 Changes in.Mammal Numbers on the Kellogg Bird Sanctuary and Farm Between 19314-35 am 1951-55........OOOOOOOOOOOOOOOIOO 171 FoxFoodIiabitsObservations-ooooooooooooooooooooooooocococoo173 Winter Food Habits of Kellogg Bird Sanctuary and Farm COttontaiISOOOCO.......OOOOOOCOO00.0.00...00.00.000.000... 176 LITmATIJRE CITEOOOOOOO.O.......O......OOOOO......OOOOOOOOOOOOOO 182 TABLE 1. 2. 3. h. S. 9. 10. ll. 12. 13. LIST OF TABLES Cover Type Composition at the Kellogg Station, Hickory Corners, MiChigan....O..O...00............OCOOCOOOOOOOOOOO Summer 1951 Trapline Summary Kellogg Bird Sanctuary and Farm, HiCkOU Corners, PliChigan.......OIOCOOCOOOOOCO...... Late Summer and Early Fall Trapline Summaries, Kellogg jBird Sanctuary and Farm, Hickory Corners, Michigan....,... Spring Trap Line Summary--Kellogg Station, Hickory corners, MicmgaHOOOOOOOOOOC00.0.0000.........OOOOOOOO.... Prehunting Season Trapping Summary, Kellogg Bird Sanctuary andearm, Hickory Corners, Michigan....................... Prehunting Season Trapping Summary-~Kellogg Forest, ‘Augusta’ MiChigaHOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO...... Summary of Rabbit Hunts, Kellogg Bird Sanctuary and Farm-— 1951-1952, Hickory CornerS, Michigan...................... Summary of Rabbit Hunts at the Kellogg Bird Sanctuary and Farm--l952-l953, Hickory Corners, Michigan................ Summary of Rabbit Hunts, Kellogg Bird Sanctuary and Farm-— 1953-l9Sh, Hickory CornerS, Michigan...................... Summary of Rabbit Hunts at the Kellogg Bird Sanctuary and Farm--l95b and 1955-56, Hickory Corners, Michigan..,..,... Conditions Under which Hunts were Conducted at Kellogg Bird Sanctuary and Farm-1951-1952, Hickory Corners, MichiganOOCOOOOO......OOOOOOOOOOOO.......OCOOOOOOOOOOOOOO. Conditions Under which Hunts were Conducted at Kellogg Bird Sanctuary and Farm-~1952—l953, Hickory Corners, MiChigaDOOOOOOOCO......OOCOCOCOO......OIOOOOOOOO00.0.00... Conditions Under which Hunts were Conducted at Kellogg Bird Sanctuary and Farm—~1953-195h, Hickory Corners, MiChiganOODOO00.0.0000...00.0.0.0.........OOOOOOOOOOOOOCOO Page h 10 ll 15 16 25 26 27 28 29 3O 31 LIST 1h. 15. 16. 17. 18. 19. 20. 21. 22. 23. 2h. 25. 26. 27. OF TABLES - Continued Conditions Under which Hunts were Conducted at Kellogg Bird Sanctuary and Farm-~195h and 1955-56, Hickory Corners, MiCIIiganOOOOOOOO000......0.00.0000........IOOOOIO Summary of 1951-1952 Lincoln Index Data--Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan............. Data Used to Estimate the Preseason Populations, Kellogg- Bird.Sanctuary and Farm, Hickory Corners, Michigan........ Data Used to Ebtimate the Preseason Populations--Kellogg ForeSt’ HiCkOW Corners, MiChiganO.....OOOOOOOOOOOOOOOIOOO Summary of Marked-Unmarked Ratio Changes During the 1953- 1955 Hunting Seasons--Kellogg Forest, Augusta, Michigan... Lincoln Index and DeLury Preseason Population Estimates Compared--Kellogg Bird Sanctuary and Farm, Hickory corners, MiChiganOOOOCOOOOOI000......00.000.00.00.0.0.0... Kill Statistics and Indexes of Abundance, Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan............. Kill Statistics and Indexes of Abundance-~Kellogg Forest, ’Aug‘lsta, MiChiganO......OOOOOOOOOOOOOODO00.0.0000.0.0.0000 Trapping Data Indexes of Abundance-—Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan............. Trapping Data Indexes of Abundance-~Kellogg Forest, Augusta, MiChigan.......COOCCOOOOOO0.0.0.0....00.......... Summary of Rabbit Population Data--Kellogg Bird Sanctuary and.Farm 1950-1955, Hickory Corners, Michigan............. Summary of Rabbit Population Data--Kellogg Forest 1951- 1955, «AugUSta, MiChiganooooooooooooooooooooooooococo-coco. Distribution of Birth Dates Among Juveniles Live—Trapped in August and September--Kellogg Bird Sanctuary and Farm, HiCRory Corners, Micmgan000000.000......OCOO...00.0.00... Summary of the Kellogg Forest Crippling Loss Experiment-- 1955-56, Augusta,MiC11iganOOOOOOOOOOO......OOOOOOOOOOOOIOO Page 32 to h2 LB St 57 58 65 66 71 71 86 9O .-l. -S J. O. 'qu. .D‘ 2'7 ‘VUAI S -,. .1“ 3a.". 3 fl 3' ‘ ‘ ‘.:';I Q J. 17/:-‘ a. 0‘ . .~ .(J a. .... u C7 0. 1?; . {-1 3 i. ll“ Foroc 4‘ 37. l ‘.c «Au. lv d a ‘1 LIST OF TABLES - Continued Page 28. Summary of Non-Planted Rabbit Recoveries, Kellogg Forest 1955’ AugUSta’ Micrliganooooo00.00.000.00.ooooooooooooooooo 92 29. Distribution of Rabbit Ages in Hunting Kills, Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan............. 96 30. Distribution of Rabbit Ages in Hunting Kills, Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan............. 96 31. Age Distributions Following Years of Heavy Hunting Kills Compared to Those Following Lighter Kills................. 98 32. Life Equation Kellogg Bird Sanctuary and Farm Rabbits, 1951-52, Hickory CornerS, Michigan........................ 100 33. Relative Importance of Mortality at Various Times, 1951—52 Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan 101 3b. Daily Hunting Effort and Rabbit Kill l95l--Kellogg Forest, ‘AuguSta’ MiChiganOOOOOOC.CO.....C......................... 10; 35. 1952-1953 Daily Hunting Effort and Rabbit Kill-~Kellogg ForeSt’ AuguSta’ MiChigano00.000.000.000...OOOOOOOOOOOOOOO 106 36. 1953-19Sb Daily Hunting Effort and Rabbit Kill-~Kellogg ForeSt, AuguSta, MiChigaHOOOOOOOOOOOOOO0....0000.0.0000... 10? 37. l95h-l955 Daily Hunting Effort and Rabbit Hill--Kellogg Forest, Augus-ba, Michigan...................O...........OO 108 38. Yearly Hunting Effort, Kill and Success--Kellogg Forest AuguSta’ Mic‘higanoooooooooo00000000000000.00000000socoo... 11]- 39. Monthly Effort, Kill and Success Summary l951-l95h SeasonS--Kellogg Forest, Augusta, Michigan................ 113 to. weekly Distribution of Hunting Kill, Effort and Success l951-l95h Seasons--Kellogg Forest, Augusta, Michigan...... 118 bl. Daily Distribution of Hunting Kill and Effort-~Kellogg Forest, Augusta, Michigan................................. 120 L2. Distribution of Hunting Effort, Kill and Success 1951 Season--Kellogg Forest, Augusta, Michigan................. 123 xiii LIST OF TABLES - Continued Page D3. Distribution of Hunting Effort Among the Hunters l95h-55 Season--Kellogg Forest, Augusta, Michigan................. 125 hh. Distribution of Hunter Visit Frequency and the Mean Visit Length, Success, Dog Use and Percentage of Visits with Snow'Cover for Each Visit Category, 19Sh Season-~Kellogg Forest, Augusta, Michigan................................. 126 kg. Distribution of Visit Lengths 19Sh-19SS Season-~Kellogg ForeSt, AuguSta’ MiChiganOOOOOOOO0.0.0.0....00.00.000.000. 128 to. Distribution of Rabbit Kill among Hunters 195h-1955 Season-~Kellogg Forest, Augusta, Michigan................. 130 L7. Distribution in Hunting Success 195h—1955 Season-~Kellogg ForeSt, Aug‘JSta, MiChj—ganooooooo0000000000000...cocoons... 131 D8. Observed Distribution in Kill in Various Effort Categories Compared to What would be Expected Due to Chance. 195b- 1955 Season--Kellogg Forest, Augusta, Michigan............ 132 h9. Relationship Between Cloudiness and Hunting Effort and Success 1951 and 1952 Data.Combined--Kellogg Forest, AuguSta, MiChj-ga’noooooooooooooooooooooooooooo00.00.000.000 1'36 50. Relationship Between Minimum Daily Temperatures and Hunt— ing Effort and Success 1951-1953 Data Combined—-Kellogg Forest, AuguSta, MiChigan...C.......OOCOOCOCOOIOCOCOO.0... 137 51. Relationship Between Maximum Daily Temperatures and Hunting Effort and Success 1951—1953 Data Combined—- Kellogg ForBSt’ AuguSta, MicmgaDOOOOOOOOOOOOO......OOOOOO 138 52. Relationship Between Mean Wind Velocity and Hunting Effort and Success 1951-1953 Data Combined, Kellogg Forest, Augus‘ba’ MichiganOOOOOCOCOCOOOOO.........COOOOOOOOCOOOOOOI 139 53. Relationship Between Precipitation as Rain and Hunting Effort and Success 1951-1953 Data Combined——Kellogg Forest AuguSta’ MichiganOCOCOCOCOOOCOOCO00.0.0000...0.00.00.00.00 lilo 5h. Relationship Between Snowfall and Hunting Effort and Success 1951-1953 Data Combined, Kellogg Forest, Augusta, Michigan...........COOCCOOCCCOOOCO......OOOOOCCOCCCCOOCOOO 1’41 55. Relationship Between Snow'Cover and Hunting Effort and Success 1951-1953 Seasons--Kellogg Forest, Augusta, Micmgan................0.0............................... lh2 LIST 56. 57. 58. S9. 60. 61. 62. 63. OF TABLES - Continued Hunter Occupation and Success, l95h--Kellogg Forest, AuguSta, MiCkligan......................................... Effect of Dog Use on Hunting Success, l95h--Kellogg Forest :‘iugus‘ba’ MiCkligan.....................O...‘......000...... Dog Use and Hunter Success, l951--Kellogg Forest, ‘iugusita’ MiChigan................OCOCO......COCOCOOCOOOOOO Instantaneous Rates of Kill Per lOO Gun-Hours at Various Success Categories, l95b--Kellogg Forest, Augusta, Michi- ganOCOOCOOOOOOOOOCOOCCOO......OOOO.........OOOOOOOOOOOOOOO Comparison of the Observed Weekly Distribution of Kill (1) ‘With the Distribution Expected if the Kill Depended on the Relative Amount of Effort by Various Success Categories Each of which Caused a Different But Uniform Instantaneous Rate of Kill (2) And the Distribution if the Kill was the Product of a Uniform Instantaneous Rate of Kill During the Entire Season (3). 195h-l955 Season—-Kellogg Forest, itlgusta,.Michigan......................................... Influence of Various Factors on the Instantaneous Rate of Hunting Kill, l95h--Kcllogg Forest, Augusta, Michigan..... Relative Abundance and Use by Rabbits for Winter Food of 'Wood, Vegetation, Winter 1951-52 Kellogg Bird Sanctuary and Farm, Hickory Corners, Michigan....................... Mbody Plant Abundance and Use as Food by Rabbits From Quadrat Data, Winter 195h-l955--Kellogngird Sanctuary, HiCkory Corners, Michiganooooooooo000.000.000.000.0.0.0000 Page th 1H8 1119 158 161 163 178 180 - u ...... ........ vvvvv oooooo ............ I") C U) .r‘ "I D 1 (IN _a- In F ‘ D. 31... I I F S‘P’EQ 4. u- “b; . 7. 33.23:: «4.1 / PA .n‘ .. Halal E q LIST OF FIGURES FIGURE 1. 2. VI 4:- w 0 O\ 7. 8. 9. 10. ll. 12. 13. 1h. 15. l6. 17. 18. surruner Trap Lines 1951.00.00.0000000000000O0.00.00.00.00. Spring Trap LocationS.................................... Fall Trap Lines 1951000000000...000000000009000000000000. Fall Trap Lines 1952-19550.00...0.0.0....0.00.00.00.00... . Fall Trap Locations 1951-1955............................ Spring Trap LocationS.................................... Rabbits Shot 1951-52..................................... Rabbits Killed 1952-53................................... Rabbits Killed l953-Sh................................... Rabbits Killed 19Sh...................................... Rabbits Shot 1955........................................ Fall Population Estimates by DeLury Method--Kellogg Bird &nCtuary and Farm 1952.00.00.0000000000000......OCCOOOOO Fall Population Estimates by DeLury Method--Kellogg Bird SanCtuary and Fam 1953......0.00.0000.........IOCOOOOOOO Fall Population Estimate by DeLury Method--Kellogg Bird Sane-tonal? am Farm 1951, 1955.00.00.00.........OOOOOOO... Fall Population Estimate by DeLury Method--Kellogg Bird SaHCtuary and Pam 191160.000...0............OOOOOOOOOCOOO Total Kill Compared to Fall Population Estimates......... Kill During Constant Hunting Efforts Compared to Fall Population EstmateSOOOOOOOOOO......OOOOOOOOO00.0.0000... Rates of Kill Compared to Fall Population Estimates...... Page 8 12 13 lb 17 19 33 3h 35 36 37 50 51 52 55 S9 62 6b LIST OF FIGURES - Continued Page 19. 20. 21. 22. 23. 2h. 25. 26. iFall Trapping Data Indexes of Abundance Compared to Pre- season Population Estimates--Kellogngird Sanctuary and Farm.0.0.0.0...OOOOOOOOOOOOOOOIOOOOOOOOOOOOOOOOOOOOOOIOOOO 67 Fall Trapping Data Indexes of Abundance Compared to Pre- season POpulation Estimates--Kellogg Forest............... 69 Hunting Kills--Kellogg.Bird Sanctuary and Farm-~1932-1955. 7h Hunting Kills Kellogg Forest, 19140, l9hs-SS............... 77 ‘weekly Hunting Effort and Rabbit Kill Kellogg Forest 1952 and 19514 seasons CombinedCOOCOOOOOOOOO......OOOOOOOOOOOOOO 115 'Weekly Variation in Hunting Success Kellogg Forest 1952- 1951‘SeasonsCombinedOOOI......‘COOOOCOCOCOO......OOOOCOOO117 Hourly Distribution of Hunting Effort Kellogg Forest l95h- 1955 Seasons.........C...........0............C..........O 122 Fluctuations in Hunting Success and Factors Influencing writing Success, Kellogg Forest, 19514.00.........OOOOOCOOI 1514 xvii INTRODUCTION Because of the complexity of the factors affecting a wild animal population, it is difficult to evaluate precisely the relationships of population characteristics during the span of only a few years. 'Major objectives of this study of the cottontail rabbit (Sylvilagus floridanus) at the Kellogg Station were threefold: 1. Evaluate the economy of currently available census methods. 2. Appraise the effects of factors influencing rabbit abundance, especially hunting. 3. Analyze hunting pressure as it occurs on a public area. The investigation of the effect of hunting on rabbit abundance was emphasized because this mortality was most readily controlled. Furthermore the recreational value of rabbits is realized largely from their harvest by hunting and it was desirable to evaluate the influence of this harvest on future abundance. Despite only five years duration, this study has begun to indicate the character of interrelationships between the extent of the hunting kills, spring and fall population densities and fall age ratios. Description of the Study Areas Kellogg Bird Sanctuary and Farm The Kellogg Bird Sanctuary and Farm is a continuous area of 500 acres located in section 8, Ross Township, Kalamazoo County, Michigan. "at 31rd Salazar; Kind Park, a 3‘ Some 2337.59: of the area size. time of Allen‘s marines dam to 931‘s 20-;0 fee plantations f1“; fonarly 0‘5“ m " run 6 cover. To obtain , ‘1 . , Ker°og Scam“ 19.110 58 Far] .4 The Bird Sanctuary is surrounded on three sides by the Kellogg Farm, Midland Park, a residental area adjoins on the west. Allen (l938b)described in considerable detail the history, physiography, soils, climate and plant and animal communities of this area. Some changes have occurred, however, in the vegetation and fauna of the area since Allen's study. The conifer plantations during the time of Allen's work were not over eight feet tall and the trees had branches down to the ground. During the present study, these conifers were 20-h0 feet tall and had lost their lower limbs. Consequently, near the ground much less cover currently was afforded by conifer plantations than was the case during 1935 and 1936. In contrast, formerly open grassy fields have now developed considerable shrubby cover. To obtain a description of current vegetative conditions the Kellogg Sanctuary'was cover mapped during the summer of 1951 and the Kellogg Farm during the winter of l95h. Cover types were classified according to the following outline which gives symbols used in.Table 1. OUTLINE OF COVER TYPES AT THE W3 K. KELLOGG STHTION OF M.S.U. 0 Open Land Herbaceous vegetation, or woody vegetation less than 2 feet high. woody vegetation over 2 feet high covering not more than 10%. Not cultivated 1. Grass - 70% or more grass c - scattered conifers 2. Low shrubs -'10% prostrate shrubs 3. Mixtures of 1, 2, and weeds - none over 70% on area of more than 30 yards by 30 yards . S Over grown land Open land overgrown with shrubs and young trees on more than 10% of area. No woody clump more than 30 x 30 yards. 1. 10% - h0% covered with woody plants 2. h0% - 90% covered with woody plants Note: Cover'type of open land between woody plants will be found in parenthesis ' T Thickets Dense stands of shrubs or small trees less than 20 feet high covering 90% or more of ground W’ Deciduous Woods 1. Young - mostly DBH 12 inches or less - over 20 feet high 2. 01d - DBH - over 12 inches Underbruch density a. light — 0% - 10% ground covered b. medium - 10%- 60% ground covered c. Heavy - 60% - 100% ground covered C Conifer Plantation 1. Pine 2 . Spruce 3. Larch Accompanying vegetation a. none, trees closed b. not closed - per cent not covered and its type noted in parenthesis A Crop Land This method of classifying cover is patterned after that used by Bump (1950). The distribution of cover types on the Kellogg Sanctuary and.Farm is indicated in Table l. Much.more woody cover was present on the Sanctuary than was present on the Kellogg Farm. It consisted of well- interspersed conifer plantations, open brushy hillsides and dense lowland thickets. Kellogg Farm vegetation consisted largely of open crops and pastures with woody cover limited to woodlots and scattered 0.9.: 11.12.. CHEF Toe. ~.:;_ {.33 9.3. ... 9.; fig Con N..fi.~ h.“ mofifi £. Q. “China. UCI Q§.HUH.fi-m~ 9.3 ..wofifi mu. 3.“ Nonvm. JoHN 309'..- QOQQ 0.: fine men H42: O.N: muses N. N. < vale teens: H30 N. o.H m. o.H m. o.H emu oomoao “coapmpcwaa man no Amoco mosham ;.o o.mm o. m.~ m. a.H m. a. nae mono .eospnonnaa mesa n.HN N.HH :.m a.nH o.H m.n .~.OH 4.0H «so ennoao .noapnpnnan seam h>mon :mshnhooc: ~.m s.mm a. o.m m.m o.m one .m.m.n =~H can» once noon: nsoeeaooa Eafivoa snapshocs: a.m m.ma m.m a.wa a.m o.mH m.m m.m nne. .m.m.o .NH gasp once neon: neosefiona pawns snsnnnmoqs H. m. N.H m.o n.H m.o «we .m.m.n ems snap ones neoos mensesona h>won suspnhooqd H.H a.m m.H a.» m.H 4.5 m. m. use. .m.m.a gas can» nnnH neon; nsoenaoma Edfioos gmsnnnmocs a.~ a.sa s. a.H m. o.H a. a.; nu: .m.m.o gmH can» nnna neoo: nsosesona pawns ansnnnnen: ~.H w.w w. ~.m m. H.H m.m o.m max .m.m.a smH can» mood .moooz escapwoon N.N m.HH m.~ 0.4a ~.H m.o 0.5 m.a a possess w. o.m m.m ©.m mm oospam memnoa «ocean mouse: :zoawnm>o m.n o.mm o.a s.mm a.s a.®a H.eH a.oH mm noses neon: Remus: .enna neonmno>o a.sH p.05 o.m m.ma m.H o.a m.» o.a Hm no>oo neon: moaned .enna gzonmnneo 0.0H m.Hm oao mpoMHcoo oonoppeom .msooosnpom a.HH N.He 4.0m m.mmH s.mm ~.Hsa m.HH a.HH o emsn>spaso son .nsomomnnnm Memo pom mono< psom.amm monpfl Each one Waco Mom momma, mmmu Mom. moped coco mama umboo pmopohlmmoaaox handpoqdm chflm shah wwoaaox thSpocwm opamlwmoaaom wmoaaoe asses Zion—”mas «mmmzmoofimoxog «zen—“Hana 806.5% a9 94 ZOHBHmOgo mafia. 5500 H aqmde 0.00H n.2Hm ~.mm H.: m.0m o. 03” o.m m.o m.mm H.~om «.ma 9% H.wea m.mm :. m.om m.~oa :. :.Hm N. 0,00 ENG awash pad mwafioflfism hope: some 2: salesmen mono .cowpaPQdHQ haw ho file at the L :wwrs‘ty. :eez: badly erl 3C illustrate mien practi, K Wide Vi hagtat imprc‘ l the “I" 9%? of ”35119215 e .~ 2‘ ‘1 &: Eleve ”011:; A; of this (1:. WM“. “a O - Phroaf. A“ conifers and shrub wildlife plantings. A cover map has been placed on file at the Department of Fisheries and'Wildlife, Michigan State UniVersity. Kellogg Forest The 515 acre Kellogg Forest, located about two miles southeast of the Sanctuary, was donated to Michigan State University in 1932 by ‘W. K. Kellogg. It has rough topography and sandy soils. The area had been badly eroded. It waser. Kellogg's desire that the area be used to illustrate the rehabilitation and use of such land by proper conser- vation practices. . A wide variety of coniferous and deciduous Species have been planted. Many of these are species which commonly are used in wildlife habitat improvement plantings (Gysel and Lemmien, 1955). Because of the variety of plantings and natural vegetation, this area presents considerable difficulty in defining cover types. Table 1 lists the relative amounts of various types of cover found on the area. A cover map of this area is also on file at the Department of Fisheries and Wildlife, Michigan State University. Hereafter the Kellogg Bird Sanctuary and.Farm area is referred to merely as the Sanctuary and the Kellogg Forest area merely as the Forest. The several areas together are termed the Kellogg Station. fraying Perfizj Live tra; Grief EJ'GI'pOS-a ‘ for census par; $31.35 were :13: a W33 of the a. quently 33:9 0: tn}? ing '35 L). it the $5 alaost coma“ . r ‘ ‘ 604319360 54 .. (IA? 9 it“ ‘1 ‘ \ ...e ‘1] Orion Dwain infle GENERiL PROCEDURES Trapping Periods and Locations Live trapping played an important part in the investigation. The chief purpose was to enable a portion of the population to be marked for census purposes but, much additional information also was obtained. Traps were placed so as to capture as many rabbits as possible from all parts of the area. They were set in the most likely places and conse- quently were often unevenly Spaced. At the Sanctuary, the bulk of trapping was done on the central part of the area because most of the rabbits were located there. The more even distribution of rabbits at the Forest enabled the traps there to be located throughout the area. Ear corn was used for bait at all times. During 1951 apples were also used in addition to corn but did not seem to increase trap success. At the Sanctury trapping during 1951 started June 7 and continued almost continuously until August 29. Five trap lines were operated. Dates, numbers, and success of traps are shown in Table 2. Trap locations are shown in.Figure 1. That same year two additional trap lines were conducted simultaneously from October 17-29. One was on the Bird Sanctuary and the other was across the north end of the Farm. This was done in order to compare population densities on the two areas and to obtain information on rabbit movements. In 1952 and 1953 the entire central part of the study area was trapped during late August and early September to determine the age composition of juveniles and the incidence of warbles (Cuterebra sp.). Numbers of traps used and other details can“... ,1. ...:< Ca]: 37;- . _ r. J5. ..-.J . .IHIAH {and I; .1321 On 1r. m: an .2. 1‘1””; _ 1:. .OH 6:3. 07s 0:3. 93.5 0:3 C. .21: 3< EH! 52.-n Adz< ZJn—JHu—HUHZ a we: ZU~OU N wag KfluHaa NUNANDhnuUZd‘m h~x h n— aysuogqhui vn ¥ van‘iT—ui ..kz Hunflfifivb .Hunenm H. in :unnfi‘tr '3 EV.“ nofipsaesnoe H H H o H any». mo .02 m m H m OH 3.358 03533. m s m a m needs misses. N a m o 2 nodes“ 33 m H m o a 832 9.93 «a H w 1. mm amends 62 mm 3 am a Q. nonsense fleece b H MH m mm .338 sewage enemas 5 ma we e um Hessian: .oz 23 8e Rm 0% mam nee»? gene S we Hm om mH mags .oz 3 S S S Hm gen? .02 swam S .34 oTfi. 32. SA 38 SA .36 23-3 .msa Am ads on .8 cm .aw r: 1: .3 8% .03. 8% 3.38 83 .a as: 3 ea 25a eza E3873 Ema 80.33 53.? eggs GS. g Emma? . msezeoo amoeba N M393. nfi—N. I a 1110 -,,._.V . bl.“ . . .u @\ O r/ - . n o. wxj Qua... inseam: . an 1 1 1E 9...}. .. g. o wz... ... 0 m2... 0 0 m2: 0.. 115. wt... 0 etCfiWI-d a .u 6.: H ”0.3% #3.... 00¢ >21 amt: ate» «...—{£3 >m.