THE VALUE AND PREDICTION OF ACORN CROPS FQR DEER Thesis for Hm Degree of DH. D. MICHIGAN STATE UNIVERSITY Jerry Paul Duvendeck 1964 THESIS This is to certify that the thesis entitled The value and prediction of acorn crops for deer presented by Jerry Paul Duvendeck has been accepted towards fulfillment of the requirements for Mdegree mflihsnigs. and Wildlife LIBRARY * E Michigan Staw a University PI JET-7"?" III | .I 1 'IlIlllrtv {III 'II Ilnl'IIL’I ABSTRACT THE VALUE AND PREDICTION OF ACORN CROPS FOR DEER By Jerry Paul Duvendeck In a study conducted from 1952 to 1956 at the Houghton Lake Wildlife Experiment Station in Michigan I investigated the food value of acorns to white-tailed deer (Odocoileus virginanus) by controlled feeding of penned animals. I also investigated the possibility of developing a method of predicting acorn creps for deer. Acorn production on 6h selected oaks was studied intensively. Extensive investigations of acorn production and deer use of acorns were conducted throughout the oak forests of northern Michigan. Deer will eat approximately 1.5 lb of acorns per day per hundredweight, when they are available. Of deer fed a basic star- vation ration of poor browse foods plus acorn supplements only those given a 0.5 lb supplement of acorns per hundredweight would have had a chance of surviving a 90~day'winter period with a weight loss less than the critical 30 per cent. White oak acorns are apparently more palatable than red oak acorns, but weight loss was not significantly different between two groupsof deer, one receiving white oak acorns, the other red oak acorns, as a supplement to a poor food browse diet. Deer entering a winter in excellent condition may be better able to resist a critical winter weight loss than animals that are in poorer, lighter condition. There are wide variations in acorn crops produced between years, between different oak stands, and between individual trees in the same year. Several factors may affect the quality and quantity Jerry Paul Duvendeck of acorns available to deer. However, it appears that in years when medium or better acorn crOps are produced, and medium or better crops may be expected on the average in two out of five years, adequate quantities of sound acorns may be available to deer through the fall and part or all of the winter. Counts of oak flowers, frost damage surveys, counts of fertilized flowers, and counts of aborted acorns did not lead to an acorn crop prediction method. Observations of mature oaks, made with the aid of binoculars in late June and early July of the year an acorn crop will mature, to visually estimate the per cent of the crown area of oaks producing acorns, the per cent of twigs within the producing portion of the crown bearing acorns, and the average number of acorns per twig can lead to a general prediction of the number of pounds of sound acorns that will be produced in an oak stand, and a fairly accurate estimate of acorn availability to deer through the fall and.winter. Based on findings in the deer feeding studies and observations of acorn production and deer use of acorns in the oak forests of northern Michigan some suggestions for oak management for deer are made. THE VALUE AND PREDICTION OF ACORN CROPS FOR DEER BY Jerry Paul Duvendeck A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 196k ACKNOWLEDGMENTS I thank Mr. H. D. Ruhl, Chief, Game Division, Michigan Department of Conservation, for making this study possible. The research was conducted under Pittman-Robertson Projects, Michigan W>63-R and WL9S-R. Dr. R. A. NacMullan and Dr. Ralph Blouch directly supervised.my work while they were Biologists-in-Charge of the Houghton Lake Wild- life Experiment Station. To Dr. L. L. Eberhardt, former Game Division Biometrician, I am deeply indebted for guidance and help on statistical design and analysis. I am.also indebted to Dr. Dale Fay and William G. Youatt for analyzing deer blood samples. I thank Dr. C. T. Black for editing the manuscript. I wish to thank my many other Game Division colleagues who assisted throughout the study. I thank Mrs. Dorothy Watling and Joyce Messenger who typed and re-typed the manuscript. I wish to express appreciation to Dr. George A. Petrides and Dr. Leslie W. Gysel of Nfichigan State University for their guidance and advice. 11 TABLE OF CONTENTS Chapter INTRODUCTION . . . . . . . . . THE VALUE OF ACORNS IN THE DIET OF MICHIGAN DEER . . GENERAL METHODS . . . . . . . . . . . . . . . . RESULTS OF FEEDING EXPERIMENTS Controlled Feeding-~l9Sh . Controlled Feeding-~19SS . Controlled Feeding--l956 . Food Preference Observations O O O Q o 0 0 O O o O o O o o o 0 o o o o A SUGGESTED METHOD OF PREDICTING ACORN CROPS FOR.DEER . . . Some Factors Affecting Acorn Crops . . . . The Study Area . . . Soils . . . . Topography . Climate . . . Vegetation . 0 O O 0 REVIEW AND DEVELOPMENT OF Methods of Measuring ACORN Acorn Acorn Crop Measurements . Extent of Damage to Acorn Crops by Various 0 O O O O O O O O O O O O O O O CROP PREDICTION Crops . . . . . METHODS Factors Insects O O O O O O O O O O O O O O O O O O 0 Animal- O O O O O O O O O O O O O 0 O O O O 0 Disease 0 O O O O O O O O O O O O O O O C O 0 Methods Investigated for Predicting Acorn Crops Observation of Oak Flowers . . . . . Estimating Frost Damage to Oak Flowers . Counts of Fertilized Female Flowers . . . . . Counts of Maturing Acorns on Sample Twigs . Collection of Aborting Acorns in Mast Traps . Acorn Crap Prediction Through Visual Estimates iii 0 O O O Page Chapter EVALUATION OF PREDICTION METHODS . . Counts of Oak Flowers . . Frost Damage Surveys . . . Fertilized Flower Counts . Counts of Maturing Acorns on O O O O 0 Sample Twigs and Collection of Aborting Acorns in Mast Traps Acorn Crop Prediction Through Visual Estimates SUMMARY AND RECOMMENDATIONS . Summary of Field Studies . . . . . . . . . . . Recommendations for Oak Management for Deer LI TEP-A. trill-RE C I m 0 O O O O O 0 0 O O O O O O O O O O O O 0 0 iv Page 72 72 72 72 73 76 76 78 82 Table 10 12 13 15 LIST OF TABLES ‘Weights of Deer and the Food They Consumed on Starvation Browse Diets Plus Acorn Supplements, Jmum'MBrCh,l95hoooooooooooooooo Analysis of Covariance of Weight Data from l95h Deer Feeding Experiment . . . . . . . . . . . . . . Body Weight Loss of Deer on 19Sh Feeding Experiment weights of Deer and Food They Consumed in Feeding Experiment Comparing Food Value of Red and White Oak Acorns, Januarbearch, 1955