<:._.uz> a amour. > w are provided in Table 3. The location of these traps was the same as during the spring. From 1952 through 1955 a ten day trapping period during the last two weeks in.March was established as routine. Data concerning these periods are found in Table h. Trap sites are shown in.Figure 2. Each fall live trapping was conducted largely during November. To insure more complete coverage of the better rabbit habitats on the Sanctuary, the trapping was done in two portions. Trap locations used in 1951 are mapped in Figure 3. The distribution of traps was then changed and the same locations used from 1952 through 1955 (see Figure h). This change involved a reallocation of trap locations between the two trap lines rather than a change in trap locations. The two lines trapped were shaped somewhat like a horseshoe with'Wintergreen Lake in the center and the two ends coming together. It was thought that the number of rabbits captured on'both lines might indicate the extent of movements during the trapping period. Details of this trapping are given in Table 5. At the Kellogg Forest trapping was conducted each year from mid- September to mid-October. The area was divided into three portions and each third was trapped about ten days. Details concerning this trapping are listed in Table 6. The trap locations were approximately the same each.year. Those used in.l9SS are indicated in Figure 5. 'Varying numbers of traps were located by Spacing the traps somewhat closer when larger numbers of traps were available. Because the south trap line was shorter than the others, during years when more than “‘0“ ." - . v. a 1! ‘1‘3‘ 0' F;_ r- ir-Uaa “In FAQ. ‘ th“rn ‘ ‘H 0a .‘_.3 H . . ..‘a: mgr-2 0 111231- ”1.4:.- . V .A. A- v ., .3 . He“ 31 oJ‘A. CIT-“931- reca;:; TABLE 3 LATE SUMMER AND EARLY'FALL TRAPLINS SUMMARIES KEELOGG BIRD SANCTUARY AND FARM HICKORY CORNERS, MICHIGAN Line T F 1.1.0. w.c. K.B.S. Farm Dates Aug. 26- Aug. 15- Oct. 18-25 Oct. 17-29 Sept. 5 Sept. 8 1951 1951 1952 1953 Number of nights l2 13 11 2h Number of traps 30 39 57 73 Trap nights 360 507 ' 627 1752 Number of individuals 29 ll 27 38 handled Current recaptures 26 h h 15 Total captures SS 15 31 53 Number marked 2b 11 28 36 Adult males 1 3 3 3 Adult females 3 l 11 3 Juvenile males 13 3 11 13 Juvenile females 12 h 9 l6 ll NH mm ma Om Omw mp OH HH ma mm ma 8 mom my HH SA :3: mad ..az NH 4H 3 mm Om . Own Ow m RAN :3: m samumm 33: omuam £82 HTS 38: 8.3 .3: HN ON mm 3 3 OOO Ow OH JO—ZTM one 4w OH Omw mm OH mm mm ma mm gm gm. 0mm my OH mmHmEmm mo pmnasz OOHUGMQ Have» we moHda Mo hmnEdz Umxhds pmnasz mmndpmmO Haves mmhdpmmomh pamphso v3.25: mamsflfiefi go 3952 mpzqu mane mqup mo umpeazp mesmen mo umnesz v.88 .1mmmH mmmH pmmyom mwOHme mm”? mm“? Sham £.hhmdpo:wm UAHm mmOHHmM ammH anfi mmmHl ficEBz .mmmzmoo $3on 534% cocaiemsgm mEq n3; gamma a MHm¢H .F \K J; G ‘ a...» cow neg: 3.3m . . 203(004 at a . eow . Q . a > gauzfinz ... b . . ngm 80.33.22, .N W130; 6.... RH ml... smm§tz “Sr—021%. . 35 83322? sq auteur. 15 ma ow am mm 0m cm 0: mm N4 em mmawsmg mHHcgsn Mo .8952 NH 4H ow mm mm em ON ON gm mm mmams maflcm>sn mo pmneaz w m HH m m m NH m mH m moHdEmu 933.. Mo $252 HH N. m a m o m a O m mmHmE pHsum mo 39:32 am i am mm B R mm as S om Bias $952 Hm we mm MHH ooa mHH «am we mam mam eoaecea proe mm mm mm mm Na Na mmfi mm 00H zma ampspamomy pamgnsnv a: om No am am He me mm am we weaved: maasee>fleqfl mo amnesz cam mow mmw ogaa mooa ooNH gmoa HQOH New mam mpnwfic megs mo “mneaz aw 05 me we me me aw we we 05 mampp mo umpesz OH HH HH ma 4H 0H mm ma HH ma magma: mo amnesz .1maum 1: .>oz mmuma wen: H .omm mm .poo e emuo m .>oz oa-:.omm mfl1m -4 .>02 -mm .900 .>oz .>oz -mH .poz ~H.N .>oz .>oz -Jm .poo amuaw .>oz .>oz mmmm, pmmm, swam paw: pmmm paws gum: swam, npaoz mpuom 1mmma «mma HMQH 1’ ‘ $12 l‘l‘l‘ zeonOHz .mmmzeoo Mmouon Swarm Q24 Hfiraozfiu Sim 80% Mg wage zomgm clan—”Emma m mamas o m N m mmHas HHseH N 04 Nm mm sOHHmm emNe pmnssz O am mm O: Ommmwp pmnssz O NO mm Om mmhfipamo Hdwoe H NH mH OH mmpdpmmomp psohndo N m: mm Am>HH NNOOH emHeam: mHmsvH>HOcH mo hangsz ONN mm» me mpan: dupe NN mO mO mQMNp no geneaz OH HH NH mp£MHc Mo AmQEdz AmcHH m.HmmNoV wHam.poo N.poo.NN.pamm NN-OH.pamm mmpwq epuoz Hmmx epsom :HNoz maHH mmmH Admw N m 4H mH NH Om OH :H mdeEmH deum>dw a O NH OH HN ON MH N mdeE mHHcm>sw H N a o a m a H mmHmsmH pHseH O H O O O N O m mdeE pHdUd O O am ON Nm mO an on Ommmmp Mo umOEdz MH mH N4 N4 NO ON Nm 4m megapamo Hdpoe H H NH NH OH MH NH : mohspdwomh pamppdo NH NH mm mm Nm mo mm 0m emHecmz meseH>HeaH Ho umpssz HNH ONN ONN OHO mNO paw OHO OmO mpzmwc nape NH NH om Om mN NN NO no mmmpp no Nonssz m NH mH HH m HH NH 0H mpstc mo pmnezz NH-HH.poo N.poo-ON.pamm m-N.poo O.poo-ON.paom mN-NH.paom NH-N.poo NN-NH.pamm OHnN.pamm mmpam appoz epN021‘ pmmx zpsom empoz gpgoz gpsom paw: mcHH NmmH HmmH Admw znuHmon .campcsa emu mom.OOqumxllwmézzbm OZHdeme ZOO 3m OZHBZOmmmm O mqmde H J 859“ch HN NH HH NN OH mH mmHNEOO OHHcmpsO HN ON OH HN OH HH mmHas OHHOO>OO O HH O m N O mmHNsmO OHOOH O NH O H H m mons OHOOO Hm Nm Nm OO Nm Om OOHHoN OONO NOOEsz NO ON mm NO Nm Om OONNNO umnssz NO NN NN ON OO Hm OONOOONO proa OH mm OO Om HH NH mmaspamoog OOONAOO Hm NO mm OO Nm Nm OOHOSNO meHHUgflUGH HO hmflfifiz NON OmN OON OHN ONO OOO NOONH: NOON OOH - m.OoO.ON.OOOm OO - NN-ON.OOOO OOH OOH ON . NN-mN.OOOO HN NO OO . mNnoN.OOOm mOmNp No NOOEOZ ON . ON O mN.OOOm N N HH OH OH OH NOONH: No gmnssz NH-mH.OoO NH-O.OOO m.OoO-mN.OOOm NH-OH.OoO N.OoO-Om.OOmm NN-ON.OOOO OOOOO Ommz Ognoz Opsom One: Oppoz OOOOO oaHH mmmH OmmH Admw 8.995%;— O OH OH NH OOHNEOO OHHao>OO m ON mH ON mdeE OHHHHESH. H O H m mmHmsmO OHOOO r'bUNE 5 KELLOGG FOREST FALL TRAP LOCATIONS l9fl~l952f 17 ...{s :38 55- =9 3:. :02. .....;:... ::o ‘3. I. ... r . I: .3. .... . .° ‘1.:.:. 3.. .0: 0:“ .3: .. ’° 0 a. . 5" ':: :;~ ' ° "° .-.-' O ., .. ..”~ . . ... . . . . . . :: fl . 3." .5 .1 1’ s 0.. :: ............ f. ..t'. ...: .. ". :5 Q ‘I .9 . O... .0. n - .-:...;,..;,,-..,‘.,.;-.-.-.-::. O . ‘ OM egrgooooao 3 - mm m?!" 3 soup I INCH: 53.5 FEET + 0 $0070 LINE 0 WEST LINE :go.oooooo.o.o‘oé\u. o . 6‘.{o o‘.oo‘oo...ooo.o :- ‘- 0 “-1“ z- '3- :g. -; S:- :5 35- I‘- 05: :2 53. >5: .‘0 53 .E: ooooo ,0 0.0.00 '5; ".50 ""1733 ‘ :: .--'¢3§ -E ‘5: .: .'.' '3 = ° ‘22 . .20 '3 gg-z- ‘3 " "uncut-V 18 80 traps were available the trap locations indicated by green crosses in Figure 5 were added. In 1952 two additional trap lines consisting of straight intersecting lines of traps were set up to obtain inform- ation on.movements. During early March 1955 a trap line comparable to that conducted at the Sanctuary was operated. Some details and results of this trapping are given in Table h. The trap locations are indi- cated in Figure 6. Types of Traps Two types of traps were used. Fifty were unpainted wooden traps (as described by Hickie, l9h0) and the remainder were size 3, wire mesh traps manufactured by the National Trap Company, Tomahawk Wisconsin. During the first year, wooden traps were three to five times less effective than the metal traps. However, as the study progressed wooden traps increased in effectiveness while metal traps seemed to become slightly less effective. From 19Sh until the end of the study, during September and October, wooden traps were about twice as successful as metal traps, but during March, November, and December the two traps were of equal effectiveness. The efficiency of wooden traps early in the study apparently was reduced by the odor of 'penta” wood preservative with which they had been treated. Cutting three inch holes in the ends of half the wooden traps did not increase their effectiveness. The success with which metal traps took rabbits during warm months was greatly reduced because during the day they often were sprung by red squirrels (Tamias hudsonicus), chipmunks (Tamias striatus) and striped ground FIGURE KELLOGG FOREST LocAv IONS . l9 SPR‘NG TRAP .133 "w -°.- 3.5-4. :3 ‘5. ’00:: 000:. u a... 0000 :‘3. £3 = 0.: .3 0: :0 b :. :. :: .0 o: .0: 2: :5 z: “a. ;, .: : :‘ .' '.a :0 fl .0 0‘ ‘. n .U ' '3 "a a. :U : :- r. 33 °' 3 . o I. ll 1P... gm ozsy... . o. o. ¢:.~o~ 0.0.... 0. 0.. . .... SCALE: / INCH! 535 FEE 7’ 2O squirrels (Citellus tridecemliniatus). hboden traps took these animals only rarely. Traps were checked during the morning except during the summer of 1951 when they'were also checked during the late afternoon to reset Sprung traps. The effectiveness of metal traps was somewhat reduced during the study because they became progressively more battered by raccoons and from handling. HandlinggMethods Rabbits were removed from upturned wooden traps by grasping them by the head (as described by Petrides, 1951) and placing them in a cloth bag. This procedure was unsatisfactory for the removal of rabbits from metal traps because they would run up the sides and escape when the trapper opened the door. Rabbits were best removed from metal traps by placing a cloth bag over the open door and frightening the rabbits into it by striking the opposite side of the trap. The date, location, age, sex, weight, breeding condition, and presence of conspicuous parasites were recorded for each rabbit handled. The nose-rump and hind foot lengths of rabbits not fully grown also was recorded. Sex and Age Determination Aging and sexing was done by the criteria described by Petrides (1951) except that additionally the epiphyeeal cartilage at the distal end of the radius and ulna that mark juvenile, was located by palpation I in many live Specimens. The bump caused by the epiphyseal cartilage was detected.mere1y by sliding the end of the forefinger along the lateral side of the distal end of the forearm. Because nipple size was 21 not a reliable age criterian in females (see beyond) and penis length was sometimes a questionable criterion in males, the aging of some live rabbits was uncertain. The writer has the impression, however, that deepite some difficulties,very few rabbits were incorrectly aged. No rabbits which were later shot had been incorrectly aged when first trapped during 1951 and 1952. Shot rabbits were aged by the degree of ossification of the epiphyseal cartilage at the proximal end of the humerus, as described by Hale: (l9h9) and Petrides (1951). During the last three years of the study obvious aging errors of live rabbits (eSpecially males) were occasionally made by less experienced men assisting in the trapping. These errors did not effect the results of this study because age ratios were established from the hunting kill which could be accurately aged. MarkingiMethods Almost all rabbits were eartagged in both ears with numbered monel metal finglerling type tags (style 1005, size 3), National Band and Tag 00., Newport, Kentucky). The ear was folded double when the tags were inserted (Hangen, 19h2). Thus, the ear was pierced in two places so that when it was unfolded the tag was completely inside. This method was very satisfactory since only one tag was lost. Each year several rabbits had one of their two tags ripped out by shot, but in no known case were both tags lost in this manner. During 1951 and 1952 small juveniles were marked with National Band and Tag COMpany ear tags. These were numbered metal surfaces with two metal extentions. The extentions were placed through a small hole punched in the ear, then through a light metal washer placed on the other side and spread open. An ear tag gauge (also obtained from National Band and Tag Company) was placed under the washer before the extentions were opened to avoid cutting off circulation and to provide room for growth. Since only one hole was necessary these tags were better than fingerling-type tags in the rapidly growing ears of very small rabbits. In 1953 it became desirable to have rabbits marked so that they could be visually identified without retrapping or shooting. Fifty percent alcohol saturated with picric acid was found to dye rabbit tails a bright durable yellow; Rabbits that had their tails dyed yellow in November could still be clearly distinguished as marked rabbits the following June. The mark apparently was lost during the summer followu ing marking. Shortly after they had been dyed with picric acid during late January, 1953, three rabbits were found dead. This raised a question concerning the toxicity of the dye solution. Cohn and Githers (1928) reported that picric acid is toxic and that the fatal dose by mouth for various Species is approximately .5 gm. per kilo (h grains per pound). Yellow stains around the mouths of recaptured rabbits indicate that they had licked the dyed area. However, calculations indicated that it would be impossible for a cottontail to ingest a fatal amount even if it had been completely covered with 50 percent alcohol saturated with picric acid . 23 To get some idea if lagomorphs were unusually susceptable to picric acid poisoning, an experiment was conducted with 18 domestic rabbits (Oryctolaggs caniculus). Fourteen were administered picric acid in various ways and amounts and four held as controls. Only two rabbits that had received large oral doses died. G; L. Bowers (personal communication) reported no evident mortality from picric acid among many rabbits marked in summer in Pennsylvania. Later, at the Kellogg Forest up to 60 percent of the rabbits marked with yellow tails were later bagged by hunters. It would have been impossible to harvest such a high percentage of the population if a very high mortality had occurred. These observations indicate that when an alcoholic picric acid solution is applied to the tail and adjacent light colored hair no harm is likely to occur. The three recently-dyed rabbits that were found dead possibly expired due to the chilling effect of having about half their body (tail, sides and belly) soaked with a fifty percent alcohol solution when the air temperature was below'freez- ing. No information is available on the effect that color marking has on rabbit behavior. Malachite green (blue-green) and Nyanzol A (purple-black) dyes also were tried. Both were reported by'Fitzwater (l9hh) to show'promise. Also Flyger (1955) used the later with success in marking grey squirrels. Both of these dyes faded rapidly on rabbits. One animal whose tail was dyed black with Nyanzol A September 23 had its tail color recorded as white when shot by a hunter December 28. 2h Harvestinngethods The major method used to collect rabbits was shooting. At the Sanctuary almost all hunting was done under the supervision of the writer. Records were kept of the dates of the hunts, numbers of hunters, areas hunted, numbers of rabbits bagged, kill locations and cripples that escaped. Notes were also made of factors which might have influenced hunting success. Data concerning the hunts are tabulated in.Tables 7 to 1h. The locations at which marked and un- marked rabbits were bagged are shown in.Figures 7 to 11. During the winters of 1952-53 and l953-5h some unsupervised hunting was done by Kellogg Farm employees. The kills they recorded are indicated in Tables 8 and 9. These kills are probably minimal since they kept records in a very unsatisfactory manner. Also during those two winters, in order to greatly reduce the rabbit population, live traps and snares were used after hunting was no longer sufficiently productive. During the 1952-53 harvest a ferret also was used. Rabbit removals by these methods also are indicated in‘Tables 8 and 9. Hunting at the Kellogg Forest was open to the public and is described elsewhere. 25 .ofie chess one eds: pod ONO 23252 HHO.s H o N O NO ..2 2 ON 2. O mHOH ...N NH O O OH NN.OH .2 2ON.O ”2.2 ON ON NH H N HH NON .2. 2 NH.N ..2 a NHNOH 2N NH O N N O.N ONNHH .2.... 85H N O N O O N0.0H ..2 2 OO.H ..2 .1. ON. OH O m .52. O N N OO.N ..2 2 ONO .2.2 ONNH H ON O N N OO.N ..2 2 ONN ..2 2 SN N NN O O H OO.N ..2 2 NNNN .2.2 NNNH m HN O H H OO.N ..2 2 BO .2.2 ONN N NH O N O 8O ..2 2 NO. N .2.2 NO; O OH O O O ON ..2 2 ONO .2 2 ONN N NH N O NH N.OH ..2.e. NONHH ..2 a NHNOH N OH H O OH NN.OH ..2 2 ONcm .2. 2 mHNH. O OH H HH ON NH.N ..2 2 NNN ..2 2 NONH N OH H O NH NN. O ..2 2 NH.NO ..2 2 SN N NH O N m ONO 2. a NNOH ..2 2 ON N NH H N N 8.2 ..2 2 ON.N ..2 2 ONN O N .25 eodcoom pose OoHHNN 338.2 HHS 282-23 case on: 232:2 38 moHddHao $822 powwow $2232 H302. H.302. $32352 msfladpm mo nonfisz N. mqmdm. 2.62282 @2228 228622 NmNH..HmNH--§...2 222 22.225222 Bum 83.222 2222 2222.2 20 2222222 H o o o OO.N .2.2 OO.H .232 OO.NH m NH H O N N OO.N .2.2 OOONH .2.2. OO.HH N NH N O O O OO.N .2.2 NH.O .2.2 NHOH N OH O O O H NN.N .2.2 ONOO .2.2 NNON O NH O O O N OO.NH .2.2 OOON .2.2 OO.N O HH H H O N OO.OH .2.2 OOON .2.2 ONON O N N O H N ON.N .2.2 NOON .2.2 NHuNH N N H O H . O NO.NH .2.2 ON.O .2.2 OO.H N N O O O H NN.OH .2.2 OH.NH .2.... NNON O N .52. N O O H N0.0 .2.2 OOON .2.2 ON.H N HN O H O N NO.HH .2.2 OH.NH .2.2 ON.N N HN H O O H NN.NH .2.2 NHON .2.2 ON.H N ON NH N H N O0.0H .2. 2 NHONH .2.2 NH.OH O ON NH N N NH OO.ON .2.2 NHOO .2.2 OHONH O NN O O N NH ON.OH .2.2 NOON .2.2 NHONH O ON N O H N OO.OH .2.2 OOONH .2.2 OOOOH ...O ON O O O O OO.H .... N H NN OH H H O OO.NH .2.2 NOONH .2.2 NO.N O NN HN .N O HH OO.OH .2.2 NOOO . 2 NO.H NO ON N H O O ON.N .2.2 NOOHH .2.2 ONOOH O ON N O N N OO.N .2.2 OOON .2.2 NHOH O ON N O N N NN.O .2.2 OOONH .2.2 NNOOH O ON OH H N N NN.O .2.2 OOON .2.2 NHOH N NN N H N N NN.O .2.2 NOOHH .2.2 ONON N NN OH H N OH O0.0 .2.2 OHOO .2.2 OOuN O HN O H H O NO.N .2.2 ON.N .2.2 ONOH N OH N H H H OO.O .2.2 ON.HH .2.2 ONON N NH N O H H ON. .2.2 NHON .2.2 OOON H OH N O N O NN.N .2.2 NHOH .2.2 ONONH N OH OH H N NH N0.0H .2.2 NNOHH .2.2 NOON N OH .82 Umwwamm 202 .52 328mm 322. OOOONO 33932 HHH2 Oéo2éOO 952. 252. 2352 BOO 2.58 28.252 3.3930 2.29282 GOmwma 200,252 .7309. .2309 mcHnmfiQm wcfipuwpm Ho .3852 m 3%...9 2.23282 .O2222OO NOOOOHO NNNH-NNNH€22....2 22.2 22282.2 922 83222 O22. .3. 8222 22222.2 2O 2232,52 26 .opm .mmpmsm .mmwnp wcflxomgo maazz *2 .OENp 222220 02p was: we: ONO mumpnds HH2$ HNMN H (x. (I) (\o Hot—1 Ooo (\IOr-IOO 000 H: mr—ir—IN oOO moo nor—«OH OOOOr-l mam (“Own-+0 OOO r-lr—HA 3388.2 NN.OO2.NN.§O . .. n 8.8% NN.OO2..NN.ONO - - . $0.02.." NN. OO.N.NN. 5:. I I : vmamdme NN. OO.N.NNOOO « 5 Nu mhmxhohs Edam HH OOO-ON.>02 NN.O .2.2 NHON .24. OOONH N ON OO.N .2.2 NOOO. .2.2 NOOH N NH OO.HN .2.2 NOON .2.2 NOuNH .2.2 OHONH .2.2 OOuOH O NH OO.NH .2.2 NHOHH .2.2 NOuO O N O0.0 .2.2 OOOO .2.2 OO.H N N NO.O .2.2 ONOH .2.2 OOON H H .82 OO.O .2.2 OO.NH .2.2 OOuHH O HN ON.N .2.2 OOON .2.2 OO.H .2.2 OO.H .2.2 NOOHH N NN OOO.OH .2.2 NNON .2.2 ONOHH NO ON OO.OH .2.2 NOONH .2 N NOOOH N OH OO.NH .2.2 ONON ..22 ON” NH O NH 27 ( c ...ISIISE 7.25 I .. O O I I I 8.8% I I NH NN I I I 82222. O82 I N N HH I I I mpmxhgs £222 .222 H O H O OO. NH .2. O ON. HH ..2 .... NOON O ON O H N N NO.OH ..2 2 OO. .O ..2 2 ON.H 2. 2 ONONH ..2 O NHOOH N ON N O O O NN.ON ..2 2 NNONH 2 O NNON N HN N H N N ON.HN .2.2 8H 2. O N0.0H OH NH O O H H ON.OH .2 2 OO.O .2.2 NN. H O OH H O O O ON.O ..2 2 OOONH .2.... ON.OH N OH O O H H NN.OH .2.2 N0.0 .2.2 NHOH O OH O O O O OO.N ..2 N OOONH .2.... ONOOH N OH N H H O ON.H ..2 2 NOOO .2.2 ON.H- O OH O O H H ON.N .2.2 OO.O .2.2 ON.H N OH O O N N NN. OH 2. 2 OO.N .2.2 ONOH N NH O O N N NN.O ..2 2 OHONH .24 OOOOH O OH N O H O NN.OH ..2 2 NHON .2.2. ON.H N OH O O O O OO.O .2.2 NO.NH .2.2 NOOOH O OH H H H H OO.N .2.2 ONOOH .2.2 ON.N N OH O O H H OO.NN .2.2 ONOO .2 2 NO.H .2.2 ON.NH ..2 O ONOOH O N N O H N O0.0H .2.2 O0.0 .2.2 ON.H .2.2 OO.NH ..2 O ONuOH O N N H N O OO.ON ..2 2 ONON .2.2 ON.H N N O H H H ON.N ..2 2 ON.N .2 2 NOOH N O N N H H *NNN ..2 2 NH.NH .2.2 ONON «O N H O O O OO.N .2.2 ONON .2 2 ONONH H H .52. H O H N OO.OH ..2 2 NO.N ..2 2 NO.H O HN H O N O ON. N .2 2 OOON .2 2 NO.H N ON H O O O NN.O .2.2 ONON ..22 NOOH N NN O H N N SO .2 2 ONONH .2.2 ONON N ON NH O N N NN.O .2.2 ONOO .2.2 NH.N N NH O O N O OO.NH .2.2 OOON .2.2 OO.N O NH N N O N NH.OH .2.2 NO.NH .2.2 N .N N NH N N N N HN.O .2.2 NOON .2.2 NNON N OH N O O N OO.N .2.2 NN.N .2.2 OOOH O OH N O N O ON.O .2.2 NOON .2.2 NH.N N OH .82 JOOOONO 202 1 OONNOOOO $22. OOOOOO 32232 HHH2 22202-55 222. O2: 28.52 SS 3.2 :32 282:2 OOHOONOO 222:2 OOOOON 232:2 H.302. H302. OOHOOHOH2 OOHONOOO no .8952 2.28282 622228 228822 ONNHINNNHIIOEE Oz... Nagozé OOHO OOOHHOO 2222 222222 2O 22.22% m mamda .933 mhfipcm map 955 so: 6% whopfifi H.322. 28 .vmupoomu 902* HH O HH NN HO.HO .2.2 NNuN .292 NN.O OuN NNNH .NH .022 OH H N NN OO.NO .2.2 NNuO .2.2 NOON N-N NNNH .OH .022 * +N ON NO NN.NN .2.2 NHuN .2.2 OO.N O-N NNNH .OH .002 HH N HH HN OO.ON .2.2 OO«O .2.2 NouH N ONNH .ON .002 OH H N NH NN.NH .2.2 NHuNH .2.2.OO2OH N ONNH .ON .022 N O NH Om22222 ONNH .OH-OH .OOO Um mwmm 1 202 OOO Oo2womm 2222 OOOOOO 2222222 HHHO 22202-2:O 2222 2222 2222222 2222 zmmm 229522 mmammfi2o 229252 Ummmme 229522 H2909 Ndpoa wnw9mflcfih mcfipnwpm 20 209222 2282 O22 .222228 25282 ON-NNNH 222 ONNH--2222 Oz. 22222O22m 2222 OOOOHO2 O22 22 m2222 NHOO22 NO 2222222 OH mqmde l! -1 Il'. 29 .888 52 23H 893.38 3 802 ..2 m mfm ..2 H mHSH mH .Hsmmam: hpm> mom hvsoao mcoz :w an mawmmn H .2. m on“: .2.« maum ma .808 282 ..OH mm .pom 538 H ..2 m mdm ..2 H mHSH NH .338 282 ..m on ...Em 538 H om.HH .24. 8.2 0 ....OH 5:333... 32% .888 282 ..m mm .03 538 H H: 8H .24 85H m .82. .833 282 ..OH mm 332 H. .2. m on”; .sz omuH om .smHo 8% sow mH ....fi .838 H ..2 m oflm .zd ooum mm .338 395 .82 .8 NH .pmm $38 H ..2 m mm.~ .z.m mm; Hm .833 282 ..mH 8 ...6m 538 H ..z m 83 .z.m Ra aH .530 282 ..mH 8 5% $38 H mHmmB H .x. m 3% .zé 92H mH .8208 38m Ems SH H. .me 538 H ..2 m on”; H: omum NH hvdoao ..HHpmuHm Hum» Empcsm Siam 282 EH m 282 .24 miHH J: mHSH 0H .vHoo an nzov p50 hocoHoHme nopqdm hnzoao 30am h>dom INH ma ocoz .z.m Omum .z.m mHuH ma .Ummmmp 3392 8:93 no aowlom .hm>oo mmnmv CH Umpmnpcmocoo mpHnnmm hvsoao 30cm h>wmm :0 ON mcoz .z.m mmum .gfim mnua :H .m. 8a co gasp umppon hocm -HoHtm pmpaé .24 5.” mm 23.... mm. mnHSHm 302 .388 :86 92 .2.: ..m mm 282 4.: H3 42 8& NH 2 Hp .3? 95.3.3 339% .888 38m 293 .N\H N am 282 .24 mMSH .24 mo& NH .83 honoHOHtm 98:3 338 282 802 mm .pmm 538 H .zd omum Jam 8% m .03 mpcoesoo nmpomnmzo pcsm wQHnsa pm>oo .mo mm: mom mEHe msHmr mama GOHHdeQHoon .3ocm onfipmpmasoe mcfinmdcfimfl msHpndpm 5.352 .mfizmoo $3on mmmanammanuzm¢h 22¢ HMdDBQde QmHm cwoqqmm 94 QQBDDQZOO EMMS mezzm muHm3 mmmz: mZOHBHono HH mqmde lillllllllnlll|li IIIII.‘ 11". a: HszHo mcoz =N\H N am mHNamn H .z.m oo.m .z.m OO.N HH AUSOHO 0coz an mm .cwmm pmxooo H .pmm::muHoo H .z.m oo.m .z.m om.N m stOHu mcoz copmppmom mm mmHmmmn N .z.m m;.N .2.m mH.NH m zusoHo ocoz 0:02 mm mmHmmmn N .z.m om.a .z.m oa.H ,N Nu:0Ho meoz acoz an mmeme N .z.m OH.NH .EHH mm.m N .cmw Nu90Ho 0:02 0:02 Nm monme m .zwm o:.m .z.m ON.H Hm husoHo maoz mcoz om muoz .z.m OH.NH .z.H omfim Hm HusoHo msoz maoz Hm acoz .z.m mH.m .z.m omuH om HusOHo ocoz 0:02 mm ocoz .:.m mHuNH .z.H mH~OH om xszHo ocoz ozoz mm mcoz .z.m mH.a .z.m OHuNH mN hUSOHo thme mcoz mcoz mm AOprmp H mHmmmp H .z.m mowm .z.m mHuNH NN pmoHo ocoz umpmppmom mN NOHANmp H .z.< OO.NH .:.H O0.0H mN ammHo mcoz umhmgpmom mN mqoz N N NN NUSOHO Hprmm wcoz umpmppmom mN .Hmm :mnHoo H .z.m m:«NH .2.: mnum NN HUgOHo mcoz wmpmapaom mN .pmm :muHoo H .:.m mg“: .z.m mgnH 0N mfi30Ho 0:02 vmumppmum om mHmwmn H .pmm cocHoo N .2.H mguHH .z.H omuOH 0N MUHSHU mhmkozm umpmppaom 0:02 on 0Hmmmn H .z.m oonm .z.H mH.H 4N HusoHo ecoz 0:02 mm oHNMmp H .z.« OO.NH .z.H mm~0H :N HUSOHU :ng panH ocoz m: .pmm amnHoc H .z.m ooum .z.m mHuH mN HNSOHQ chN pgmHH mcoz mg 0:02 .sz mauHH .2.H ON.N MN HUHOHQ cwa 0:02 o: azoz .:.m 0H.4 .z.m oouN HN pmmHo mcoz mmzopmm mm .gmm_:onHoo H .z.m ON.H .z.m cm.H NH HusOHo mcoz =N mm .pmm cmquo N .z.H omuHH .2.H om.m mH Nu=OHo msoz .N mN acoz .z.m NH.N .z.m OO.N :H HUSOHQ scam pngH .N NN mcoz .z.m mHuH .z.m OH.NH 4H kH258 scam pgmHH .N NN anoz .qu mm.HH .z.H mgum 4H .oma mHCmEEoo um>ou USOHO gnaw wQHnda pm>oo.3ocm .mo mm: woo mEHH mEHa mpma :owpmHHmHomum waspwpmmsma MCHzchHm mchndpm ZHonon .mmmzmoo HHOHOHm mmmHuNmmHa-zm«m 22H HHHDHQZHm msz GOOHHHH EH omaopmzoo mmm3_mez:m mOHm3_mmaz: moneHmzoo NH mqmde 3O .vmm: 9022mm .Umm: pmhgmm .Umm: 2088mm .menH m20HpHccoo .mmmpp wCHMomzo 0H8: 88 AUSOHO 2Hp8¢8 hfidOHo 2Hpnwm nmmHo hvfiOHo hvSOHo ndeo .Hdeo 28:0Ho 888 .888 .888 .888 naoHo 888 28:0Ho 30cm 30cm ocoz mcoz mcoz 30cm mcoz mcoz 302m 050m 8:02 0202 @202 @202 mcoz qup p2MHH mumps modpe one” mcoz one 2 2H 0:02 .2.. ..w. 00H N. @0889 coupe momne 0:02 @202 02 oN 02 3 02 NN 2 mN mm on on mN mm meoz mcoz mcoz mcoz mHmmmp H 0:02 0:02 mmHmmmn J pmHnump H mmHmmmn m mcoz .pmm amuHoo H .pmm.couHoo H .208 :mvHoo H onmmn H mHmmmn H .2..» mHNm .amma; .28 mouN .28 OH.NH .2.2 mH.HH .28 83 .28 om; .28 oonNH .28 oouN .28 09H .28 $8 .28 38H .28 9.8 .28 09H .24 OO.NH .28 HS .28 83 .2.4 OONNH .28 E: .2.2 mquNH .2.2 8.3 .2.2 38 .28 OO.H .2.2 88 .2.... 8.8 .28 08H .2.2. SSH .2.2 ON.HH .2.2 SSH .28 QTNH .2.2 8. NH .2.2 00.8 .ywm mHNH .z.m mw.m H Hm mN 2N 2H 8H NH NH 0H mH . 8.2 31 E .88 .2288 OO.H 8N 88.82 m .2.... omuHH . mam Om 8H 88 ..22qu 882 2m 88882 N .28 8.2 .2. 2 ON.H .28 OH.NH .2.2 mH5H 2N o «82 ..o.H 2H 88.82 2 .28 mNnNH .2.2 $8 HN o 802 ..m.m H 802 .28 OO.H .2.2 m25H 2H 8H 88 .. .2 2N 3282 H .28 8.2 .28 mNQ 8H 8H 82m .. .2 2N 282 .2 2 8. NH 2.2 3.3 2H 8H 88 .. .2 2N .....onme N .2.2 8.2 .2 8 mHuH 2H 8H 88 .. .2 2N 8882 N .2.2 oouNH .2 2 ON.OH 0H 8H 88 .2223 .. .m on 8882 N .28 m2.2 .2. 8 09H 2H 8H 88 ..22qu ..o.m on 888..., N .28 02.2 ..2 8 09H 2H 02 82 ..m.2 H 388, H .28 88 .2 8 ON.H Q 8 802 ..o.m oN 3882 H .28 OH.NH ..2 .. 8.2 0H OH 282 ..o.m oN 3282 H 8882 H .28 mHum .2.2 09H OH 02 282 ..o. m 8 .888» H . .3282 H .28 mouNH .2.... 8:: OH SH 882 ..o.m oN 08.2 .2.2 omSH 2 .. oma 0H 8H 882 ..o.m om 88.23 H .28 092 .28 m2~H .3282 H .28 oTNH . 2.2 082 m 8H 882 .08 cm 8882 N .28 8.2 2. 2 08H .28 O22NH ..2 2 8.3 m 8H 982 ..m.m mm 8882 2 .28 88 2.8 omuH 2 8H 82 ..m.m Nm 98.2 ..28 RR 28 m2.H 2 8H 88 8H2 ..m.N mm 88.82 N .28 mH.NH 2.2 88 m E 88 .08 on 3282 H .28 oTN .28 om.NH H .82. SH 88 ..o.N mN 3282 H 28 38 28 m2; Hm m 82 ..o. N on .82 838 N .28 8% 28 m2.H on 03 802 so.N mN 802 .28 on; 28 m2: MN 8H 88 ..m. 02 982 .28 QTNH 2.2 092 8 m 882 ..2. mm .82 838 H .28 092 28 mH.N 2H m 282 ..2. on 3282 H .28 88 28 OO.N 2H m 282 ..2. on 282 .28 m2uNH .2438 2H 00H .880 302w .... mN 882 .28 38 .28 mm.m 2H 0 . 982 ..m. 8 282 .28 $3.. .28 o2uH 0H 0 282 ..m. ON 82 .28 man .28 mHnN 0H .0 .2980 £32 @8245 .8260 beam .220 mm: mom 92.2. 92.2. 3.3 vsoHo pcmoumm 2032:3098 0.3928982. 25..ansz 2:385 282282 882.28 2888 JmQHImmmHllzzdm Qza... gHDBonHm. am 80am 94 ameosazoo @323 @2222 mng 5.3222 mZOHBHono m H 2228. 32 8.0 852 8.2 H 8388 2-N .2.8 mNam .2.2 mNé mmmH .8H .88 om 282 EN 8 802 .2.8 mN,2 .2.... m2; mmmH 6H .88 m8 88 ..H-m. on 82 .2.8 mHum .2.2 02.8 mmmH .OH .88 88 2qu 888. on .882 .2.8 022 .2.8 mouH 2m8H .wN .88 882 88.5. on 882 .2.8 mHuNH .2.2. 092 2m8H .mN .88 .880 stMo and: @5th 88500 .80 «.69 won 05.8. 8.5“. 33 £898 8 203839388 305m 8.3.8.8838 we wfizmflcfim mcaphmpw 1“ 23382 522288 229282 omummmH 22.... 83-28.88 822 2558.028 BB8 8322.2 8.... 8:88:28 282 9252 28.23 22822 9288828 2.2” mama; 33 JL fi J; 33) .923 SSV 53%“: N Sukkovl .502 QW¢QTF’:- ¥ QMGQTK 0 Us...» 03 .12.: 3.3» N??R~ L22 2 Sat: >K> my . mz<4 \ ./ 5395‘ \ + / .r. .ka < , I. F Q». ”J *4- + 4. +- + 4 fl ”Q 003 ¢>\QV §>¢V \kw‘Q? Ox ++ 0 ++ +0 + o + 56 o + baa Q o o o I 0 [.0 o _D I +00 I o u + + o o m . mx<4 26565.3 I ’ ' ’ I ’ ‘ emits knxkawo). N 93.33 u kmmsikk o? a 9320 um .32 _ Qm¢o<523 + Qmwutk 0 «new.» 00% ....u 32.3. ”0.00% hm...~..n2 .53.: «K RSV? >¢BW ‘KQD‘ m. Q $5.6 u kwwwkkkg N Qua! w o u kmhh‘fiw 9< w QWKKO NR k0? Qwuuofiz: + QWoath 0 “New 00¢ ":25.“ ”0-10” $.79»? .33; 92%ka >m§huz<0 am. 9 .08..qu 2 .>> m WSEE 36 wxxbmwk 53:2» ...; 3003 @204 N QQLAO at so? Qm‘wmvfb Z: + .333». o a...» 8w "2.2.: 3.3m. T2: 3.3.: mtmmvé >m<3uzoz 3m em mm e3 83:83. geeseeez M» 5m ma mm vases nonsm>oz 3m em fl mm .2 {a m 2&3 mg 8 5 mm i=3. es; 333e,,“ poem heeeez poem 2e: sees peels: emr eegeem eefieaeeem fleece eereez e852 tease 38 e528: .mmmzeoo 205% area Be Eraozfi 9mm smearifiea N85 5825 madame mo wag ma mama. hi In 1953, in order to test the degree of agreement between the marked fraction in a shot sample and one based on sight observations of dyed tails, hunters were asked to report the number of rabbits seen but not bagged that had tails which were (1) white, (2) yellow or (3) not clearly seen. The agreement between population estimates obtained by the two methods was extremely close (Tables 16 and 17). On both study areas in 1953 the two population estimates differed by less than one rabbit. During l95h, both methods again were compared. At the Sanctuary the two estimates were 2h9.2 and 2h8.6. At the Forest the estimates were 395.9 and h13.8. In 1955 the agreement was again very close at the Kellogg Forest with the two estimates being 310.7 and 303.9. At the Sanctuary only 27 sight observations were made; consequently, the lack of agreement shown is not surprising. Thus in a total of five valid comparisons the total difference between estimates based on the two sampling methods was only 26 rabbits. This very close agreement indicates that both sampling procedures yield comparable results. Therefore, the yellowhtailed:white-tailed ratio from sight observations can be added to the tagged-untagged ratio in the hunting kill to increase the sample size upon which the marked fraction is based and thus increase the reliability of the population estimate. By dying tails at the same time that tagging is done, more efficient results are obtained with little increased labor. ‘When Lincoln Index population estimates are based on samples obtained by shooting or visual observation, several types of evidence that might indicate whether or not sources of error are being TABLE 17 h3 DATA USED TO 13.911114;st THE PRESRASON POPUIJLTIONSuKELLOGG FOREST HICKORY CORNERS, MICHIGAN Year 1951 1952 1953 195h 1955 Number bagged 267. 196 2h5 206 176 Number recently marked rabbits bagged 66 67 72 6h 81 Number tagged preseason 130 118 118 123 lh3 Number clearly seen but not bagged White 23k 232 1149 Yellow 96 97 129 Total 200 120 98 Number dyed preseason O O 117 122 lhl Population estimate based on tagged-untagged ratio 526.0 3115.2 tors 395.9 310.7 Population estimate based on sight observation h02.2 hl3.8 303.9 encountered. One would be whether or not a change took place in the marked-unmarked ratio during the sampling period. If the marked fraction remained constant, it would.seem likely that during that period (1) both marked and unmarked animals were equally sampled, (2) movement of animals to and from the study area did not cause a change in the marked-unmarked ratio, and (3) mortality was the same for both marked and unmarked animals. Data from the Kellogg Forest indi- cated eSpecially well whether or not a change in the marked-unmarked ratio was likely to occur during the samplirg period. This seemed true because: 1. Sampling was conducted over a long time (October 20 to January 31) and started the day after marking. Thus time was allowed for a differential movement or mortality between untagged and tagged animals to take place if it was going to do so. 2. Rabbit habitat was fairly uniformly distributed over the area, thus creating a situation more favorable for rabbit movements into or out of the study area than existed at the Sanctuary study area where the excellent rabbit habitat in the center of the area was largely surrounded by more open farm land. 3. More than half of the population was bagged by hunters each year. Therefore, if marked or unmarked rabbits were more likely to be collected a change in the marked fraction would certainly be evident during the hunting season. LS Table 18 gives the marked-unmarked ratios observed during the 1953, 195%.and 1955 hunting seasons at the Kellogg Forest. They are tabu— lated.here by three week:intervals, exeept for the last six weeks which are combined because of the small numbers of observations late in the season. Neither the taggedauntagged ratio in the kill nor the yellowa tailedxwhite-tailed ratio among rabbits clearly seen but not bagged showed a consistent change during the season. It had been thought that a gradual decrease in the marked fraction.might occur due to interchange of marked and unmarked animals around the edges of the area. No statis- tically significant tendency for this to take place was evident. Even the greatest drop in the marked fraction observed between the beginning and end of the season was not significantly different from that at the 'beginning. These data indicate that if such movement occurred it involved too few animals for the marked fraction to be seriously affected. It also seemed evident that marked animals did not differ from unmarked animals in their probabilities of being seen or shot or of dying naturally. A second type of evidence concerning the reliability of Lincoln Index population estimates based on a shot sample has previously been published (Geis, 1955). In that study the known number of rabbits shot on two areas also was estimated from the rabbits' trapping records by the method described by Hayne (19149) . Due to a non-uniform trap response these estimates were hS and h3 percent of the population values known to be correct. The total population estimates obtained in the same way for the two areas were hO and h? percent of the total TABLE 18 SUMMARY OF MARKEDeUNMnRKED RnTIO CHANGES DURING THE 1953-1955 HUNTING ssiSoNS--KELLOGG FOREST AUGUSTA , MIC HIGnN .14 Time Interval Oct. 20— Nov. 10- Dec. 22- Nov. 9 Nov. 30 Dec. l-2l Jan. 31 l953-Sh (Tagged 26 8 13 22 Kill (Total killed 82 he 35 70 (Percent tagged 32 17 37 31 (Yellow tail L9 l2 l9 16 Seen (Total seen clearly 133 65 5h 78 (Percent marked 3? 18 35 20 19Sh-SS (Tagged 23 l8 l6 7 Kill (Total killed 75 h9 55 20 (Percent tagged 31 37 29 35 (Yellow tail 39 21 2o 11 Seen (Total seen clearly lhl 5? 91 DO (Percent marked 28 37 28 28 1955-56 (Tagged 3h 23 17 8 Kill (Total killed 73 b3 33 23 (Percent tagged b6 53 53 35 (Yellow tail 8? 16 19 7 Seen (Total seen clearly .180 Lb 36 20 (Percent marked b8 36 53 3S Hill-In. .riastvj b7 population estimates based on the taggedeuntagged ratio in the hunting kill. Thus, there was relatively close agreement in the degree of error (1) when estimating the known number of rabbits shot from the trapping records of that partial population and (2) when estimating the total number present from all trapping records in comparison to the number estimated by means of the ratio of marked rabbits among the shot animals. If the trap reSponse of shot rabbits is representative of that of the total population, then this consistency of underestimation indicates that the total population estimates based on the tagged- untagged ratio in the hunting kill were accurate. A third type of evidence concerning the accuracy of this census method was obtained by estimating the pre- and post-hunting season populations and determining that the difference between them approxi- mately equaled the known hunting kill. During the fall of l95h, 96 rabbits were marked on Kellogg Bird Sanctuary proper. Sight observa- tions made before shooting indicated that 56% of 113 observations were of yellowbtailed rabbits which yielded a pre-hunting season.population estimate of 172 for that area. Fifty-three rabbits then were bagged or trapped of which 29 or 55% had yellow tails. Therefore, the number of yellowetailed rabbits remaining in late December was about 67. During the following May and June, 62 or 58% of 107 observations made on Kellogg Bird Sanctuary were of yellow cottontails. This led to a (post-hunting season population estimate of 116 or 56 less than the pro-hunt estimate. The agreement of the known kill (53) is regarded as evidence supporting the accuracy of this method. The agreement of hB the marked fraction in December observations (56%) with those in the shot sample (55%) and those five to six months later in.May and June (58%) is another way of expressing these constant results. All evidence gathered in this study indicated that population estimates made by the Lincoln Index method in which rabbits are live trapped, marked, released, and then sampled either by shooting or by making tail color observations, were reliable within the limits of sampling variation. Not only did quantitative evidence indicate the accuracy of this method but also its application yields logical results which were in harmony with field observations. HuntingiResults DeLury (l9hl, 1951) presented methods for estimating the size of an animal population from the decrease in yield per unit effort that occurs as the population is harvested. His methods are based on the assumptions that: l. Harvestibility remains constant. 2. The population is closed, i.g., natural mortality, recruitment and the like are insignificant. Rabbit hunting data would seem to satisfy these assumptions if these largely interrelated circumstances prevailed: 1. Hunting conditions that influence success are fairly constant. 2. Hunting pressure is heavy over a short season. 3. The fraction of the population harvested is large. h. Movements during the season.ras no influence on hunting success. 5. Unknown natural mortality is insignificant. h9 Rabbit hunting seasons in Michigan, however, are long and occur under widely varied conditions. Often hunting success is better at the end of the season than at the beginning. Consequently, the DeLury method under most circumstances could not be used to eStimate rabbit numbers here. At the Sanctuary study area, however, hunting was conducted in an atypical manner. Hunting did not start until December but then was intense. A pronounced drop occurred in the rate of kill as the harvest progressed which indicated that the DeLury census method might logically be applied. For the 1951, 1952, 1953 and 1955 hunting seasons, hunting effort data were divided into groups of approximately 25 gun-hours and the rate of kill calculated for each interval. Preseason population estimates were obtained graphically by plotting the rate of kill on the abscissa against the number of rabbits known to have been removed on the ordinate. At the point where the line fitted to these points crossed the ordinate the entire population theoretically would have been.harvested, and therefore that point indicated the pro-hunting season population level. Figures 12 and 13 show both the kill per 100 gun-hours and rabbits seen per 100 gun- hours plotted against the number of rabbits previously removed during the 1952 and 1953 harvests. Both indexes of abundance yielded about the same population estimates. In 1951 and 1955, less hunting took place and only the kill per 100 gun-hours was used as an index of abundance. Observations for these years are plotted ianigure 1h. In l95h only one hunt took place and consequently this census method could not be applied. €701 RABBITS PER mo GUN~HOURS 200‘ I704 I601 I50« [20‘ ”0' I001 504 401 20‘ I04 1 DE LURY METHOD KELLOGG BiRD SANCTUARY AND FARM FlGURE IZ FALL POPULATION ESTIMATES BY 50 H52 V . 80 130 Iio H0 Interiors and RABBTTS PER 100 GUN-HOURS I70 1 [to T 150* m ISM I20‘ I/o« laoi 90‘ w 20‘ 104 FIGURE (3 FALL POPVLATlON ESTIMATES BY DE LURY METHOD KELLooG BIRD SANCTUARY AND FARM v V V 1o so so Ibo [20 mo TOTAL KNOVN KILL 1953 51 IOO GUN-'HOURS PER KILL m HO‘ “0' :504 [30‘ [20‘ ”0‘ (00‘ ?0' 80‘ 70‘ 601 ‘Ioi 304 201 I01 FIGURE H FALL PoPuLATIou ESTIMATE 3y DE LURY METHOD KELLOGG BIRD SANCTUARY AND FARM 115:,m55 2'0 «'0 62> so :60 women nmuu nFAn do No I60 52 53 Table 19 compares population estimates by the DeLury method with those obtained by the Lincoln Index method. If the Lincoln Index values are correct, and the evidence overwhelmingly indicates that they are,then the population estimates based on changes in hunting yield consistently underestimated the population from h3 to 6b percent for the four years considered. The most likely explanation for this dis- crepancy is that harvestibility did not remain constant. Harvestibility may have changed during the shooting period because of an unequal probability of being shot among the various animals in the population. Although this census method did not yield accurate population estimates, it was fairly consistent in its degree of underestimation. Therefore, if no trapping had occurred before the shooting period at the Sanctuary study area, a fair approximation of the fall population probably could have been obtained by doubling the DeLury population estimate. To illustrate this procedure hunting statistics for the 19h6 hunting period collected by Pirnie (unpublished data) were used.‘ The DeLury population estimate from these data was 120 (Figure 15). Doubling this a preseason population estimate of 2hO was obtained which indicated a 37 percent kill. This population estimate and percentage kill is reasonable judging from observations made later when more data were available. It can be concluded that the DeLury census method yields popu- lation estimates that are far too low; ‘Values obtained seemed to fairly constantly underestimate the population and may provide re- liable indexes of abundance. 1ft... again... ITT 5b TABLE 19 LINCOLN INJEX AND DELURY PRESEASON POPULATION ESTIMATES COMPARED KELLOGG-BIRD SANCTUARY AND FARM HICKORY CORNERS, MICHIGAN Percent Error in Year Lincoln Index DeLury DeLury Method 19 51 388 135 —65 1952 365 190 -h8 1953 230 120 -u8 1955 19h 110 -L3 IOO ouu~HOUR$- KILL PER 55 FLGURE :5 FALL POTULATION ESTUMATE BY DE LURY METHOD , KELLOGG- BIRD SANCTUARY .AND FARM me 50. ”1 30¢ 20‘ 101 ‘ v ' fl ‘70 60 80 I00 /50 1710 :50 3. TOTAL KlLL 56 Indexes of Abundance ' Several types of data that could be used as indexes of abundance were collected during this study. These indexes were compared to population estimates believed to be reliable in order to appraise their accuracy. These data were of three main types: hunting kill statistics, live—trapping results and fall age ratios. Hunting Kill Statistics Hunting kill statistics were viewed in several ways. with which the total kill, kill during a constant effort, and average The accuracy rate of kill, reflect population levels was evaluated for both study areas. These indexes and the values upon which they were based are shown in Tables 20 and 21 . 1. Total kill Figure 16 shows a very strong positive correlation between the total kill and population level at the Forest. At the Sanctuary the total kill was also directly related to the population present but not as closely as at the Forest. In 1952 the kill was higher and in 19511 lower than would be expected from the known population level. These discrepancies can be explained by variations in hunting effort. 1952 had over twice the normal effort and 1951; had only about one-fourth the normal amount. It is important to observe that the total kill closely Jreflected population trends deSpite great fluctuations in hunting efiffort. The reason for this was that after about 1110 gun-hours the rate of kill was very low. Consequently, the great increases in TABLE 20 KILL STATISTICS AND INDEXES OF ABUNDANCE KELLOGG BIRD SANCTUARY AND FARM HICKORY CORNERS, MICHIGAN 57 Lincoln Total Kill/Total Hours Kill/Normal Effort Year Index Total Hours K/T-H Kill Hours Estimate Kill per 100 G. H. 1951 388 123 1h6.l 8h.2 123 lh6 1952 365 th 377.6 38.h 105 1h? 1953 230 9h 37h.0 25.1 65 167 19Sh 2h9 to 36.2 110.5 1955 19h 86 l2h.2 69.2 86 l2h2 58 a.s Hoes as m.a ammm mas Ham mmma, o.w 4am ow m.m 04mm sow can emma N.a mam Ha H.0H oamm mam Nos mmma 4.a woos ma m.w omoN was mam Nmma m.HH Nam mHH H.NH :HHN New owm Hmma A.mn.m.ooa nomv A Amnsosucdw OOH podv .m-.o oooa pmflmmw annex Hasmn m-mme mssom Haas mpeermm new» .m-.u oooH pmH\HHHe masom Hmpoaxmwflx Hmpoe Hence eonsuasaom zeonomz .aemaope . ammeoanoucsgmg--mozanzzma so mexmozH 92a mOHamHesgm AQHe Hm m4m¢a li‘ . I FIGURE I6 .59 TOTAL Km. COMPARED To FALL POPULATMN ESTIMATES «woos am: SANCTUARY AND FARM ' ( [$01 051. 05! (00‘ j 0:: i o 35' .3 < '— O "' 504 OJ"! 1% t 200 560 Too POPULATION ESTIMATE KELLOGG FOREST o 5" 501 0 5’3 - so 2001 o 5). .1 0 55 SE .1 E I50‘ 0 p. I00' 360 460 560 POPULATION effort in 1952 and 1953 caused relatively small increases in the total kills. Therefore, the total kill was a somewhat useful index of abun- dance even when the total effort was not constant; and probably was an accurate index when the hunting effort was fairly uniform. 2. Kill during a constant effort. Some of the variability shown above was due to differences in hunting effort. To avoid this difficulty the indexes considered here are based on comparable amounts of effort each year. At the Forest this effort was the first 1000 hours hunted. At the Sanctuary it was the “normal effort" of about lhO to 150 hours. This was selected as normal because it approximated the annual effort expended during the period 1932 to 1951. Because the effect of hunting pressure was being evaluated, hunting pressures other than the normal amount were applied in 1952, 1953 and 195k. During 1952 and 1953, a much heavier pressure was exerted so the kill for the normal effort was that made during about the first 150 hours of hunting. In 195h only 36 hours were hunted. Since the kill during the normal effort would have to be esti- mated by a questionable extrapolation none was estimated for l95h. In 1951 and 1955 the total effort approximately equaled the normal effort so both kills were the same. At the Sanctuary this index is expressed as the total kill so that the values are comparable to those recorded from 1932 to 1950 when hunting effort data were not available. At the Forest the index is given in terms of kill per 100 gun-hours. When the kill during a constant effort was compared to known population levels, a close positive relationship was evident on both 61 study areas (Figure 17) except for 1955 at the Sanctuary when the kill was too high. The reason for this discrepency seems to be as follows. During 1951 through 1953 most of the hunters were hunting for recreation. Some were elderly men physically unable to flush the 'maximum number of rabbits per hour hunted. Also, shooting accuracy often was poor. During 1955(and l95h when total effort was decreased) rabbits were hunted only by those people connected with the rabbit research project. .habbits were collected as quickly and efficiently as possible. The rates of kill in 195h and 1955 (Table 17) were relatively highertmanin 1951, 1952 and 1953. If there had not been a marked change in hunter effectiveness during this study, it seems reasonable to believe that the kill made during fairly comparable amounts of effort would be a reliable index of abundance. Thus the total kills recorded by Pirnie (l9h9) from 1932 through 19h? and by Staebler from l9h8 through 1950 are regarded as fairly accurately reflecting the fall population levels that existed from 1932 through 1950. It was the intention of both of these successive Kellogg Bird Sanctuary directors to keep hunt- ing pressure constant each year and it is very unlikely that small variations in the effort from year to year could have greatly influenced size of the kill. This is especially true when it is realized that since the rate of kill at the end of the shooting period is low, even moderately large fluctuations in effort would influence the total kill very little. On both study areas the kill during a constant hunting pressure was apparently'a satisfactory index of fall rabbit abundance. 62 FIGURE I? “U. DURING CONSTANT HUNTlNG EFFORTS COMPARED I0 FALL POPULATION ESTlMAYES KELLOGG auto "SANCTUARY AND FARM I251 ." a g I00‘ 05'). la b 055' 75‘ 4 < Z on a: 0 5M 2 > SE —\/\ 500 {00 :00 POPULATlON ESTH‘MT E KELLOGG FOREST |OOO HOURS KILL/100 GUN nouns ouame FlRST lly H‘ ‘0' 05! 051 039 051. o I" a/s v . . soo ‘ «on 500 POPuumow ESTIMATE 63 3. Average rate of kill. At the Sanctuary the rate of kill based on the total kill and total effort indicated very little correlation with population density (Figure 18). Wide variations in hunting effort and the change that occurred in hunter effectiveness in l95h and 1955 was responsible for this. In contrast, rate of kill at the Forest showed a close corre- lation with population density. This would be expected since the effort was fairly constant each.year. ‘When based on wide variations in total effort the rate of kill is an unreliable index of abundance. Trapping Data Four indexes to fall rabbit abundance were considered for both study-areas: (1) total individuals tagged, (2) total captures(including repeats), (3) individuals handled per 100 trap-nights, and (h) total captures per 100 trap-nights. These data are tabulated along with population estimates in Tables 22 and 23. At the Sanctuary (Figure 19), the total rabbits captured seemed to be a fairly reliable index to fall abundance. The total times rabbits were handled was less satisfactory because 1951 was high, and the two rates of capture also did not showra regular pattern when compared with population size. The chief reason for the unreliability of these indexes was the very high rate of capture in 1951. A heavier snow cover then perhaps was responsible for this. Allen (l938b)indicated that snow cover had a marked influence on trap success. These data indicate that the total number of animals captured probably accurately FIGURE :8 _ 61‘ IRATES 0F KHL COMPARED TO FALL POPULATION ESTIMATES Q KELLOGG BIRD. SANCTUARY AND FARM of. 0 I00‘ 0 '; m 75I d 056' 3 o I $4 1 ‘ 052 y. a 251 0‘3 §> ; N‘ - , f t 200 ' 300 II00 POPULATION ESTIMATE KELLOGG FOREST '2. 05'! III IOI of.) 9‘ '05). .5, g 8‘ .55 w 7‘ ‘I o 5‘ z _1 ‘14 E. a o l— 2. \ :3 |‘ .; 1 v ‘ v '_‘P 300 900 500 POPULATION ESTIMATE 65 s.e m.m mesa amH am sea mmea s.oa H.o moefl wow ONH esm smea s.e o.m woNN mam mod omm mmafl w.oa m.s meow cam ems mom mmea m.mm w.m mmwa mo: Hoe wmm Hmma mesmez arse essmnz ease mesmez seduces esessdm sesssaeo sameness ease ooa\eesssdeo coe\.esH sees moses asses .esH Hesse seeseasaea 1| 5&8: .meezmou @81on NN mqmda swam Qza Mmdoeozmm QmHm.OOOqgmmlumozz<3hvzoomm emsm>oomm sopssz empqua poems when OH mama qano>oomm new mswpcdflm pmppoaom ambEdz psooamm popesz open mange? hnm>ooom somapmm Hw>nmesH mafia nonfisz no apHHHnmnona zaOHmon «Hmaoba omlmmmatugzmaHmumNm mmoq OZHHmmHmo awmgoh_oooqnmm mme mo HMdESQm pm mumda 91 planting was .00125 for the October 30 planting, while the mean prob- ability for the December 2b and 26 plantings was .00266. Thus the December plantings were more than twice as likely to be found than were the October plantings. This difference was statistically signifi- cant.. But becauSe much less hunting was done in December the overall percentage recovery was only slightly higher than in October, despite the higher probability of recovery. Table 28 lists the recoveries of non-planted rabbits. It indicates that there was from seven to nine true crippling losses found. The exact number is not known since circumstances indicate that one rabbit could have been lost from a hunter's coat while another possibly could have been hit by a car. It seemed reasonable to adopt eight as the number that were recovered. The following proportion then could be used to estimate the total crippling loss: number of planted rabbits found a number of true crippling losses found total number planted total crippling loss ZFrom the values observed in this study: .29....8 103 ‘2 X = 27.5 Confidence limits around this value indicate that between 20 and 38 .cripples were lost unless once chance in 20 had occurred. Thus the crippling loss was from eleven to 23 percent of the known kill. Since this estimate is minimal because some cripples probably die in locations 92 TABLE 28 SUMMARY OF NON-PLANTED RABBIT RECOVERIBS, KELLOGG FOREST 1955 AUGUSTA, MICHIGAN Date Tag Number Comments Oct. 22 325 Reported and found Oct. 23 ? Reported Oct. 25 -- Possibly hit by car Nov. 5 -- Hit by car Nov..2h -- Brought in Nov. 27 -— Brought in--shot holes in ear Dec. 27 -- May have been lost from hunter's coat Jan. 1 3329 Reported and found Jan. 3 -- Brought in--found dead sitting in form Jan. 11 3bhh Brought in 93 that make their recovery highly unlikely, it is probably safe to assume that 20 percent of the known bag were killed by hunting but are not recovered. The dense cover and heavy hunting pressures at the Kellogg ‘Forest created a situation that would be expected to favor a high crippling loss. Judging from the results of this experiment, the number of cripples found can be multiplied by four to obtain an estimate of the total crippling loss. If a significant portion of the cripples that die do so in places where their recovery is not possible, even this correction factor may be too small. The reliability of this method of estimating the crippling loss depends on the uniformity of the followe ing factors. 1. Cooperation of hunters in reporting dead rabbits. 2. The accuracy of hunters in reporting the locations of dead rabbits. 3. The ability of game area managers to find reported rabbits. h. The proportion of crippled rabbits which die in places where it is unlikely to find them. The last point merely influences the uniformity by which this pro- cedure underestimates the total crippling loss. Since about 35 rabbits probably died as a result of shooting but were not bagged, it is some- what surprising that only eight rabbits were reported to have been crippled and escaped. This indicates that many rabbits are probably fatally hit but are assumed to have been missed. This view is supported in that frequently hounds catch or find crippled rabbits that showed 9h no evidence of being hit when shot at. The significance of these findings concerning the crippling loss at the Kellogg Forest is that they indicate that hunting caused the death of about seventy percent of the fall population during each of the past four years. Unless the rabbit population there soon fails to maintain its present level, fare is additional strong evidence that even very severe hunting kills do not limit the populations in succeed- ing years. Non-huntinnginter Mortality It is difficult to accurately determine the non-hunting winter mortality because spring population estimates are only approximate. If it is assumed, however, that about half the population is handled during the spring trapping period (see Table h) and that the crippling loss is ten percent of the hunting kill, then non-hunting mortalities on the Sanctuary for the winters of 1951-52, l9Sh-SS and 1955-56 were 29, Sh and L5 percent reSpectively of the fall populations not killed 'by hunting. l9Sh-SS had a very light hunting kill of 22 percent while "the other years had moderate hunting kills of 32 and hh percent. Thus it seems that the light hunting kill was associated with a higher non- hunting mortality than occurred during years of heavier kills. No estimates of the non-hunting mortality during the winters of 1952-53 and 1953-Sh could be made because very few animals were trapped in the spring. Since Spring population levels seem to be strongly dependent upon the extent of the previous year's hunting kill, it seems likely that 95 the hunting kill was the most important source of winter mortality on the area except possibly for years when that kill was very light. Age Composition of the Sanctuary Rabbit Population The age composition in the hunting kill indicated the high mor- tality rate in the rabbit population (Table 29). The numbers of animals in each age class except juveniles was based on the age at which shot rabbits had been trapped. Since no trapping was done before 1951 the age distribution then can only be divided into the juvenile and one and one-half or older age groups. In 1952 six rabbits were shot that had been marked as adults in 1951, thus they were two and one-half years old or older. Every year a number of adult rabbits of unknown age were shot. Table 30 tallies the age distributions with those animals in the one and one-half years or older category prorated into one and one-half and two and one-half and over categories according to the ratio observed in 1952. This may have tended to exaggerate sur- vival because 1952 followed only a moderate hunting pressure. Based on five year totals 82 percent of the winter population was made up of juveniles, lb percent rabbits one and one-half years of age and.only four percent two and one—half or more years old. There was no record of any rabbit actually living longer than two and one-half years. Only 17 percent of the juveniles alive one winter survived until the next winter and only about five percent lived until the second winter. Atzenhoefer and Martin (l9h9) found survival rates of 12.5 and five percent over comparable periods of time in Ohio. It can'be TABLE 29 DISTRIBUTION OF RABBIT AGES IN HUNTING KILLS KELLOGG BIRD SANCTUARY AND FARM HICKORY CORNERS , MIC HIC‘mN 96 Age in Years Year JUV 17172 1 1/2 or more 2 172 or more 1951 100 O 23 1952 1ho 16 31 1953 103 3 9 l95h SO 0 3 1955 73 8 h TABLE 30 DISTRIBUTION Of" RABBIT AGES IN HUNTING KILLS KELLOGG BIRD SANCTUARY AND F ARM HICKORY CORNERS, MICHIGAN Age Year JUV 1 1/2’ 231/2 or more 1951 100 17 6 1952 1140 39 in 1953 103 10 2 195h 50 2 l 1955 73 11 1 Totals h66 79 2h Percent 81.9 13.9 h.2 97 concluded that the life expectency of a rabbit that lived to the first winter was less than a year. Also, the turnover rate (as defined by Petrides, 1951) of a population was less than four years. 'When the age distributions in years following heavy (1953 and l95h) and following moderate or light kills (1951, 1952 and 1955) were combined (Table 31), it was apparent that the rate of survival was less during years of heavy hunting pressures. This difference was statistically highly significant. (It seems likely that differences in hunting pressure were responsible.) It has previously been shown that Spring population densities were strongly influenced by the size of the hunting kill during the previous year. These age distribution differences further indicate that light hunting kills were not com- pletely compensated for by non-hunting winter mortality. 1951-52 Life Equation Observations. A major objective during the first year of this study was to establish a table of animal gains and losses for Kellogg Bird Sanctuary and Farm cottontails. Such a table has been termed a life equation (Leapold, 1933) and shows the relative magnitude of mortality at various times during the year. Most population values in the life equation were obtained by (Lincoln Index population estimates and were calculated for several times during the year. These values are summarized in Table 15. The production of young was estimated as the number of adult females times the average number of young produced by each adult 98 TABLE 31 AGE DISTRIBUTIONS FOLLOWING YEHRS OF I-IEENY HUNTING KILLS ‘ CQIPARED TO THOSE FOLLOWING LIGHTER KILLS Following Heavy Following Moderate or Light Age Hunting Kills Hunting Kills Number Percent Number Perc ent Juveniles 153 91 313 78 1 1/2 years 12 7 67 17 2 1/2 years or mo re 3 2 2l 5 Totals 168 100 101 100 99 female. The average litter size of eight observations made in this study was 5.6. Allen (1938a)working at the Kellogg Bird Sanctuary found that the average litter size to be 5.1. Haugen (l9h2) determined that the average litter siZe in twelve litters in Allegan County, Michigan, was S.h and Trippensee (1936) also working in southern Michigan.ca1cu1ated the figure to be S.Oh. Five young per litter was judged to be a sufficiently close approximation for practical use. The number of litters per female per year has been reported as follows: Four (Bedell, 193k); three or four (Leopold, 1933); two common, four possible (Trippensee, 1936); two or three (Dalke, 1937); two certainly, three probably (Gerstell, 1937); two (Allen, 1938a); five as a maximum (Hickie, l9h0); and 3.8 (Schwartz). Hangen (l9h0) came to the conclusion that three or four litters a season are common in Michigan. Three and one-half litters per year was adopted as a reason- able figure for use in calculating production. The hunting kill at the Sanctuary was known and crippling loss was estimated at ten percent of the kill. The extent of mortality was estimated from the differences in population estimates for various times. Table 32 gives the resulting life equation. The difference of three rabbits between November population estimates based on all observ- ations and that obtained by adding the estimates for juveniles and adults was due to sampling variation. So that the relative extent of mortality that occurred at various times during the year could be appraised, Table 33 was constructed. This lists the percentage mortality that occurred at various times lOO .oEHp oqo pm m>HHd ohm: sows: Mo Ham pozs oeH MHH oeH mpHseH NmmH mgHeam mmm NH NH mmoa msflaadfiwo mom mmH mmH HHHm qupesm usages mmm mam meme dam ponso>oz Hmm mm mw wedded honfioboz om: Hm me moHHcm>sn nobsoeoz pm: 0mm mam monso>sw pmsmsa ammoa mam mam chop mgow .8856. B mswaam ©0H QOH mpHsea HmmH .mese haemopso cofipwasaom whoa cams :ofiudasaom coflpdficomon no mm< mama Haeoe no maHm eopmHsoHao mm mHmae zeeHmon .memzeoo_»moe0Hm mm-HmmH .meHmmam seam aza emapeozam mm m woosqmm ZOHHaaam mmHH TABLE 33 RELATIVE IMPORTANCE OF MORTALITY AT VARIOUS TIMES, 1951-52 KELLOGG-BIRD SANCTUARY.AND FARM HICKORY CORNERS, MICHIGAN lOl Percentage of Percentage of Mortality Total Annual Population Category Mortality Lost Between birth and age 2 l/b months 6S.h 63.1 Between age 2 l/h months and November 3.h 8.9 Adults between June and November 3.8 32.h Hunting kill 13.5 31.7 Crippling loss 1.3 3.1 Non-hunting winter loss l2.h 29.1 102 based on the total annual mortality and the population to which it applied. It can be seen that the greatest mortality was among juveniles between birth and 2 l/h months of age. This is probably a minimal estimate because the production of young by juveniles in their first year was disregarded. It is clearly evident that the size of the fall rabbit pOpulation is largely determined by the success of juvenile production. Two other points also are of interest. A rather large number of adults died between June and November. And even during this year of "light hunting pressure, the shooting kill made up the biggest part of the winter mortality. If sufficiently detailed data were available so that life equations could have been constructed for other years of this study, it seems likely that they would differ from that in 1951-52 in several respects. During years following heavy hunting kills juvenile production or survival would have been higher. During years with heavy hunting kills the overall winter mortality would have been greater but the non- hunting winter mortality would have been numerically less. 103 HUNTING EFFORT AND SUCCESS AT THE KELLOGG FOREST I. Introduction The Kellogg Forest is a very heavily hunted public area on which the cottontail rabbit is the principal game animal. Hunting data on this area were collected to: (1) determine patterns of hunting effort, kill and success, (2) determine the distribution of effort, kill and success among the hunting public, (3) evaluate factors influencing hunting effort and success, (h) predict the probable effects of hunting seasons of various lengths, and (S) explain why rabbit hunting statistics display seemingly illogical relationships. The chief reason for seeking this information was that hunting statistics from this area indicate what can be expected from other similar public hunting areas. This is especially true because the Kellogg Forest is located on rough, very poor land that is typical of the sort that falls into public ownership. A serious problem in public wildlife management is the provision of hunting areas for an constantly increas- ing number of hunters. II. Methods Hunters were required to obtain a hunting permit each.year and to report at the Kellogg Forest office after each hunt. For the l95h season, the hunters' occupations and whether or not they had hunted the area during the previous season also was determined when'permits were issued. 10h The data collected after each hunt varied somewhat each year. Every year, however, the starting time, finishing time, party size, numbers of tagged, untagged animals bagged and tag numbers of marked rabbits were recorded. During the 1951 and 19st seasons information on the use of dogs also was gathered. For the l95h season, each hunter was assigned a permit number. Hunter's names, permit numbers and individual success was recorded each time they checked out. The following weather data were collected at the Kellogg Forest as a routine procedure: maximum and minimum daily temperature, kind and amount of precipitation, snow cover, wind velocity and cloudiness. The hunting data were organized as follows: The daily hunting effort and kill werecross tabulated by week of the season and day of the week. These tabulations are shown in'Bables 3h, 35, 36 and 37. A small amount of effort and kill registered by neighbors who reported only season totals could not be included in these tabulations. This accounts for the discrepency between the total annual kill and effort shown in Table 37 and that shown in these tables and others which depend upon knowing when during the season effort and kill occurred. The daily and weekly cross tabulations permitted the daily and weekly patterns of effort, kill and success to be readily determined. For the 1951, 1952 and 1953 seasons the hours hunted, kill and kill per 100 gun—hours experienced each day during the season was written on and punched into a punch card along with the following weather data: minimum temperature, maximum temperature, total precipitation, total snow fall, depth of snow on the ground, average wind velocity and cloud cover. This was done to 1C5 ongoglgc I 939.0 fl. omN Ha m0 Hm eN NH NH oN HHHe m.mHHN w.Naa 0.eam N.HmH N.NeH o.oNH e.mHH m.HNN .m.0 oHocoe a H 0 N N H 0 H HHHe N.0m 0.mH 0.e 0.m w.m 0.o 0.H 0.e .m.0 Hm-eN .oo0 HH N 0 H 0 0 H 0 0 HHHe N.0N a.oH m.0 0.0 0.0 0.m 0.0 0.0 .m.0 NN-NH .oo0 0H d H H H 0 e 0 N HHHe N.00H 0.NN m.e m.m m.m N.am m.HH 0.e .m.0 oHuoH .oo0 e oH m m 0 0 0 e e HHHe m.eeH N.mm m.me N.H N.H N.0N 0.mN 0.aH .m.0 mum .oo0 m 0N o e N N N e 0 HHHe 0.mmH N.Ho m.0o m.HH m.eH m.HH m.0H m.» .m.0 N .oo0uoN .eoz a Nm HH a m N 0 N N HHHx N.HoN m.00H m.ao w.mm m.0m 0.4 a.mH ®.e .m.o mN-eH .soz o oH m e 0 m e 0 0 HHHe m.amH 0.0o 0.em 0.5 ®.m m.mH 0.0 0.0H .m.0 wH-NH .eoz m as mH NH H m 0 0 a HHHe m.meN 0.00H m.0e m.m m.oH m.N m.NH 0.HN .m.0 HHnm .>oz s cm NH HH N e N H N HHHe m.eoN ®.mHH m.oe N.@H m.mH 0.0 a.mH N.mN .m.0 e .eoz-eN .so0 m 0: MN HH e N 0 0 0 HHHe . e.mNN 0.NmH e.ea N.eN N.mH N.a m.H w.e .m.0 mN-NN .oo0 N mm s 4 HH N m H m HHHe 0.mHe w.ae 0.5HH ~.eN N.mN m.MH m.mN m.m0H .m.0 HN-mH .Ho0 H nHocoe .ssm .eom .Hwa .ncdea .ens .noee .soz open some zs0Hr0Hz .aamaopa ammmoiflo00HHmeuuHmmH HHHe aHmmam mza emoasm ozHezem HHHeo an Emma. 106 14:" medogncso u .m.o: an. .mnH mm mm mH mH NH mH oN HHHe .emoN .H0o m.mmm m.NoH m.N5H ®.0eH 0.55H m.eoN .m.o nHeeoe a H m m 0 0 0 HHHH 0.No 0.eN m.HH m.NH 0.m 0.0H 0.H .m.0 Hm-cN .eoe mH H 0 0 0 0 0 0 H HHHe m.m5 0.NH N.NN a.HH 0.m 0.3 0.0 m.5 .m.o mN-mH .ece 4H H 0 0 0 0 0 H 0 HHHe m.me 0.eH 0.0 m.0H 0.5 0.w 0.m 0.m .m.o mH-NH .eoe NH e N e H 0 0 H 0 HHHe N.Nm m.oH 0.NH N.N 0.0 0.0 0.HH 0.: .m.0 HHum .coe NH oH o 5 0 m 0 0 0 HHHe m.5oH m.5m m.oe 0.5 0.NN 0.m 0.NH m.o .m.0 e .seenmN .oo0 HH : H m 0 0 0 0 0 HHHe - 0.mm 0.mN 0.0m 0.5 m.5 0.0 0.0 m.e .m.0 NN-NN .oom 0H m N 0 N H 0 0 m HHHe N.Hn 0.oN mm.e 0.o 0.m 0.0 0.5 m.eN .m.0 Hm-mH .oon e eH e H N 0 0 m e HHHe m.®mH m.wm m.0m m.HH 0.5 0.0 m.0N m.w .m.0 eHnm .oom w oN HH 4 H N e m H HHHe N.NwH N.e5 m.mm m.5 0.NN m.mN m.5H 0.o .r.o 5-H .ooa 5 a m e H H 0 0 0 HHHe m.HH 0.NN 0.Ne m.NH 0.e 0.0 0.0 N.NH .m.0 0mteN .eoz o HH m N 0 H H N 0 HHHM N.wd 0.HN 0.mN m.w m.m 0.©H 0.5 mN.m .m.o NN-5H .>oz m a H m 0 0 0 N H HHHe N.NHH a.Hm m5.mN ww.m 0.H 0.w m.mH m.0N .m.o oHaoH .>oz e mN HH N m m N N e HHHe N.emN 0.e5 m5.em 0.eH 0.NH N.NN m.eH 0.:m .m.o m-m .>oz m MH m H H 0 N m H HHHe m.moH m.5N m.mm w.w N.HH 0.5H N.mH 0.eN .m.0 N .soz.5N .so0 N mN e m H H m 0 HH HHHe m.eNe 0.NOH m.Nm N.oN m.Nm 0.eN m.em 0.50H e.m.o oNtoN .eo0 H nHoeoa .ssm .eem .Hsm .nssse .ooz. .noea .eoz open some smme0e_000HHeet-HHHe aHmmHm 024 emoamm ozHezem HHHHQ mmmHuNmmH 262on 49203 no flansa n55$c30..ao4 107 NNN oH as om NH mN a oH HHHe 0.0HmN m.mHH N.Hom m.mHm 0.N0N w.oHN 0.JMH m.eHN .m.e nHocoe 5 e 0 H N 0 0 HHHe N.oe N.5e m.m m.mH m.NH 0.0 m.w .:.o Hm-0N .soe mH o. 0 m m 0 0 0 0 HHHe .eo 0.0 m.eN m.NN 0.N 0.0 0.0 m.d .m. mNth .coe 4H oH 0 N m m e N 0 HHHe 5.50H 0.0 0.mm m.mH 0.NN N.mH m.o 0.0 .m.0 mHtNH .eoe mH m 0 e H H H H 0 HHHH ®.m0H 0.0 m.e5 0.mH m.m 0.e N.o 0.0 .m.o HH-m .soe NH 5 0 m 0 H 0 0 H HHHe m.5mH m.mH 0.5m m.5 0.mN 0.NH m.m 0.eH .m.o e .eoe-eN .oo0 HH oN N wH m 0 N H 0 HHHH m.m5H N.mN N.He 0.em 0.0 0.oH 0.m 0.: .r.0 mN-NN .oo0 0H 0H 0 N w 0 0 0 0 due m.Nn 0.0 0.em N.Hm 0.0 ®.e 0.0 0.N .m.0 HN-mH .oo0 e o H N N 0 H 0 0 HHHe ®.0HH N.r 0.5H m.oN 0.0N m.N 0.N m.e .m.0 eH-N .oo0 m NH H e m 0 e H N due N.NNH m.mH 0.me 0.mm 0.N m.mH N.m 0.© .m.o 5-H .oo0 5 oN m eH o 0 m 0 0 HHHe m.mmH 0.NH N.No m.0m N.HN m.me N.cH 0.H .m.o onueN .>oz o o m H 0 0 H H 0 HHHe . m.5o m.5 m.© w.mm 0.0 0.NH m.H 0.H .m.o «N-5H .eoz m eH N m m H H H H due m.HmH m.m 0.Nm 0.He 0.N 0.0H w.NH m.NN .m.0 oHtoH .soz s 5H m 5 m 0 H 0 m due w.m0N m.w m.mw 0.®m 0.5H 0.eH m.d N.eN .m.o can .>oz m Nm 0 oH m N H 0 e HHHe m.N0m 0.mH m.5HH m.55 N.5N m.wH 0.4N m.NN .m.0 N .eoz-5N .coo N mm H 0H d N e N m HHHm 0.00m m.NH 0.mHH N.cm m.eN w.0N 0.mm 0.0m *.m.0 oNnoN .poo H oHscoe .eoz .sem .som .Hta .onsre .ooz .ooea . econ more ‘1 I.“ l zaonomz .Hem002a emNm0a_000HHmH--HHHe BHmmHm 02a aromas ozHBZDm HHHH0 emeH-mmeH om mama; 108 2:97:50 n .m.o s NNH NH N No mm oH 5H NN HHHe N.NNNN N.5mH m.om 0.5NN N.50o mN.5NN 0.NNN N.:NN .m.0 oHosoe H 0 0 0 H 0 0 0 HHHN 0d: 0.0 0.N N.HN N.NH 0.0 0.N 0.0 5.0 HNuoN .dee NH 0 0 0 0 0 0 0 0 HHHe N.NN 0.0 0.0 0.0 0.NH m.oH 0.: 0.0 .m.0 NN-NH .coe :H N 0 0 N 0 0 0 0 HHHe . m.0: 0.0 0.0 N.NN m.0H 0.0 0.N mN.N .m.0 NH-NH .coe NH H 0 0 0 H 0 0 0 HHHe 5.NN 0.5 0.0 0.N N.5H mN.H NN.H 0.0 .m.0 HHnm .eee NH N 0 0 0 0 N 0 0 HHHe N.NNH 0.N 0.H N.NN N.5H N.No 0.0 0.HH .m.o : .eoe-NN .oo0 HH NH H 0 N 0 N 0 o HHHH m.oNH 0.:H 0.0 N.Nm 0.: 0.:N 0.5H m5.:H .m.0 NN-NN .oo0 0H :H N 0 o : 0 0 N HHHe N.5NH m.0N m.0 N.5N m.N: m.N N.HH N.NH .m.0 HN-mH .oo0 N 0N 0 H 0H m H N H HHHe N.:NH 0.0 N.NH N.5N N.m: 0.oH m.N N.NH .m.o :H-N .ooo N HN N 0 o 5 N : 0 HHHe N.oNN 0.NH m.o 0.5N 0.No mN.5 0.0N m.0H .m.0 5-H .ooo 5 5H N 0 N H H m 0 HHHe N.N:H 0.0N 0.0 0.N: 0.5N m.NH N.oN 0.0 .:.0 0N-:N .>oz o 0H N 0 : N 0 H 0 dam 0.:0H 0.N m.N N.o: N.H: 0.0 N.0H .H .m.u NN-5H .eoz m NN N H HH N 0 N N HHHe 0.::H N.m N.H N.05 m.:N 0.H 0.: N.5N .m.0 oHtoH .soz : NH N H m o N H H HHHN N.0NN N.NH N.HH N.NN m.o: N.NH N.NH N.5N .m.0 N-N .eoz N 5N H N .N HH 0 0 : HHHe 0.NNN N.NN N.NH N.NN N.NN m.m N.NN N.NN .m.0 N .sozn5N .soo N 0N H 0 m NH : N m HHHe N.0:: N.NH 0.H 0.NHH N.NHH N.om N.Nm N.5o s.m.0 oN-0N .coo H oHosoe .oose .eoz .eom .eom .Hce .ntdea .ess open soc: 33:82 .1880... emma0a_o00HHNeu-HHHe BHNNHN mza Broads ozHezem NHHHQ NNNH-:mNH NM H.513 109 permit climatic factors which might influence hunting effort and success to be evaluated. In l95h a different punch card was made out for each hunter that indicated the daily record of hours hunted, kill, dog use, effort with snow cover, the total effort and total kill. The fraction of total visits on which dogs were used or snow cover present was indicated on the card as well as the average visit length and kill per 100 gun-hours. Punched into each card was: 1. hunter occupation by a classification outlined beyond 2. whether or not the hunter had obtained a permit the previous year 3. number of visits h. total hours hunted 5. total kill 6. number of cripples 7. week numbers during which hunting was done 8. kill per 100 gun-hours 9. average hunt length. These cards were used to determine the effect of hunter occupation, previous experience with the area, number of visits and hours hunted on hunting success. They also revealed the distribution of hunting effort, kill and.success among the hunters. In l95h another type of punch card was also made out for each hunting party. These had recorded on them the: date, number and types of dogs used, party size, number of unre- covered cripples, kill, starting time, finishing time, hours hunted and 'the total gun-hours for the party. Punched into these cards were dog 110 use, amount of snow on the ground, whether it was a Saturday, Sunday, ' weekday or opening day, and the period of time during which the party hunted. From these cards the hourly distribution of hunting effort on Saturdays, Sundays, weekdays and opening day was determined as well as the effect of snow cover, party size and dog use on hunting success. III. Patterns of Hunting Effort, Kill and Success A. Yearly Table 38 summarizes hunting effort and success for 1951-5h. Hunting pressure each year was slightly over 2000 gun-hours. The effort expended in 19Sh averaged b.56 hours per acre. Total kills varied between 196 and 256. The mean yield over the four year period was 9.9 rabbits per 100 gun-hours, or about ten hours of hunting per rabbit bagged. Success rates seemed to be strongly associated with the size of the kill. Apparently the number of hours hunted during the entire season was not associated very strongly with either the kill or rate of kill. ‘With other things equal the lowest total kill and highest rate of kill would be expected to take place when the least hunting effort was expended. Contrary to this expectation, 1952 which had the least effort experienced the second lowest rate of kill and l95h which had the greatest effort had the second lowest kill. From the data in Table 37 it is apparent that some other factor influenced the hunting kill and success besides hunting effort expended. This matter is discussed beyond where success data and total kill are evaluated as indexes of abundance. TABLE 38 111 YELRLY HUNTING EFFORT, KILL mu success-..KELLOGG FOREST AUGUSTA, MICHIGAN Effort Kill/100 Year Hours Kill Gun-hours 1951 leb 267 12.1 1952 2056 196 8.9 1953 2310 2h5 10.1 195h 2 3A9 206 8 .8 112 Although the rabbit open season was increased from 77 days in 1951 to 10h days in 1952 no increase in hunting activity occurred. In fact 58 fewer hours were hunted in 1952 than in 1951. The average annual effort during the three years of th-day seasons was 2238 hours in contrast to 211k hours hunted during the 77-day season in 1951. Thus a 35 percent increase in season length was accompanied by an average increase in effort of only 5.9 percent. It should be pointed out, however, that the rabbit population was higher in 1951 than in any of the following three years. Effort in 1951 might have been less if the population level had been comparable to those in the following years. B. Monthly The monthly distribution of effort, kill, and success is tabulated in Table 39 for the individual and collective 1951-5h seasons. There was a regularly decreasing average hunting pressure of h7.1, 25.0, 18.2 and 11.3 hours per day respectively during each month from October to January. Hunting success over the four year period averaged highest (11.5 rabbits per 100 gun-hours) in November, next highest (10.1) in December, next in October (8.6) and was lowest in January (7.9). A better compari- son was probably obtained when 1951 was eliminated from consideration because it had no January hunting and five more days in October. November and December are equally good with October success slightly poorer than in January. Probably because of seasonal changes in vegetation and snow (see beyond) the rate of kill in January was greater 113 m.a m.m H.6H a.m H.~H .e.w ooa\aewe mm.HH®m om.rmmm OO.OHMN oo.amom ma.mesm nnnom .Han .eea .mmm .mmH .omm Hana annoy m.a e.a m.m o.m N.m eoao eoz .e.m ooa\aaam om.mmOH om.mmoa oo.®mH mm.oer ma.0mm eoao eoz manor .mr .mn .m .mr .mm eoao eoz Hana anoneoe fl.oa H.6H H.a w.HH m.a o.oe .e.m ooe\eaae 00.4mma OO.NoNN mw.aaa om.mmm ma.mmm .mme manor .ewfi .wmm .0a .No .mm .44 Hana nongoooa «.6H m.HH 4.0H m.oe o.m m.rH .e.m coaxaaae mm.m~0m mm.~mmm om.qao oo.ama m~.::b 00.4mm mnsom .HHN .mrm .ON .aa .mo .ema Haae .aoQEoeoz m.a o.m o.r ®.e a.o r.oa .r.m ooa\aaee ma.eeafi Nr.oanm mm.nmo ma.mmm ma.eom ao.Hma manor .ama .mHN .rm .ar .rm .ra Hana noeoeoo owmmmw oaoeoa mmnrmaa rm-mmma mm-~mea Hmafl ergo: ZMOHmoez .aamooea anemoa oooque--mzomemw emmH-HmmH Mmezzpm mmmooam gze que .amoeam Mnrazoz mm mamaa 11h than that in October even though.by January each year close to 50 percent of the population had already been bagged. C. weekly A consideration of weekly fluctuations in effort, kill and success 'probably‘best shows the changes that occur during the season. Figure 23 shows the total weekly effort and kill for the 1952-5h seasons combined. 1951 was not included because different opening and closing dates were then in effect. Peak effort occurred during the first week (October 20—26) then gradually tapered off until the fifth week (November 17123). Effort sharply increased for the next two weeks reaching a second peak during the seventh week, (December 1-7). This second peak was slightly less than half as high as the first. Effort then tapered off until it was uniformly low during the 12th week through the end of the season (January 5-31). Late season effort was very light compared to the expended early in the season. During each of the last four weeks effort was only about 1h percent of that regis- tered during the first week of the season. Figure 23 shows the close association between the weekly kill and the effort expended. Easentially the pattern Of kill was the same as for effort. Peaks occurred during the first (October 20-26) and seventh (December 1-7) weeks, however, the second peak was about 70 percent as great as the first. The reason for this relatively greater late peak in kill was the greater rate of kill experienced during the seventh week. Because of the low effort during the last four weeks the kill was also low. A total of only 61 rabbits were killed during the last 115 3.: m_ 1 a I171. mdé. t K I :5 a; “2-3 :.m .z w~,NN 2-3 I-m duo 8,3 m_ .2 : o. a m N o xl ,. , A V/l/U—‘l. tivfllw r/ 7/ T 4/ H w. W 7 ., /,r A i 71 r /,_, c J A f n 7/ a t rL 825:8 wzomfim rm: + Nmr . j; in; a: EE: 2. were: ¢-m N>oz £2 r. s. E ow memod 00013. 02; 2:: Evan; 2-: 3-2 .52 -250 ..SO when m r m N _ .oz xmw) o E //U m/ N A . / rt 1. d 3 0 HO 3 3 N l m 0 ..J o. l O l V n) 116 four weeks of the 1952-5h seasons while the total kills for the lst, 2nd, 3rd, and 7th weeks all exceeded that level. The hunting success as measured by rate of kill is indicated in Figure 23 by the difference in length'between the kill and the effort bars for each week. For example,in the first week the effort bar is much longer than the kill bar indicating a relatively low'rate of kill. In contrast the seventh week had a kill bar longer than the effort bar indicating a much higher rate of kill. Figure 2b shows the weekly variation.in hunting success in terms of kill per 100 gun—hours. It indicates a gradual increase in success between the first and seventh weeks and then an irregular decrease for the rest of the season. The last week experienced a slightly higher rate of kill than did the first week; Table to gives the numerical data on which.Figures 1 and 2 are based and provide additional details. Since fluctuations in these values were generally similar each season the total mean values used in Figures 23 and 2h probably give a reliable indication of the fluctuations that can be expected to occur in the future. It is noteworthy that hunting success did not decrease as the season progressed. It might be expected that the rate of kill would be proportional to the population present and that consequently it would decrease as the population was reduced in size by hunting. This was Observed to happen on the Kellogg Bird Sanctuary and Farm where hunting did not start until December and then was conducted in an intense manner under fairly constant conditions. The reasons for this lack of a 117 M3332 rum} omzfiiOu $20w23) 4‘ 9 : N. 33d TH)! 5800*!“ N09 00! 118 .Hm awesoooa oomoao one ma ponopoo oocoao condom MGHpcsn one cos: Hmma new names: x. OOOO OHOO OOON HHON OOON OHHN musoO OHO HOO OOH OON OOH OON HHOO anOoO 0.0 O.N OON OON OO OO NO musoO 0.0 O.N O.N OH OH H O O HHOO OH HOuON O.N O.N OOH OOH OO OO OO nusoO O.O H.H O. O O O O H HHOO OH ON-OH O.N N.N OOH OOH OO OOH OO mwsoO O.OH 0.0 O.N ON ON O OH H HHHO OH OH-NH O.N H.N OOH OOH ON OOH OO muOoO 0.0 O.N O.N OH OH H O O HHHO NH HHaO .ssO 0.0 0.0 OOO OOO ONH OOH OOH HO nsOoO 0.0 N.O 0.0 ON OO O O OH O HHOO HH O .qseuON 0.0 0.0 OOO OHO ONH OOH OO ON nuOoO O.OH 0.0 H.O NO OO NH ON O N HHHO OH ON-NN 0.0 0.0 HNO HNO OOH NO HO OOH menoO O.OH N.O 0.0 NO HO OH OH O O HHOO O HN-OH O.O O.O OOO OOO OOH HHH OOH OOH essoO N.O 0.0 0.0 OO OO ON O OH OH HHHO O OH-O O.O N.O OOO ONO ONN NNH OOH OOH unsoO O.NH O.OH 0.0 OO OO HN OH ON ON HHOO O O-H 0.0 H.O OOO 0.0 OOH OOH NHH NON ousom O.HH 0.0 0.0 NO OO OH ON O NO HHHO O O0-0N O.O 0.0 OON OOO OOH OO OO OOH heuoO O.OH 0.0 0.0 ON OO OH O HH OH HHOO O ON-OH H.O 0.0 OOO OOO OOH OOH NHH OON nusoO O.HH 0.0 O.OH OO NO NN OH O OO HHHO O OH-OH 0.0H 0.0H NOO OOO ONN OON OON OON mesoO 0.0 O.OH N.HH OO OO OH OH ON OO HHHO O 0-0 .eoz O.HH O.HH HOO HOOH OON NOO OOH OON hence 0.0 O.HH O.NH NO NHH ON NO OH OO HHHO N N .eoz-ON 0.0H 0.0H OONH OOOH OOO OOO ONO OHO mnsom 0.0 0.0H 0.0H OO HNH OO OO ON OO HHOO H ON-ON .eoO OOOHuNOOH OO-NMOH OmuHmOH OOOH OOOH NOOH HOOH HOOH sonssz tnnphO .O.O OOH Hneoe Oo eeoouom Hoeoe HOoes Ooez nod HHHO ZOOHOOHO .OOOOOOO Om OOOOHOOOOOOO--OOOOOOO OOOH-HOOH OOOOOOO O24 OOOOOO OOHO OOHOOOO OO OOHOOOHOOOHO OOOOuO O: mqmde 119 decrease in yield as the season progressed will be considered later after the factors influencing hunting success have been discussed. D . Daily Over a four year period about 35 percent of the total effort occurred consistently on Sundays, 25 percent occurred on Saturday and weekdays each had about 8 percent. The distribution of the kill was almost identical to that of the effort expended. This indicates that the kill is almost precisely a function of the effort expended when based on season—long totals over a h year period. The rate of kill was nearly the same each day. 'Week ends, however, had a slightly lower yield of 9.7 rabbits per 100 gun—hours in contrast to 10.2 rabbits per 100 gun-hours on weekdays. Statistical analysis indicates that this great difference could be expected 50 percent of the time due to chance alone. Therefore, the difference in success between week ends and week- days is of doubtful significance. A summary of daily hunting statistics is given.in.Table hl. It was thought that the amount of effort expended on the weekends might vary during the season. This was checked and it was found that except for the first week the amount of effort expended on weekends consistently ran about 60 percent of the total. During the first week only h2 percent of the effort was expended on the weekend. This was no doubt due to the great hunting pressure applied opening day which was always on a weekday. 120 0.00 O OOH O OO O.OOH O OO OHOO ONON HHON OOON OHHN Hepoe H.ON O.NN O.ON O.ON O.ON OONN OOO OHO OOO OOO asosspsm H.O 0.0 0.0 H.O 0.0 OOO ONN OON OOH HOH OOOOOO 0.0 0.0 0.0 0.0 0.0 HOO ONN OHN NOH OOH OnenhsOO 0.0 0.0 0.0 O.O H.O ONO ONN OOH HOH ONH Oseooeeee. O.O 0.0 O.O O.O O.O OOO OOH ONN OOH OHH aneumue O.N N.O H.OH 0.0H 0.0 OOO OO ONH OON NNN Onoeoea N.OO O.OO N.ON 0.00 N.OO NOH ONO HOO HOO OOO Odessa mgom HdPOB MO osfimoumnm HwPOB HMQOB mhfiom MO LED—O52 «whomfim mo pcooaom 0.00 O.OOH O.OOH O.OOH H.OOH HOO OOH OON OOH OON HspoO O.O O.ON O.ON O.HN O.ON O.ON OHN OO OO OO OO Onenessm 0.0 O.O N.O N.O H.NH 0.0 OO OH NH OH HO Onoene N.HH 0.0 O.OH O.OH 0.0 0.0 OO OH ON OH ON enemusee 0.0 H.HH 0.0 0.0 0.0 0.0 OO NN O NH OH Onohoeooz. N.O 0.0 0.0 O.OH 0.0 H.O NO OH OH OH NH Onenosa 0.0H O.N 0.0 N.OH N.OH 0.0 OO O OH ON ON Onesoz 0.0 0.00 O.HO 0.0N O.HO 0.00 NOO OO OO OO HO aneesm .O.O OOtheoO zoom tom HHHO Hseoa mo eeoouom HHHO Hence HHHO Oowwnm tongsz_ uHHOO \HHHM mo pcooummp Hmpoe OOOH OOOH NOOH HOOH OOOH OOOH NOOH HOOH OsO hdmN hdmfi OOOHOOHO .OOOOOOO OOOOOOHOOOOOOO--OOOOOO OOO OOHO OOHOzOO OO OOHOOOHOOOHO OOHOO a: mqmde 121 E. Hourly The hourly pattern of hunting effort on weekdays, Saturdays, Sundays and opening day during the l95h season was determined by tabu- lating the number of hunters on the area at one-half hour intervals. Fluctuations in effort are shown in.Figure 25. A peak of hunting intensity occurred in all cases about 11:00 A.M. A lower second peak occurred about 3 P.M. on Saturdays and weekdays. On Sundays no after- noon peak was evident. Friley (195k) presented graphs showing changes during the day in.pheasant hunting pressure at the Rose Lake Wildlife Experiment Station near Lansing, Michigan. These hunting pressure fluctuations were similar to those observed in this study except that the afternoon peak was relatively higher and about one hour later. In 1951 the amount of effort expended between daylight and 10 AWM., 10 A.M. and 12:30 P.M., 12:30 PJ‘I. and 3 P.M., and 3 PM. to dark was tabulated (Table h2). This tended to obscure the peaks of effort but it affirmed that the greatest effort was in the morning. The rate of kill was greatest for the earliest period and decreased as the day progreSSed. This possibly indicated that early morning was the most successful time to hunt; however, the difference was not statistically significant because as great a difference could be expected to occur between.50 and 30 per cent of the time due to chance. To further test the possible superiority of early morning rabbit hunting the effort and kill that took place then during the 195k season was compared to that experienced later in the day. It was found that 600 hours were hunted and 59 rabbits bagged between daylight and /,V W! , II/ , (I f. !f I a /w/. ./ o. , O I h>(fid)h(w OOOOIO OOIOOO. OOOOOO OOOOOOO 28.3 OzEzOI OO 20:29:53 OOOOOI ON :3; O O O m ~ \ N. : on r O 'I"...Ilrll ' .. ’ D II'\ . 'IlLIIt D ' Ir " ’1 ' l l I D 14 h I ,,'l "' ullr b k P ' - b b 1,1IIO/ I.||".llu // 10: I //l/ \9! l \ I/‘I .14, [III \ \ t \ /.;.I I‘ll" lu+"+'ll‘|‘llill.|l'nll.ll‘1|'+lin"! I \ t. A / l - I‘lll r I‘l I‘...\ \ .u/ u ’ / I I ‘ l ‘I II /I., ,, /¢.t l l9..\ \ . A ,/H,, ,V O. \\ \ . I; f \ N ‘ ../ - 23:3) \\ \ A A; A o. d t A: 8 2 OHUV WHINOH JO uaflunu 39V83AV 123 TABLE )12 DISTRIBUTION OF HUNTING EFFORT, KILL AND SUCCESS 1951 SEASON-~KELLOGG FOREST AUGUSTA, MICHIGAN Effort Kill Kill per Time Hours Percent Rabbits Percent 100 Gun-Hours Light - 10 1.14. 60h.3 28.3 82 32.2 13.6 10 A.M. - 12:30 P.M. 671.7 31.5 8b 32.9 12.5 12:30 AM. — 3 P.M. h86.0 22.8 51 20.0 10.5 3 P.M. - Dark 325.5 15.3 33 12.9 10.1 Unknown 1:51; 2.1 5 2.0 10.9 2132.8 100.0 255 100.0 12.0 12h 10:30 for a success rate of 9.8 rabbits per 100 gun-hours. During the remainder of the day, 1,725 hours were hunted and lhO rabbits bagged for a yield of 8.1 rabbits per 100 gun-hours. When data from the 1951 and l95h seasons were combined, the early morning yield was 11.7 per 100 gun-hours and for the remainder of the day 9.6. This difference was almost statistically significant at the 5 percent level. It seems likely that a larger sample would have indicated that a significantly higher rate of kill occurred early in the day and lessened as the day progressed. IV. Distribution of Effort, Kill and Success Among the Hunters A. Effort The distribution of effort expended during the season by individual hunters ranged from four men who hunted only one-half an hour to two that each hunted over 78 hours. Most hunters Spent a relatively short time on the area with two hours being the modal effort and four hours and 26 minutes the mean. Table h3 shows the distribution of effort among the hunting public with the effort grouped into hourly intervals. Eighty-five percent of the hunters hunted less than six and one-half hours. Possibly a.more natural way to classify the distribution of effort is by the number of visits made during the season. Table hh shows this along with hunter success, average hunt length, and the extent to which dogs were used and hunting was done on days with snow cover. Sixty-five percent of the hunters visited the area.only once and the number that hunted increasing numbers of times tapered off rapidly. Only seven 125 TABLE 113 DISTRIBUTION OF HUNTING EFFORT AMONG THE HUNTERS 1958-55 SEASON-—KELLOGG FOREST AUGUSTA, MICHIGAN __.._ Length of Effort Number of Percent of (Heurs) Hunters Total Hunters .50 - .75 19 3.58 1.00 - 1.75 16b 30.9h 2.00 - 2.75 128 21.15 3.00 - 3.75 73 13.77 h.00 - 8.75 32 6.0b 5.00 - 5.75 28 ~ 5.28 6.00 — 6.75 16 3.02 7.00 - 7.75 7 1.32 8.00 - 8.7h 8 1.51 9.00 - 9.75 10 1.89 10.00 - 10.75 9 1.70 11.00 - 11.75 5 .9b 12.00 - 12.25 3 .57 13.50 - 13.75 3 .57 1h.25 - 111.50 h .75 15.25 - 15.75 2 .38 16.25 - 16.75 h .75 18.50 1 .19 19.00 1 .19 21.00 1 .19 22.25 1 .19 25.00 - 25.25 2 .38 27.50 l .19 28.50 1 .19 33.00 1 .19 3h.25 l .19 36.75 1 .19 118.50 1 .19 75.25 1 .19 78.25 2 .38 530 100.01 126 mN ooa m.ma 4H.m 0H N.mm H 2N ooa ooa N.mN cm.N NN N.Nw N NH om OOH m.ma mm.m 4N m.omH N :H o o e.mN om.N N m.NN N HN oo 05 O.N NN.N N N.NN H OH ooa ooa s.m No.4 N p.0m a m NH mm m.4 Nm.N a m.mm : w oN mm N.NH om.N NH w.mw m N ON 00 3.6 NN.N N N.Noa m 0 cm NN O.N N:.N ma m.0ma ma m 0N Nw m.® NN.N NH O.NMH mH : NH mo m.m NO.N NH N.NNH mm m NN am w.m ON.N mN m.NNs mm N R 66 o. N mm. H 2 meme in H pm>o0130cw on: mom madcmnqzo 00H you pHmH> pom HHHx musom mumpcsm mpamw> spa: mpfima> Hamopmm Hafiz new: mHSom :mmz Hence annoy honEsz hmpsdz namepom . zeoemo z .eempope Nmmmoavoooqqme--zomemm :mNH .Nmoomeeo NHNH>_mo«m mom mmsoo.3ozm NNN: QBHmH> mo mowezmuwmm qu mm:.ooa Nmmmoobm amabzmq BHmH> 24m: Mme Q24 Mozmbwmmm aHmH> maezmm mo ZOHBDmHmemHQ a: mqmummno OOHOOQXM Oo>pompo ompooaxm. Um>pmmno ompooaxm oobpomno oopommxm oo>pmmno HHHM whopcdm mo ponfisz._ manage: mo Ampszz mhmpcsm mo amnesz mampcsm mo honesz myopadm mo hmpEdz Hence muses mN.Oa-Om.mH enaom mN.NH-Om.O ensue mN.O-mN.N eneem OO.N-m. H 525% .eemOOOH ammuONHOOOHHmeunzomemw mmmthmOH umoznmo ON OOO OHNOONHH mm OHOO3 Name ON OHNHNEOO mmHmOOneHO NNONNH NOOHNH> 2H HHHM 2H ZOHNOmHmNmHO.Om>NmmmO w: mqmde 133 category for the four distributions was then added to yield an over— all distribution of the kill that would be expected if chance alone was responsible for the variability in kill. This approach takes into consideration variability due to differences in effort and assumes that there is a single probability of success. The resulting distribu— tion based on these conditions was then compared with that which was actually observed. The observed distribution had more hunters which killed no rabbits and which killed seven or more rabbits than would be expected where it is assumed that a uniform probability of success existed. Statistically, this difference was highly significant. Thus, there was less than one chance in 100 that the differences in success which were observed were due only to chance. When differences in effort are taken into consideration, the vari- ation in success due to chance accounted for more variation than might be expected from a more casual consideration of the matter. It could be expected that with a uniform rate of kill 7.1 per cent of the hunters would kill two or more rabbits and that their kill would make up 50.h percent of the total bag. It was actually Observed that 6.8 percent of the sportsmen killed two or more rabbits and that they succeeded in killing 65.8 percent of the total taken. If a uniform probability of kill existed it would be expected that 7h.3 percent of the hunters would kill no rabbits. It was observed that 80.h percent killed no rabbits. Next the expected Poisson distribution kill for each effort cate- gory was calculated based on each category's average rate of success. Thusiknu~different probabilities of success were used. These expected 13b distributions were then added as before and compared to the observed distribution. Again there were more hunters with no kill and a bag of 7 or more than would be expected due to chance. However, as would be expected, the difference was not as great as that observed previously. In fact it was not quite significant at the 0.05 level of probability. Thus it can be concluded that much of the variability in success observed in the first analysis was due to hunters in the different effort cate- gories having different probabilities of success. This was probably due largely to the greater success experienced by those persons that hunted most often. I When the kill distributions in the individual effort categories were each compared with the Poisson distributions based on their indie vidual probabilities of success, only the greatest effort category dis— played a statistically significant departure from a uniform probability of success. The wide range of effort of from 13.5 to 78.25 hours in this category possibly contributed to this. The other effort categories extended over a much shorter period of effort. It was necessary for the greatest effort category to include a long period because so few hunters hunted more than 13 hours. Because of the small number of hunters involved it was not feasible to further subdivide the groups and make an analysis like that above. However, it was possible to divide the category into four parts each with seven hunters and test statistically the homogeneity of hunting success. The significant chi-square value which resulted indicated that hunting success within this group varied more than could be attributed to chance alone. 135 This analysis considered the variation that existed in effort expended so that that factor could not have been reSponsible for the differences noted. This investigation into the homogeneity of hunting success may be summarized as follows: 1. Considerable variation in the distribution of the kill can be attributed to chance and to differences in amount of effort. 2. Observed variation in hunting success was significantly greater than could be attributed to Chance when differences in effort were considered thus implying that real differences in hunter effectiveness existed. 3. The non-uniform rate of kill was due largely to the relatively few very successful hunters that hunted on the Kellogg Forest more than 13 hours. VI. Effect of hbather on Hunting Effort Tables h9—55 summarize the average daily effort expended under various climatic situations. The various climatic factors were con- sidered separately for each month and the daily effort weighted so as to eliminate the influence of day of the week on hunting pressure. During Octobers there was a longer hunting effort on clear days than on partly cloudy or cloudy days (Table h9). During the remainder of the season no trend was evident; however, there were only 8 clear days during the 153 days considered. The daily minimum temperature (Table 50) had no noticeable influ- ence on daily hunting effort. Perhaps this is because minimum 136 H.N N ON N.NNH O.NHH HN NOOOHO H.NH O m N.NN N.HN H NOOOHO NHenmN H.NH N H N.NN N.NN H teeHO antenna N.H H NH H.OHH O.NHm NH NOOOHO H.HH N OH N.NHH H.NHH ON NeeeHe NHenHN O.HH n NH H.NH N.NN H neeHO tenseeeO H.HH mH NO H.Nmm H.NHN NH NHHOHO H.NH HH NH N.NOH 0.0NO NN NHHOHe NHeeeN H.HH HN N O.HN O.HN H neeHO menseeoz H.HH NH NH H.NOH H.OOH N NOsOHO H.N HN NH N.NHH O.NNN HH NOOOHO NHeneN O.N NH NH O.HHN H.HOH O nemHO teneeeO .:.w 00H Ham mom HHHm manom undo: when mmocHOSOHo spec: \HHHH meson OeeHmHms Heeear tenssz Essa: . HNmOOOH NmmeON OOOHHNH--ONZHN2OO HNHO NNNH OHH HmNH OONOOOm OZH NNONNH OzHNzOH Oze NNHZHOOOHO Hamsemm mHmmZOHNHHme m4 mqmae .H. ...rwbm. mo ENEHHozmmHv wmarmwz ZHZESH ELEM egmmweecwmm H26 EHzaHzo gnome ....zc mcoowmm HOmH-HOmO ONH» OOszzNO--HNHHOOO.HOmNmH NOOOmHN. zHOHHONz 2 H558: 256 8. a 3.3. So» mrwma modem 5PH\ 305d: Hoawmswego Owned modem mega WHHH pod uww H00 m. :. Ooeoeee HH-Hm H ON.N ON.N O NH O HNumO N ONN.N ON.N HN m: 0.0 NH-NN N NHH.m NN0.0 NH NH N.H NN-OO N NH.N Nu.O N Hm N.O NH-Nm N NON.N Hmm.H NH NN Hw.r NONE H 56.3: Nmmb NW NH H.O NH-Nm HH NNN.N NHO.H HO mm N.H NN-mO H N.H N.N O N O zooz O N O N.O N.O N O N.HH OO NN O.NNH N.NNH N m N.OH NH OH N.ONH N.ONH N O N.N mN mN O.NNN N.ONN N N H.NH ON ON O.NNH N.OOO N N O.N NN NO O.NNO 0.0NO HH H N.N NO NH O.NN O.NNH N O NonopoO .O.N OOH NNN NNN HHHN mNmNO mnsom ONNO .N.m.z \HHHN musom ONOONHNO HNOON popesz NOHooHNN NON: EOHNOH: .NNNOOON “Emma OOOHHNN OOZHNNOO «Nam NNNHuHmNH mmmOOOm O24 NNONNN ozHNZON OZN NNHOOHO> O2H3.zOOz ZNNKNNN NHNNZOHNNNNN mm M‘Hmfi; lhO .. O O O O H NN. -ON. O N.N O N.N N.N H NN. -ON. O O.N O O.N O.N H NH. -OH. O.OH O.N O m.NN N.OO N NO.-O0NNN N.N N.N NN .NHm N.NNN Nm macs NNNOONN m.mH N.mN O N.mN N.NN H NN. uON. N.NH N.OH N N.OH O.NO H NO. -OO. O O.N O O.N O.N H NN. -ON. N.O N.HH O N.NN N.HO N NH. -OH. N.NH N.O OH O.NN O.NNH NH NO.-OoNNN N.OH N.OH ONH O.NHN N.ONNH mN Neon NOOENOOO O.N N.O N O.NH N.HN N NN. -ON. O.HH N.H H N.H N.N H NN. -Om. O.N O.N N O.OH 0.00 H NO. -OO. O.NH O.NN O O.OO 0.00 N NN. -ON. O N.O O O.NH O.NH N NH. -OH. N.N O.HH NN N.OOH N.HON N NO.:oomNN O.NH O.NN ONN N.ONOH O.NNNH NN Ono: NOOEO>oz O N.H O N.H N.H H Opoe No N.H O N.N O N.N N.N H NO.H-O.H N.NH N.O O N.NN N.O H NN. ..2. O O.ON O O.ON O.ON H NO. -OO. O.NN N.NN HH N.NN N.NN H NH. -OH. H.HH N.NN NN N.ONH O.NNN m NO.-OOOON N.N H.NN OHH H.NmN N.NmOH NN one: NOOOOoO .2 .m OON N.NNH .Hmm NN.NN 9802 mbHom mbmm 53m mmnocH Hficoz NON HHHN NEON OSONHOO H3. O.N $ng EOHNOHO ..ZNOOOO NNNNON OOOOHNN--ONzHNzOO ONNO NNNH-HmNH NNOOOON O24 NNONNO OZHNZON OZO ZHNN ON ZOHNONHNHOONN NNNENNN NHNNZOHNOONN Nm NONON lbl i 1 1 O O O O O N N.O-O.O O O O O O H .N.N-O.N N.N N.N HH N.NO . N.NNH N N.H-O.H N.N O.N H O.NH O.NH m m. -H. O.NH N.N ON O.NOH N.NNN OH OONNN N.N N.O NN N.NNN N.OHO HO Ocoz _NNNOONN O O.H O O.H O.H H N.N-O.N O.NN N.OH m N.OO N.OO N N.N-O.m O.NH N.H H N.N O.N N N.O-O.O 0.0H N.OH N N.HN N.HN N N.N-O.N N.N H.NH m O.NO N.HN N N.N-O.N N.O O.O H 0.0N 0.0N O N.H-O.H N.N N.N O N.NO N.ON m N. -H. N.OH N.N NN H.ON O.NNH HH ONNNN N.N N.N OOH N.NON N.NNOH NN Ozoz pmnsmomm H.OH O.OH NH N.NN N.NHH N N.0-0.0 N.O 0.0N N N.HO N.HO N N.0-0.m O.NN O.NH N O.NH O.NH H N.N-O.N N.NN N.OH O N.ON N.ON N N.H-O.H N.NH N.NH ON O.HN N.NOH O m. -H. N.OH N.NH NH N.NN O.NNH O Oomge O.HH 0.0H NON N.NNHH O.NHNH NN Ocoz Nopfim>oz .O.N OOH NOO NON HHHN NNOON NNOON ONON. HHNNzonm Opcoz NON HHHN mpsom OOOONHON HNOON umnesz . mmOoOH zaOHNOHz .ONNOOOO .NNNNON1OOOHHNN ONZHNNOO «NOO NNNH-HNNH NNNOOOO OON NNONNN OOHNNON OOO OONNNOON ONNONNN NHNHZOHNNHNN Om OHNNN 1&2 N.HH N.O N.NH N.O ON O.N O.N NH =H N.N N.NH O.N 0.0N N N.OH N.N NH [H No OONOOONON 0.0 O.NH O O.HH N. 0.0 O.NN N 0:02 O.NH N.N O.NH H.N NH N.N N.NH O N.HN N.H HN =H case who: H.OH N.N N.N O.N N N.OH H.N NH O =H No Ompmppmom N.N H.NH O.HH N.N NH N.N 0.0 OH N.N N.HN OH Ocoz N.OH N.OH 0.0H N.HN N H.N N.N O N.OH H.NH N =H N.HH O.OH O N.OH H -- O H H.NH H.NH N =H No OONOOONOO N.OH N.NH O.N O.HH ON N.N N.NH NN 0.0H N.OH OH ocoz hgmscmw pmpfimoma nonEm>oz .n.m OOH N.OO .Hmm mama 9960 39% 9mg HHHx mhdom .oz NNNH-HNNH .£.w OOH ham pom mhmm Amm HHHM manom .oz NNNHuNNNH .z.m OOH hum 9mg mama NOO HHHN NOOON .oz ONNH-NNNH .:.m OOH NOO pmm mom HHHx mndom 230W; I! ll“ 1 I! 1111“ ll 1‘11 ZOOHNOHO .ONOOOOO NmNNONrOOOHHON--NzONOON NNNH-HNNH NNNOOON OZO NNONNN OOHNzON OON NNNOOO3ONO ZONONNN NHONZOHNHHNN NN OHNNN lh3 temperatures occur mostly at night. ‘When the daily maximum temperature was considered (Table 51) a greater effort was evident in Novembers on cool days. During Decembers, warm days tended to receive the most effort. No trend was evident during Octdbers or Januarys. During every month calm days tended to be hunted more than windy days (Table 52) . Except for January and December when the sample size was small a marked reduction in effort was associated with rain (Table 53). On the average effort was reduced by about one-half on days when rain fell. As would be expected on days with a heavy rain the reduction was even greater. Precipitation as snow had no effect on effort (Table Sb). In fact, in December a slightly greater effort was expended on days with snowfall. The association between snow on the ground and hunting effort (Table 55) was surprising. Days during December and January with snow cover had on the average less effort than did days with no snow cover. During January there were only five days without snow on the ground. During November there was a tendency toward greater effort on days having snow cover. VI. Factors.Affecting Hunting Success A. Previous Ehperience With Area Four hundred ninety-eight of the registered hunters in 19Sh indi- cated whether they had hunted the area before. The 2h? hunters with 1M4 previous experience averaged 10.5 rabbits per 100 gun-hours while the 251 who had not hunted the area before averaged only h.3 cottontails per 100 gun-hours. A chi-square test indicated that there was less than one chance in 200 of the greater success of the experienced hunters having been due to chance. A greater use of dogs by the experienced group may have had an influence on their success. The experienced group was assisted by dogs on 69 percent of their visits while the inexperienced group were aided by canines only 62 percent of the time. This difference was greater than the percentages indicate because experienced hunters averaged 2.8 hours per visit while the inexperienced group hunted only 1.7 hours per visit. No difference existed in the amount of hunting done with snow cover. B. Hunter Gecupation PeOple hunting during the l9Sh-l9SS season were classified into the eight occupation categories shown in Table S6. Mere than twice as successful than any other category were the 13 unemployed persons. This group averaged 23.5 rabbits per 100 gun hours. On the average this group neither used dogs any more nor hunted more on days with snow cover than other groups. However, 69.3 percent had previous experience with the area compared to the h9.7 percent of the entire group. The next most successful group was supervisors and foremen closely followed by farmers, unskilled labor and skilled labor. The skilled and unskilled laborers made only 13 percent and 8 percent of their visits on days with snow cover so their success can not be attributed to this. its O.H NO. OO.N OO ON N.H H OO.ON NN ON ONOOOO N.H NO. N.H NN HN N.NN N NN.NN ON NH ONNOHOEOOO O.N NN. O.N NN NN O.HH N NN.NO NH N ONEONON was maomenmasm N.H NH. N.H NN NN N.OH N NN.NN OH N NONENNN N.H NO. N.H NN NN N.N OH NN.OON ONH NN NOONOOON N.H OH. H.N NN NO N.O NH NN.NON ONH NN OOHONNNONO pad mwochsm H.N NN. O.N N HO N.OH NO NN.NOO ONN NNH NOONH OOHHHONOO H. N HN. N. N NH NN N.N ON NN.NNN OOO NNN .8an OOHHEm hmpcsm pHmH> meH> hwpoo_3ocm pcmopmm .:.m OOH HHHM mnsom mpHmH> myopad: GOHpmmsooO NOO NOO NOO OOHs mmeH> NOO NoO \HHHN HNOON No No NO NOOOON mmeH> mpwnndm mhsom ammonmm hmnsdz nonssz nonfisz O.OHNOHO .NNNOOOO. NNNNON OOOHHON--ONNH .NNNOOON OOO ZOHNONOOOO NNNNON om mqm¢a 1&6 Their dog use was a fairly high 75 percent and 61 percent reSpectively. Only seven men were in the supervisors and foremen group and there were only nine farmers, therefore, the representativeness of these categories is open to some question. The least successful groups were business and professional men, students, and the "others" category in order of descending success. Contributing to the very low success (1.2 rabbits per 100 gun-hours) of the "other" group were six women that contributed one-fifth the group's effort but bagged nothing. The student group included some grade school children which no doubt influenced the success in that category. No data are available to explain the low success of the business and professional group. C. Dog Use In 1951 and l95h when dog use was studied dog users were roughly twice as successful as non-dog users. In 1951 dog users shot 13.2 rabbits per 100 gun-hours while non-dog users averaged.only’8.7. In l95b the hunters using dogs killed 10.5 rabbits per 100 gun-hours in contrast to only h.3 non-dog users. In both years the superiority of the dog-user's rate of kill was statistically significant. Kinds of dogs used was recorded after each hunt in l95h by the following classification: 1. beagle -- dogs believed to be or strongly resembling a pure- blooded beagle. 2. "hounds" -- all other hounds such as blue tick, black and tan, red bones and fox hounds. Ill: [Iii [III lb? 3. bird dogs -- included setters, spaniels, retrievers and other breeds usually thought of as being best suited for bird hunting. h. "mixed" -- dogs other than hounds that resembled no recognized breed. 5. "others" -- recognized breeds not mentioned above such as boxers, poodles, etc. When more than one dog was used the appropriate code numbers were recorded. a total of 27 different combinations of kinds and numbers of dogs were used. The ten combinations hunted ho Or more hours are shown in Table S7. Hunters using beagles were by far the most success- ful, averaging 13.7 rabbits per 100 gun-hours. The use of two or three beagles did not increase success. Bird dog users had the lowest success of 5.1 rabbits per 100 gun-hours which was only slightly higher than the h.2 kill made by non-dog users. The probability that the superiority of beagles over bird dogs was due to chance was less than 0.005. In 1951 the kind and amount of dog use was not recorded after each hunt according to a code system as was the case in l95h. This resulted in a less precise record of dog use but the results are summarized in Table 58. The 1951 data indicates that the I'mixed" dogs were associated with the highest success. As in 19Sh the bird dog users had the lowest rate of kill except for those that used no dogs. It can be noted that in 1951, mixed dogs were used much more than in l9Sh. It is suspected that the different methods of recording the data may have been responsible for this. 1118 N.NO O.NN O.NN N.HN .moe OOOOOHs pnommm pcoonmm O.OOH N.N NNH OO.NNNN N ON.NN NN O0.0NN NO O0.0NN NN OH.OHN mHNOON O.N N.N NH NN.NNH N OO.N OH NN.NNH O NN.NH N NN.OO N.OOOO HHN N.H O.N N ON.OO H ON.OH N OO.NH O OO.NH Omst one use memmn NEO N.N N.N O NN.NO O O0.0 N OO.NN O NN.NH H ON.ON OOst ...:O N.N N.HH N NN.NO H NN.NN N OO.NN N OO.NH senses. Ono paw mHmmmO. go 0.0 O.N N ON.NN O OO.NN H NN.NN O NN.OH =OssoO= OOO N.H O.N H ON.OO O OO.NH ON.NH H OO.HH NNOO N.HO 9:9 0.0 H.m OH mm.mNH m OO.MH m mm.MO m mméw m mN.Hm mop p.30. 95 N.N N.OH OH OO.NNH N OO.NO N ON.NN N ON.ON NOHNOOO OONON O.HH H.NH NN OO.NO HH NN.NOH N NN.NN NH ON.NN NOHNNOO baa O.NN N.NH NN .NN.ONN OH ON.ONH HN NN.OHH NN O.ONN OHNNOO 0sO 0.0N N.O ON NN.NON o OO.NN OH OH.NON N ON.ONH HH NN.NNN NOO Oz ..NNNNN ..ONN. 3...... ...... a... 3...... ..N......N as... ZOOHNOHN .NNNOOOH NOONON OOOHHNN--ONNH .NNNOOON OZHNZON 2O NNN OOO NO NONNNN NN ”ENS. 11.9 ' THEESB DOG USE AND hUNTER success, 1951-431.1000 FOREST AUGUSTA, MICHIGAN Kill per Percent Type of Dog Gun-Hours Kill 100 Gun-Hours Total Effort None 828.9 72 8.7 38.7 Beagle 678.6 99 lh.6 31.8 "Hound' 182.1 2b 13.2 8.5 Iixed 185.7 28 15.0 8.7 Hounds and bird dog 117.8 15 12.7 5.5 Bird dog 11.3.8 17 11.8 6.7 Totals 2132.9 255 12.0 150 In both years the use of any kind of dog apparently increased hunting success. Also, during both years the rates of kill in the various dog type categories were not homogeneous when tested statis- tically. For 195h the amount of dog use also was determined for weekdays, Saturdays, Sundays and opening day. In each classification except opening day, about 30 per cent of the hunters' effort was without dogs. 0n opening day, us percent of the hunting parties did not use dogs. The overall average rate of dog use was 69.6 percent in l95h and 61.3 percent in 1951. D. Party Size Data from the 1951 and l95b seasons were examined to determine if any difference in success was associated with different size hunting parties. Hunters hunting alone had the best success of 12.8 rabbits per 100 gun-hours. The probability of their superior success being due to chance was less than 0.02 when compared with all other party sizes. Party sizes of four and five were next with yields of 11.5 and 10.1 reSpectively. Only 188 hours were hunted by'parties with five hunters, therefore, the results may not be representative. Party size three had the next to the poorest success of 9.3 rabbits per 100 gun-hours. The poorest success was experienced when two hunted together with an average yield of 9.0 rabbits per 100 gun-hours. The better or poorer success rate of all party sizes except party size one were not statis- tically'significant when a single group was compared to the rest. 151 The prObabilities that the better success rate of individual hunters over party size two and three was due to chance, however, was less than 0.05 and 0.01 respectively. 'Yet the superiority of party size two over party size four was not significant. It can be concluded that individual hunters are the most successful while the differences in the success of other party sizes may not be significant. The reason an individual hunter experiences the highest rate of success is unknown. Possibly an individual hunter has a higher rate of kill than two or more hunters because the number of rabbits flushed per hour by a party is not directly'proportional to the number hunting while the accumulation of gun hours of effort is proportional to the party size. E. Climatic Factors This analysis, based on data collected during the 1951, 1952 and , 1953 seasons was made on a monthly basis since it was thought that some factors might act differently at one time than at another. Also by breaking the season into shorter intervals the potentially compli- cating effects of changes during the season in rabbit abundance, density of cover, hunting effort, etc. were minimized. Tables h9-55 summarize the observations. The following factors were observed to have no noticeable influence on hunting success during any month of the season: minimum temperature, wind velocity, and precipitation as either rain or snow. During Octobers there was a tendency toward greater success on days having cool maximum temperatures. In Novembers the same tendency was 152 present though much weaker. In Decembers warm days were associated with the greatest success. Hunters were twice as successful on the 15 warmest days than they were during the 25 coldest days. In Januarys only a very slight tendency toward better success on warm days was shown. The presence of snow cover in most instances was associated with an increase in hunting success. Based on December hunting statistics during the 1951 through l95h seasons, on days with one inch or more of snow cover the average yield was lh.l rabbits per 100 gun-hours in contrast to only 9.0 for the days in December not having snow cover. This difference is statistically highly significant. The superiority of hunting success on days having snow cover was much greater in 1951 and l95h than in 1952 and 1953. In l95h, which is not shown in the table, the rate of kill on days without snow cover was 6.9 rabbits per lOO-gun-hours in contrast to 10.2 for days having snow cover. VII. Analysis of Factors Affecting weekly Hunting Success A factor that tends to limit the kill of some species, particularly pheasants (Shick 1952), is the decrease in yield per unit effort that occurs as the season progressed. As the yield decreases hunting effort becomes much reduced. is this did not occur on the Kellogg Forest it seems likely that the vulnerability of the rabbits and/or the efficiency with which they were hunted.must have increased as the season pro- gressed. The previous discussion has established that the use of dogs, presence of snow cover and hunter "experience“ all influenced success. 153 A possible explanation for the observed trend in rate of kill would be that these factors fluctuated during the season in such a way that they counteracted the increased scarcity of the population. Data to examine this possibility are available for the l95h season. Figure 26 shows the fluctuations that occurred in the relative amounts of effort expended without dogs, with dogs and by hunters that visited the area eleven or more times. Effort eXpended with and without snow cover is indicated for each of the three success categories. It was previously pointed out that hunters visiting the area eleven times or more were about twice as successful as the average hunters. Therefore, effort expended by this group was used to indicate effort by experienced hunters. Only one frequent visitor did not use a dog, therefore, this category may also be regarded as being aided by dogs. Also indicated on Figure 26 are the weekly fluctuations in the kill per 100 gun-hours and the instantaneous rate of kill per 100 gun-hours. Both rates of kill are indicated so that a rate of kill not influenced by the population level (instantaneous) can be compared with the one (kill per 100 gun-hours) that would be expected to be influenced by population level changes. All other things equal, the instantaneous rate of kill would be expected to remain constant during the entire season while the kill per 100 gun-hours would decrease due to the decreasing population. There was a tendency for this actually observed in the data. Figure 26 shows that the instantaneous rate of kill early in the season was relatively lower than the kill per 100 gun-hour while 1% .oz Xmas) .m\ lamina] A o 3 I no .1 S (.9 W 1 MN 8 3 0 IA Q I. La: A V 3 a a n 3 0 , v .. om 3 9 )0 2n )0 2m 02 fl m. \/ n can: woo *3 S u m is w J I II {I m / f8 M Q 50.; 02 I. LT one: son --..-vfioa W at?) 0.N ammgltl RZMDOWQk . out. too M S IL AW 6.4 L X 1:... N s new 1 _ 3 i mmsr \; Ni 0 S \\ 8 _ 9 I: \ n a o.— ‘7 I I.‘ VQ m N. Y m .3 8. «we j; 3.“ szEzfiwzlllullv a o H I; O 3 9 n no 0 to: an a" S .3 N y un . o. H a” mo. .v O 1 0 no 3 ¢O.P r12 3 8. 5a sill? e... r. rm! .amwmoa eschews .wqucsm azazs: 3353:: 23:: sz< 3.3:; 3:53: 2. 3.2.2333: 2 3.3: 155 in the middle of the season the instantaneous rate of kill did not taper off quite as much as the kill per 100 gun-hours. The preseason population estimate based on the tagged—untagged ratio in the hunting kill was used to calculate the instantaneous rates of kill. The population decreases at several times during the season were estimated as the preseason population minus the hunting kill. Actually the population was reduced to a greater extent than this due to the unrecorded crippling loss and natural mortality. In an area where over 50 percent of the preseason population was reported bagged by hunters other mortality would be expected to be relatively low and recently obtained information indicates that the crippling loss is not great enough to markedly change the values. Any discrepancy from this cause, however, would tend to force the calculated instantaneous rate of kill to be lower than they actually were. This may have been a contributing factor to the low instantaneous rate of kill noted during the last five weeks of the season. However, the general increase in instantaneous rate of kill during the first ten weeks indicates that this error was not important. The last four weeks of the season were grouped into two two-week periods so that the rates of kill would be based on larger samples. It should be pointed out that the weekly fluctuations in the kill per 100 gun-hours during the 195h season were somewhat atypical in that the highest success occurred during the fourth instead of seventh week and hunting success late in the season was relatively lower than 156 was noted in other years. Therefore, the fluctuations in the rate of kill during the l95h season did not depart from what would be expected if the kill was determined by the population level as much as was the case during other seasons. It is interesting to note that fluctuations in both rates of kill agree with each other closely. This indicates that the weekly magnitude of both rates are probably influenced by the same factors. Therefore, if the cause of the fluctuations in one of the rates of kill can be determined, then the same factors will be expected to also influence the other rate . The weekky fluctuations in the over all instantaneous rate of kill per 100 gun-hours were much greater than cbuld be attributed to chance. The probability of fluctuations as great as these being due to chance was only between 0.005 and 0.001 as indicated by a chi—square test. Thus, it is evident that. the instantaneous rate of kill was not constant during the entire season. To indicate if variations in the amount of effort expended by the various success categories were responsible for the weekly variations in success, the relative amount of effort expended each week by hunters in each category is indicated in Figure 26 directly below the corres- ponding rates of kill. InSpection of these data strongly suggests that the fluctuations in the rates of kill were due to the fluctuations in the relative amounts of effort expended in the different success cate- gories. For example, the marked increase in the rates of kill noted during the fourth and tenth weeks were accompanied by increases in dog 15? use and in effort by experienced hunters. In ten of the 13 time intervals increases or decreases in dog use were paralleled by the expected increases or decreases in success. In general the greater amount of effort with snow late in the season may have contributed to holding up the rate of kill per 100 gun-hours. It can be noted that the sharp increase in the rates of kill during the eighth week was accompanied by a marked increase in snow cover. The least effort by experienced hunters occurred early and late in the season. This may have contributed to the low rates of kill experienced at those times. Although these relationships are evident from Figure 26, the situation in some weeks is somewhat confused because the variations in effort by the various success categories fluctuated in opposite directions with reSpect to their expected influence on success. To obtain a clearer picture of these relationships the total weekly effort was divided into the six success categories already described. The instantaneous rate of mortality per 100 gun-hours was then calcu- lated for each category each week (Table 59). AS previously mentioned weeks 12 and 13, and Lt. and 15 were combined because the effort expended then was quite low. The weekly instantaneous rate of kill per 100 gun- hours are shown in Table 58, along with the mean rate for each category. 'When no effort was expended during a week in a particular category the rate of kill is indicated by a dash. To avoid being misled by values based on very small samples an asterisk has been placed beside all values ‘based on less than 15 hours effort. in example of how a small sample can be misleading is seen in the eighth week when a very high .1432. 11.21.?! «a . 158 .masoennsm ma cusp need no women oeam>a _mao. meo. ego. emo. moo. sac. see: N00. *000. In: 4H0. $000. 000. In: mfiuqa mmo. --- sooo. ago. Nmo. --- sooo. MH-NH eHo. ooo. sooo. Hmo. ooo. sooo. sooo. HH Nmo. sod. mmo. moo. moo. aooo. --- oH mmo. woo. --- emo. . --- ooo. ..- e ooo. omo. aooo. ego. sHHH. omo. aooo. n oso. --- omo. woo. moo. aooo. mmo. a meo. --- saga. oeo. omo. --- mmo. o omoa --- NHH. --- «Ho. --- mao. m omo. --- ooo. --- emo. --- mmo. e owe. --- omo. --- moo. --- mao. m wmo. --- sepo. Nflo. emo. sooo. wHo. m mac. nun ~50. an: :No. nu- goo. H .mpHHmuLoz mo roam 30cm. oz 30cm 30cm 02 30cm 302m oz nonssz meow .pmGH pom: won mpamfl> cmboam mpfimfl> co>oam Moms smoz Hopoa .mpfimfl>.cop swap who: can» mama anon: mom seep mmoH .moo oz 2...on on . Snags ammeoa_oooqqexuuemea .muHmOoaeso mmmooom mDOHma> 2H mmoomuzoo ooH mam que ao mmeam mooezaezeemzH mm mqmdaw 159 instantaneous rate of kill of 0.111 occurred in the dog-used, no snow, low visit category because one rabbit was killed during the four hours and 15 minutes hunted. Also, in twelve instances zero instantaneous rates of mortality occurred when less than fifteen hours were hunted. The seasonal mean values for each category are weighted means and therefore are not influenced by the small samples that occurred during some of the weeks. The weekly instantaneous rate of kill per 100 gun-hours within each success category fluctuated less during the season than did the total weekly rates. This probably occurred because the factors which. influence success within each success category were more nearly constant than were those influencing the total weekly rate of kill. ‘When the uniformity of the instantaneous rate of kill within each success category was tested statistically, none differed significantly from a uniform rate of mortality for the entire season. After the mean instantaneous rate of kill per hour for each category had been determined it was possible to estimate the kill that would be expected each week in each category by multiplying the mean instantaneous rate of kill per hour for the category by the hours hunted. The total kill expected for each week should equal the sum of the expected kills for each category. In this way an.expected kill was calculated that was based on the relative amounts of effort and different instantaneous rates of kill for the various success categories. This approach assumes that a uniform instantaneous rate of kill takes place within each category during the entire season. The weekly total, however, would not reflect a uniform 160 instantaneous rate of kill because the various weeks are made up of different.proportions of the various success categories which have different instantaneous rates of kill. When this calculated kill distribution (Table 59, column 2) was compared to that observed (Table 60, column 1) there was no statistically significant difference. If the success categories had been broken into more of the factors known to influence success, such as type of dog used, whether the hunter had hunted the area during a previous season and hunter occupation, the agreement probably would have been even closer. ‘When a kill distribution based on a uniform rate of kill (Table 60, column 3) was calculated, it differed from the observed distribution to a statistically highly significant extent. From this analysis the following points can be summariZed: 1. weekly fluctuations in the instantaneous rate of kill per 100 gun-hours varied significantly from a uniform rate of kill. 2. The weekly fluctuations in the rate of kill were caused largely by variations in the relative amount of effort expended each week by non-dog users, dog users and experience hunters, each with and without snow. 3. The weekly instantaneous rates of kill per 100 gun-hours experienced by groups constant with respect to dog use, effort with snow cover and experience did not vary significantly from a uniform instantaneous rate of kill per 100 gun-hours over the entire season. TABLE 60 161 COMPnRISON OF THE OBSERVED WEEKLY DISTRIBUTION OF KILL (1) WITH THE DISTRIBUTION EXPECTED IF THE KILL DEPENUED ON THE RELATIVE.&MOUNT OF EFFORT BY VARIOUS SUCCESS CATEGORIES EACH OF WHICH CAUSBD A DIFFERENT BUT UNIFORM INSTANTaNEOUS RJTE OF KILL (2) AND THE DISTRIBUTION IF ‘THE KILL WAS THE PRODUCT OF A UNIFORM INSTANTANEOUS RATE OF KILL DURING THE ENTIRE ssxsow (3). 19Sh-l955 ssasow--KELLOGG-FOREST, AUGUSTA, MICHIGAN (175‘ (27 I3) Expected Kill Expected Kill Observed Compound Uniform 'Week Kill Rate of Kill Rate of Kill 1 30 39.0 b9.7 2 27 25.1 30.9 3 18 18.8 22.0 h 22 13.9 12.8 5 10 9.9 8.9 6 17 12.0 12.2 21 20.7 17.0 8 20 16.3 12.6 9 1h 12.9 10.0 10 12 11.3 7.5 11 3 9.6 7.1 12-13 h h.6 3.9 lh-lS 1 h.9 h.b 162 A comparison of the mean instantaneous rates of kill per 100 gun- hours for the different categories reflects the influence on hunting success of the various factors considered (Table 61). These effects have been examined before, but this approach portrays the effects of the various factors somewhat more clearly because the influence of each factor can be indicated under constant conditions with reapect to some other factor. Table 61 indicates the instantaneous rates of kill per 100 gun- hours that occurred under various situations with respect to hunter experience, dog use and snow cover. In calculating these values only data from those weeks where at least some effort occurred in both of the situations being compared were used. This was done to reduce the complicating effect of seasonal influences and unrecognized factors. In addition to comparing the factors in question under the various situations considered, the weighted and unweighted mean values are also given. A question exists as to which mean more truly reflects the mean effect of the factor being considered. From the point of view of the impact that a factor can be expected to have on the rabbit population, the weighted mean is probably the more appropriate value, since it considers the greater effort expended in some categories than others. However, from the hunter's viewpoint the unweighted mean is probably more meaningful. The hunter would only be concerned about the change in the rate of kill that he could expect if he were in a A different situation. 163 :1, :.mm: m.mm4 m.bu: ©.mom omocaoCH psooaom Ill smmo. ommo. mono. ammo. oeeoaoeH mwoo. omoo. momo. meeo. some men es: moo . messes oH mafia moo oz omao. emeo. Neao. aoHo. enemas Ha aoeoseona paee> 302m oz 20cm :.Hea 4.5mfi e.mw e.mom monoposH pcooaom mwao. ammo. mmao. ammo. oesonoeH Hamo. memo. aomo. omeo. some moo emeo. peso. Neao. Neoo. ocoz on: moo 30cm 03 Roam eeeoee as case smog e.oea e.mHH m.meH e.mo omooaoeH pcooaom ammo. mane. meeo. pmmo. omooaecH emao. soeo. asao. opeo. messes oH oomo. memo, eemo. emeo. enemas HH assessoeo passe 30cm oz 30cm oomD.on m.mw m.sea N.moe o.Nm m.em ommoaoeH psooaom eomo. memo. mono. oeao. meoo. oesonosH Elle. ego. memo. eoeo. ammo. aoem Hemo. peso. msHo. memo. peso. noes oz coeoo zoom coma coo: mom mom mop oz scepmswam oocfisoxm eoesnees neaee> oH neaoee He sees smog meson sesame cogs who: 2..onon rages... emseoanooongme--emea .que.ozHez:m co mesa moomzeezeemZH use zo mmoeoae.moonaa ao mozeoqazH Hp mamaa 16h Table 61 indicates that the presence of snow cover was associated with an average increase in success of 86 percent or lhh percent depending upon whether weighted or unweighted means are compared. When examining the effects of snow it is interesting to note that among those hunters that visited the area less than eleven times, the non-dog users success was increased 27 percent while the dog users experienced an increase in rate of kill of 52 percent. The group that hunted more than ten times and used dogs showed a h63 percent increase in instan-v taneous rate of kill associated with snow cover. This indicates that the experienced hunters benefited more from the addition of snow cover than did the less experienced hunters. However, this indicated increase in success is probably considerably exaggerated because the success in the no-snow'cover sample was based on only 38.5 hours effort and seemed low. Hunters in the eleven or more-visit category averaged 119 percent higher success than those in the ten or less visit group judging from the unweighted means. In'both cases dogs were used. The superiority of experienced hunters was greatest when no snow was on the ground. Those hunters using dogs averaged 197 percent greater success than did hunters without dogs. When snow cover was present, dog users superiority over non-dog users was 367 percent. This value and conse- quently also the mean is believed to be too high because the success of the less-than-eleven visit, no-dog, snow category of 0.0092 is probably too low because of the small sample of 153 hours on which 165 the rate of kill was based. Evidence to support this view comes from the instantaneous rate of kill in that category when snow cover was not present. This rate was 0.01h2 and it would be expected that the rate of kill with snow would be that great and probably greater. Despite this difficulty, it is probably safe to conclude that the use of dogs is even more beneficial with snow cover than without it. 'When the effect of both visit-frequency and dog use are examined, the hunters that visited the area eleven or more times and which used dogs experienced h29 percent better success than non-dog users that hunted ten or less times. The increase instantaneous rate of kill was 0.055. When the effect of visit frequency alone was examined the increase in rate of kill was 0.0h5. Therefore, an increase of 0.010 can be attributed to the addition of dogs to high visit frequency. An increase of 0.018 was caused by dog use in the less than eleven visit category. Therefore, the experienced hunters seemed to benefit less from the addition of dogs than did the inexperienced group. It is difficult to check the reasonableness of the magnitude of the effects of the various factors because no value is known to be correct and all values are based on samples. The most reliable value, since it is based on the largest sample, however, probably is that observed for the less-than-eleven visit, dog used, no snow category. Therefore, this instantaneous rate of kill of 0.027 was adOpted as a base from which the maximum rate of kill can be estimated by adding the effects of snow and high visit frequency, and estimating the lowest instantaneous rate of kill by subtracting the effects of dogs. In this 166 manner values of 0.009 and 0.093 were obtained for the extremes in rates of kill. The correSponding observed values were 0.0lh and 0.078. The estimated values are of the same general magnitude as the observed values which indicates that the apparent influences of the various factors on success are reasonable. It is also suggested, however, that the effects of the various factors may be slightly exaggerated due to sampling inadequacies. . Another approach to this analysis is to estimate the lowest value starting from higher values and subtracting out the apprOpriate effects. For example, the instantaneous rate of kill for the category expected to be most successful i.e. more than 10 visits, dogs used, snow cover, was 0.078 rabbits per lCO-gun-hours. The instantaneous rate of kill increase in success caused by high visit frequency, dogs and snow cover were 0.0h5, 0.010 and 0.021 respectively judging from the differences in weighted mean values. Note that the dog effect apprOpriate for the more than eleven visit category was used here instead of the 0.018 increase evident in the less than eleven visit category. When these values were substracted from the value Observed when they were present, the kill rate experienced for the least successful category 1.9. no snow, no dogs, less than eleven visits was estimated as 0.002. Starting with the success rate observed in the less than eleven visits, dog used, snow cover category and subtracting out the effects of snow and dog use of the value of 0.00h5 was obtained. Beginning with the kill rates observed for the less than eleven visit, dogs used, no snow; and more than ten visits, dogs used and no snow and removing the 167 appropriate effects values of 0.009 and 0.019 were left. A value of 0.0lh was actually observed. Thus a general agreement of the same magnitude was present which tends to indicate that the estimated effects are reasonable. VIII. Probable Effects of Seasons of Different Lengths By considering the patterns of hunting effort, kill and success on the heavily-hunted Kellogg Forest, it was possible to draw some conclusions about the effects of seasons of different lengths than the present season of October 20 to January 31. Since all seasons except one were of the same length the probable effects of a longer season must be estimated largely by extrapolation. This practice is open to some question, but it probably yields a satisfactory estimate of what could be expected if the season was lengthened at the end. However, it would be very misleading to use this method to evaluate the effect of an earlier opening date. If the season was lengthened through February, little increase in either effort or kill would occur, judging from the low effort and kill observed late in the present season. Also, in 1952 when an additional month of hunting was permitted, effort increased only 5.9 percent while the season length was 35 percent longer. about all a later closing date would accomplish would be to permit the occasional hunter who may want to hunt then the opportunity to do so. Because nearly all female rabbits become pregnant in March it seems like that for esthetic reasons February 28 should be the latest satisfactory closing date. 168' An earlier opening would be confronted with many situations which would tend to make rabbit hunting undesirable. Among them are: hot weather, very dense cover, small rabbits, lactating females, rabbits with warbles and a greater likelihood of hunters contracting tuleremia. These factors would probably make rabbit hunting unpopular with most hunters and consequently effort would be lowg JMany hunters express a strong preference for hunting rabbits even later than the current opening date. Also many people are prejudiced against shoot- ing rabbits until cold weather sets in. Because of dense cover and hot weather, success during an earlier legal period would be as low or lower than that now Observed early in the season. Because of the low effort and success the total kill would probably be quite low. Since the kill during an early period would probably be low it seems likely that an earlier opening would have little influence on the kill and and rate of kill during the remainder of the season. These views are supported by observations made in Iowa by Sanderson (personal communi- cation) where the season opens September 15. There little rabbit hunt- ing is done until November. The probable effects of lengthening the season at both ends can be summarized as follows: 1. Effort during an early extension of the season would probably be low, though, this is difficult to accurately predict. 2. Effort,during a late extension of the season would be low. 3. Effort during the entire lengthened season would probably be only slightly greater than that now experienced. h. The total kill, the early and late extensions of the current season would probably be low. 169 S. The total kill for the entire season would be only slightly higher than that now achieved during the current season. 6. The rate of kill during the early and late extensions of the season would be likely to be low. 7. Rate of kill for the entire season would be slightly lower than is experienced during the current season. On the Kellogg Bird Sanctuary and Farm, it was observed that a high rabbit kill could be made when hunting was confined to a brief period between mid-December and mid-January. It was possible then to bag as high a percentage of the rabbit population there as was made on the Kellogg Forest with over five times the effort spread between October 20 and January 31. This indicates that a short season from November 15 or December 1 through January 15 would have almost as high a total kill as is now experienced between October 20 and January 31. As would be expected, the rate of kill would be much higher than that observed over the longer season. Also hunting effort during the shorter season would probably be greater than is now observed during.the corresponding period in the current longer season. The total effort expended during the current season, however, probably is greater than that which would occur during the shorter season and this would restrict the recreational opportunities for many people. If an increase hunting quality as indicated by rate of skill was idesired while still providing a long period to hunt a season both open- ing and closing the season later than at present would be in order. 15.1.1191! ... . {and 170 For example, a season extending from December 1 through February 28 would be almost as long as the current season and would offer much better hunting conditions and a higher rate of kill early in the season. 'With the current Opening of October 20, early season success was low apparently due to dense cover, hot weather, lack of snow or other factors. DeSpite these difficulties enough rabbits were killed to cause a reduction in the rate of kill later in the season when hunting conditions were better. A low rate of kill would be expected to occur late during a long season regardless of the season's opening date. Observations made on the Kellogg Bird Sanctuary and Farm support these views in that very high success occurred when hunting started in December. It tapered off as the rabbit population was reduced. An argument against having a later opening would exist if a sizeable non- hunting mortality of rabbits occurred during the fall and early winter. This apparently does not occur on the Kellogg Station and consequently a later opening would allow almost as many rabbits to be shot as would the current opening date. 171 MISCELLANEOUS OBSERVATIONS Changes in_Mamma1 Numbers on the Kellogg_Bird Sanctuary and Farm Between 193HL35 and 1951-55:' Allen (1938b) presented a rather complete picture of the higher vertebrate life existing on the Sanctuary between the fall of l93h and August 1937. The change that has occurred in the abundance of various species of mammals since Allen's study is noteworthy. Field work con- ducted during the present study, permitted observations to be made on the relative abundance of many mammal Species in addition to rabbits. Since the population changes of these Species may be interrelated, it was thought worth-while to briefly discuss these changes. Rabbits have remained at about the same general population level. Fox squirrels (Sciurus niggg) were common during both periods. Gray squirrels (Sciurus carolinensis) were not reported by Allen; however, they were trapped several times during the present study. Red squirrels (Tamiasciurus Andsonicus) were also not mentioned by Allen; however, by 1951 they were very common, probably due to the maturing of the conifer plantations. The skunk (Mephitis mephitis) was very common during l93h and 1935. During Allen's two years of trapping skunks were handled 165 times. During the present study skunks were trapped only three times. If predation of rabbits, eSpecially young, by skunks is very common, this marked drop in the skunk population should have created a more favorable situation for rabbits. In 1953 when rabbits were being snared in order to greatly reduce the rabbit population, 172 skunks fed on snared rabbits. A picture was obtained of a skunk tugging on a snared rabbit by means of a camera trap deve10ped by Gysel and Davis 1956. Judging from Allen's statements, longtail weasels (Mustela frenats)- were not abundant during his study but were more abundant than during the period 1951-1955. Allen caught three weasels in one winter. The current effort trapped only one in five years. Both opossum (Dilelphis virginianus) and racoon (Procyon lotor) were not rare during Allen's study, but were apparently less common than during the period 1951-1955. Whitetail deer (Odocoileus virginianus) were not mentioned by Allen, so were apparently not present. Deer tracks were found on the Sanctuary during the summer of 1951 and four deer were seen by the writer in 1952. Sanctuary employees have seen wild deer several times during recent years. The only evidence of red fox as (Vglpes fulva) during l93h-35 were tracks seen for several days in January, Now tracks could be found on the area any time there was a tracking snow; Foxes were flushed twice during rabbit hunts and one was shot. During the present study a reduction has apparently taken place in the number of thirteen-lined ground squirrels (Citellus treidesem- lineatus). Allen reports that they were the most abundant mammal larger than mice and they were much more abundant than chipmunks (Tanias striatus). This is no longer true. Grouna squirrels are no longer abundant although still fairly common. Chipmunks are evidently much more abundant than they formerly were. Allen reported only 2-3 pairs. The writer estimates that in recent years the chipmunk popu- lation of the area has been over 100 each fall. The reversal in the 173 relative numbers of ground squirrels and chipmunks is probably a reflection of the cover changes on the area. The formerly open grassy fields that offered suitable ground squirrel habitat have now grown brushy and are no longer used. Meadowrmice (Microtus pennsylvaninus), white footed mice (Peromyscus leucopgs), prairie deer mice (Peromyscus maniculatus) and the short-tailed shrews (Blarina brevicauda) were common on the area during both studies. The mammalian population changes that have occurred between l93h-35 and 1951-55 may be summarized as follows: The red fox, gray squirrel, opossum, racoon, deer, chipmunk and red squirrel have increased in numbers. The cottontail; rabbit, fox squirrel, meadow'mice, whitefooted mice, prairie deer mice and short tailed shrew remained at about the same population level during both studies. The skunk has undergone a very pronounced dr0p in abundance and thirteen-lined ground squirrels and long tailed weasels also apparently decreased. Fox Food Habits Observations A . Since the red fox (Vulpes fulva) is popularly thought of as an important predator on the cottontail, an effort was made to determine the extent to which local foxes fed on rabbits. During late' winter and early Spring of 1952, a total of 51 fox droppings were collected. Most of these were found on hillside pastures adjacent to the pond north of the farm dump. Many of the scats were not fresh and it is not known when they were deposited. Rabbit, was found in 28 or 5h.9 17h per cent of the seats. In 1h of the droppings, cottontail hair made up more than one-fourth the total volume. Small rodents, chiefly Microtus Sp. and Peromyscus sp., were found in 39 (76 percent) of the scats. Chicken remains were found in 15 (29.b percent). Unidentified material was found in 18 (35 percent). The Objective of the analysis was to determine the incidence of rabbit only. Time was not taken to identify many items of low'incidence. As a result a high percentage of unidentified material resulted. Latham (1950) compiled the results of ten investigations of red fox food habits on 1795 scats and stomachs. (Eadie, 1953; MacGregor, l9h2; Darrow, l9hh; English and Bennet, 19h2; Latham, l9h3. Penn. Mammal Survey, unpublished; Nfilson, l9h8; Nelson, 1933). A comparison of the percentage occurrence of rabbits and chickens in. this study with that observed in the compilation indicated a higher incidence of both rabbits and chickens in the Kellogg Bird Sanctuary and Farm seats that in the other studies. In contrast to the 5h.9 percent rabbit and 29.h percent chicken in Michigan only 35.2 percent of contained rabbit and 17.9 percent chicken in the other studies. These differences probably reflect differences in availability. During 1950 and 1951, rabbit pOpulations were high on the Kellogg Station. Chickens were allowed on open range during the summer and when layers died during the winter, they were thrown on the manure pile and Spread in the fields. Thus chickens, both fresn and otherwise, were almost always easily available. 175 During the winter of 1952 the stomachs of nine red foxes were obtained from a local fox hunter. One of these foxes was killed on the Kellogg Farm and the rest were taken in the vicinity. Only one con- tained rabbit remains. These foxes had apparently been hunting in a marsh environment, however, because 5 stomachs contained muskrats (Ondatra zibethiso) and three bog lemming (Sypaptomys cggperi). Foxes were tracked in the snow a total of 32 miles during the late winters in 1953 and 19Sh mostly on the Kellogg Forest. No signs of rabbit kills were observed, although, 6 mice and l red-winged blackbird were caught. Unsuccessful attempts apparently were made to capture 29 mice, four cotton tails, two red squirrels, two fox squirrels, two pheasants and one quail. The foxes followed visited two dead animals, a fox which was dug from beneath the snow and a weasel. The cause of death of the weaSel and fox is unknown. The results are indicative, of course, only of fox food habits during the season involved. In a sample of 18 seats collected during the springs of 1953 and 19Sh only five (28 percent) contained rabbit. This incidence is statis- tically significantly lower than that observed in the 1952 scats (55 percent). Fall population estimates for the autumn, preceding the later sample were about 225 rabbits. In 1951 the fall pOpulation estimate was 388. This difference in population density possibly explained the difference between the two years in the extent to which foxes fed on rabbits. The lower incidence of rabbit in fox scats during years of lower rabbit abundance, however, should not be taken as evidence that the effect of foxes necessarily is less during those 176 years. The lower incidence in the later years probably merely re- flected a lower availability of rabbits. There is no evidence to indicate that the individual rabbit's probability of being caught was less or that the percentage reduction of the rabbit population by fox predation was any lower during the years of low rabbit abundance than during the year of high rabbit numbers. Data collected in this study does not permit the precise appraisal of the effect that fox predation has on rabbit populations. And, of course, foxes ranged on adjacent areas as well as on the rabbit study area, and scat analysis probably reflected feeding.habits off the area as well as on it. Some conclusions, however, can be drawn. First, it is possible to have dense rabbit and fox populations living together. During 1951 the rabbit population was the densest that it was during any year of this five year study and fox tracks were evident whenever a tracking snow was present. It has previously been pointed out that rabbit populations have remained at approximately the same level during the past 23 years, and that very few foxes were present during the earlier years of this period, but later became common. Thus it appears that the increase in foxes had no effect on fall rabbit population levels. It is, of course, not known whether or not some unknown compensating factor acted. Winter Food Habits of Kellogg Bird Sanctuary and.Farm Cottontgils During the winter of 1951-1952 observations were made of the extent to which rabbits fed on the various species of woody plants 177 found on the Kellogg Bird Sanctuary and Farm. Intensity of use classes for winter food were as follows: A - heavy, eaten in preference to other nearby species. B - moderate, eaten commonly but to a lesser extent than in A when the species occur red together. C - light, rarely if ever eaten. Since the importance of a Species depends upon its abundance as well as the extent to which it is used, the following abundance classification with numerical values was used. 1 - rare 2 - infrequent 3 - common h - abundant To obtain an indication of the relative importance of the various species in terms of both use and abundance each Species was scored as follows: Importance rating equals the abundance scale designation - times a numerical value for the use classification in which A = 5, B = 3 and C = 1. For example, sassafras which had an abundance rating of 3 and a use classification of A had an importance rating of 3 x.5 ==l5. These obdervations are listed in Table 62. In general these findings agree with other Michigan studies. There is some disagreement. Both Allen (1937) and Hickie (undated) indicate that gray and silky dogwoods were either heavily or moderately eaten. This study'indicates that they were almost never eaten. Also, Allen did not even consider oaks as a rabbit food. This was probably because lbl .l l I . . .0»: . u q n. ...... . I . u a ........... I.I|l-n -l|.|o|- : .u . :n.. -.I ...: lllll I n ... .I.- .u ..l .Ilnlll « 9-. . r L !! ...")..-IQ .: 3.10.? 8.. .... 1.4.31 953:6 (3 s... ., ... 51enm..uuem. r ......i aware .... (\thfa.) .. 43.... Ark...“ . .5312. .. .-hahn‘ 9......1h3fi. H363; .3 “1.3 we) K31». 3 «as... ..9‘. as... .5419 .5... No.91. . n .3» ”421.23.“..mihhh “59.x «M: m “.... of... 3813, awe-“Mg . 43...... Awk 1.-.. . a W ... «.4. La. .43.. n ......9. £326 1 .... .. . ..ofigzml 9-41.1} r sum»... .... marina b N m wooHo coesoo mHmcmomcwo moosnewm o N m pmSOOH xouHm dHowoauooSOmm mchom 0 m m asses .385 steam use b N m new mpnums wasOHwoeo mscflxdnm o N m poozmwoppHm mcoocmom mswpdeoo o N m passwozo essence denounce b N m hahonoow>aom oncoomcdo uOHco:MHOE¢ a m m U:3 sane and. assumes a m m Anyone mxozo «CMHeHNAH> mscshm m m m OHde oom season wood 0H N a humongous somHm mHHdpcooHooo mopsm OH N m muamnxomHm mHmcoHdozmoHHd msndm OH N « omom .aw doom 3 N a 3%... .as mass: 0H N « oHddmcwoee .dm momompmao 0H N 4 :mSQOMHocH omHom smoowpsww deduced NH 4 m mass .8 s3; NH : m Anyone waHm mchowom mscdpm NH 4 m comma mcflxssa mooHoHSEowp msHsaom NH 4 m comma spoopomamq «popcooHoMpw mSHsaom NH : m onosmmocOxnsmsm .am deOHGOH NH 2 m knoxOflz xenommzm upm>o «memo mH m a mmnmdmmmm EdHofinHm mmwmmwmmm mH m . hnponmmow omm msoooH macaw mH m < omESm madam wcHHHmaoo beam mH m . 4 sec soon mchSHo> moowodd mH m c a #60 new mHHwoaop moonoso oN 4 a omesm anoamopm smegma» poem weepsm. mchsm weeps: oemz coeeOo madz OHMHpcOHom oocopwanH condensed on: * zealous: .msszsou emcee: as; oz... 35.82%. am 83am NmAmS seem: .zBSESmS .80.; mo 83 means more mammal am mm: 92 323753 mi 95% No mamas 178 .oo .oan seemeoo .mncmHm moons op whom .ommH “To .3 ..Hozomcosz «SH $93 no memo moses OHHHpsOHom wow 832933... a N N 0 sum .HHxOHwa :nmzpnoz cacaoHaoEd :BanOSpcoN N H 0 Adobe 3.33 9813.82 mHHmpcooHooo chasm. N N 0 OH ads oom use?“ 3.82 b N m amooo oom «HCHNHE agendas.» N N o Abandon opmwpmoam mmmoaaoo 35> uHchEoo portage N N 0 sends: xOmHm new? mcmesw m m 0 05d some mcmmeqmm mafia m m o .8ng mafia: .85de mOOHm m m o oooswoo nonoaoom waochOHoem 35.30 m H m rhonamaonsgnamwm msgmo 56.3ng m H m shoemacmz ommpcoq assuage» m H m EwoboSHm mcmHHHog 5500.; m H m OSHHH 3.5..ng mmchnm a a m .389on mamOHH: mobsm a a 0 mafia :opoom mHApmothm 92ch a a l 0 mafia oom mmofimow ooo.; s s o 83% asssoz 83a. s83 J a o ooosmoo haw mewgowdwa .3980 n a o oooswoo EHHm 5:55. 9:500 a 4 o smoocoepsm mHHmfiaooHooo msmpcmenaoo m H 4 goaMH :moaowsm adoHooo 5.”on m H a sogmoe measssofl seated 3&8 m H e. melanomamosooae gmmeHd mgpcng. b N m EHm coo: can". somOHwoem mSEHD 179 oak was not available to rabbits during his study. Open, recently abandoned farm-fields during illen's study are now in a shrubby stage of succession and have an abundance of black oak reproduction in them. Because black oak (Quercus velutina) ia s dominant species in the climax community of the study area and because rabbits were feeding so heavily on it, a study was made on the effect that rabbits have on oak reproduction. It was determined that many young oaks had had all new growth pruned back annually for many years. Rabbits were apparently slowing the rate with which abandoned fields were moving into the climax vegetative community (Geis, 1951.). Allen found that rabbits fed moderately on Scotch Pine (Pinu§_sylvestris). During the present study no such feeding was noted. This was true despite considerable natural reproduction which made young trees available. Unless their food preferences have changed, it appears that more preferred rabbit foods were available during the winter of 1951-52 than in l93h-35. This seems likely because of the greater amount of woody cover currently on the area. During the winter of 195h-55, 18 quadrats each eight milacres in size were studied. Three quadrats were located in each of the six cover types (T, 31, 82, 53, W1 and W2) having woody cover. Plots sampled were selected at random. A total of 1h12 stems of 30 species were counted. The percentage used for winter food and percentage that each species made up of the total stems are tabulated in Table 63. It can be seen that these data agree very well with the qualitative TABLE 63 180 mom PLANT ABUNDANCE AND USE AS r001) BY means FRCH QUADRAT DATA WINTER l95h-1955--KE‘LLOGG BIRD SANCTUARY HICKORY CORNERS, MICHIGAN * Number Number Percent Percent Species Unused FEd Used Total Stems Upon For Food Stems Rhus copallina 18 27 60 3.2 Rhus typhina 117 23 16 9 .9 Malus sp. 1 0 0 .1 Quercus velutina h 7 63 .8 Rubus idaeus 9 6 hO 1.1 Sassafras albidum 11 29 72 2.8 Carya ovata 1 1 O .1 Lonicera Spp. 522 10 2 37.7 Populus grandidentata 5 l 0 .h Populus tremuloides 1h 0 0 1.0 Prunus serotina 58 2 3 h.2 Rosa sp. 9 5 36 1.0 Rubus allegheniensis 17 h 19 1.5 Rubus occidentalis 2 l O .2 Fraxinus americana 8 3 27 .8 Robina Pseudo-Acacia 1 O 0 .l Sambucus canadensis 11 h 27 1.1 Cephalanthus occidentalis l O 0 .l Cornus Amomum 82 O O 5.8 Cornus paniculata 233 l O 16.6 Cornus stolonifera 60 3 5 h.5 Juniperus communis var. depressa 1 0 O .1 Juniperus Virginiana l 2 0 .2 Castanea dentata O 2 O .1 Vaccinium vacilans 1 0 0 .1 Salix app. 31 7 18 2.7 Prunus virginana 29 8 22 2.6 Cornus florida h 9 69 .9 Carya glabra o 1 o .1 Rhus glabra 1 0 0 .1 Picea glauca h 0 O .3 a Authorities for scientific names same as those in: Muenscher,'w. C., 1950. Keys to woody plants. Constock Pub. Co. 181 impressions recorded in Table 62. For example, five of the six species listed in Table 63 as making up more than five percent of the total stems were in the abundant category in Table. The sixth was listed as common. The five species in which ho percent or more of their stems had been fed upon by rabbits were all placed in "A“ use category in Table 62. Judging from the large amount of woody vegetation now found on the Sanctuary that is preferred rabbit food: but not eaten, it is appar- ent that winter food could not be a limiting factor of the population. Also, the food supply appears to be increasing as natural succession progresses and formerly open fields become more brushy. a}... ‘11.": n... 21.—fl 182 LITERATURE CITED Allen, Durward L. 1937. Research on wildlife populations of the Kellogg Farm. Unpublished Report. Mich. Dept. Cons. . 1938a. Breeding of the cottontail rabbit in southern Michigan. Amer. Midl. Nat., 20: h6h-h69. A 1938b. Ecological studies on the vertebrate fauna of a 500- acre farm in Kalamazoo County, Michigan. Ecol. Mon., 8: 3h7-h36. Atzenhoefer, Daniel R., and E. D. Martin. 19h9. Condensed report of rabbit studies north central Ohio. Leaflet No. 1h. Dept. Nat. Resources. Ohio Div.'Wild1. pp. 1-20. Bedell, C. S. l93h. Rearing cottontails in captivity. Game Breeder and Sportsman, 38: 196-197. Buechner, Helmut K., and C. V. Swanson. 1955. Increased natality result- ing from lowered population density among elk in southeastern Washington. Trans. North Amer. Wildl. Conf., pp. 560-568. Bump, Gardiner. 1950. Wildlife habitat changes in the Connecticut Hill game management area. Memoir 289. Cornell U. Agr. Exp. Stat. Ithaca, New York. ~ Christian, John J., and D. E. Davis. 1955. Reduction of adrenal weight in rodents by reducing the pepulation size. Trans. 20th N. Amer. ‘Wildl. Conf., pp. 177-189. Cohen, Solomen-Solis and T. S. Githers. 1928. Pharmacotherapeutics, materia medica and drug action. Appleton and Co. 2009 pp. Dalke, P. D. 1937. A preliminary report of the New England cottontail studies. Trans. Second No. Amer. Wildl. Conf., pp. She-5w. Darrow; Robert W. l9hh. (Section on.Fox food habits in Seagers, C. B. "The fox in New York.") Educational Bul. , New York State Cons. Dept., Albany. DeLury, D. B. l9h7. 0n the estimation of biological populations. Biometrics, 3: th-167. . 1951. On the planning of experiments for the estimation of fish pepulations, Journ. Fish. Res. Ed. An., 8: 281-307. 183 Eadie, W. Robert. l9h3. Food of the red fox in southern New Hampshire. Journ. Wildl. Mgt., 7: hh-77. English, P. F. and Logan J. Bennett. l9h2. Red fox food habits study in Pennsylvania. Pa. Game News, 12: 6-7, 22-23. Errington, Paul L. 19h5. Some contributions of a fifteen-year local study of the northern bobwhite to a knowledge of pOpulation phenomena. Ecol. Mono., 15: 1-3h. . l95h. On the hazards of overemphasizing numerical fluctuations in studies of "cyclic" phenomena in.muskrat pOpulations. Jour. Wildl. Mgt., 18: 66-70. I Fitzwater, William D., Jr. l9hh. Color marking of mammals with special reference to squirrels. Journ. Wildl. Mgt., 7: 190-192. Flyger, V. F. 1955. Implications of social behavior in gray squirrel management. Trans. No. Am. Wildl. Conf. pp. 381-389. Friley, V. E. l95h. Daily patterns of pheasant hunting pressure. Jour. Wildl. Mgt., 18: 255-259. Geis, lelred D. l95h. Rabbit damage to oak reproduction at the Kellogg Bird Sanctuary. Jour. Wildl. Mgt., 18(3): h23-h2h. . '1955. Trap reSponse of the cottontail rabbit and its effect on censusing. Jour. Wildl. Mgt., 19: h65-h72. Gerstell, R. 1937. Management of the cottontail rabbit in.Fennsylvania. Penn. Game News 7: 6-7, 27, 30; 8(2): 15-19; 8(3): 12-15, 26. Gysel, L. W. and W. Lemmien. 1955. The growth and :dldlife use of planted shrubs on the W. K. Kellogg multiple use forest. Quart. Bul. Mich. Agr. Exp. Sta., 38: 139-1h5. and E. M. Davis Jr. 1957. A simple automatic photographic unit for wildlife research. Jour. Wildl. Mgt., In.press. Hale, James B. l9h9. Aging cottontail rabbits by bone growth. Jour. Wildl. Mgt., 13: 216-225. ‘ Haugen, Arnold o. l9b2. Life history studies of the cottontail rabbit in southwestern.Michigan. Amer. Midl. Nat., 28: 20h-2hh. Hayne, Don.W. l9h9. Two:methods for estimating populations from trapping records. Jour. Mamm., 30: 399-h11. Hickie, Paul. l9h0. Cottontails in.Michigan. Michigan Dept. of Cons. Lansing. 110 pp. 18h Hickie, Paul. (undated). Unpublished notes. Game Division, Mich. Dept. of Cons., Lansing. Kozicky, Edward L. and G. o. Hendrickson. 1952. Fluctuations in bob- white populations, Decatur County, Iowa. Iowa St. Col. Journ. of Science, 26(3): h83-h89. Latham, Roger M. l9h3. An ecological study of the red and gray foxes in southeastern Pennsylvania. Mimeo. report, Pennsylvania Game Commission, Harrisburg. . 1950. The food of predaceous animals in northeastern United States. Pennsylvania Game Commission. Harrisburg. 69 pp. Lemmien,'W. 1951. Hunting results 1950 season Kellogg Forest, Augusta, Michigan. Unpublished report. Div. of Cons., Mich. State U. l p. . 1952. Hunting results Kellogg Forest 1951 season. Unpublished report. Div. of Cons., Mich. State U. l p. Leopold, A. 1933.. Game management. Scribner's and Sons. h8l pp. Mac Gregor, Arthur E. 19h2. Lake fall and winter foods of foxes in central.Massachusetts. Journ. Wildl. Mgt., 6: 221-22h. Moore, Don- 1956. Oral communication. Nelson, A. L. 1933. A preliminary report on the winter food of Virginia foxes. Jour. Mamm., 1h: h0-h3. Petrides, George A. 1951. The determination of sex and age ratios in the cottontail rabbit. Amer. Midl. Nat., h6: 312-336. Schwartz, Charles W. 19h2. Breeding season of the cottontail in central Missouri. Jour. Mamm., 23(1): 1-16. Shick, Charles. 1952. A study of pheasants on the 9,000-acre prairie farm Saginaw County,.Michigan. Mich. Dept. of Cons. Lansing. 13h pp. Snedecor, G.‘W. 19h6. Statistical methods. Iowa St. Col. Press. h85 pp. Southern, H. N. 'l9h8. Sexual and aggressive behavior in the wild rabbit. Behavior 1: 173-193. Trippensee R. E. 1936. The reproductive function in the cottontail rabbit (Sylvilagus floridanus mearnsii (Allen) in southern Michigan. Proc. No. Amer. Wildl. Conf., pp. 355-350. Wilson, Kenneth A. 19h8. The wild turkey of Green Ridge. Maryland Conservationist 2h: 3-h, 25-26. F. .30. o‘- . ..- -t".‘ 31' ”rise" ,1 I , ' .‘ _" ; I f 3 ‘ ”a 5 1‘ 26 Mar 59 "7"" .-..-Ag: ‘ J HICHIGRN STQTE UNIV. LIBRRRIES llIllWIM\IIWIHIMWIIUIWIWIHiIIWIHWIWI 31293100602592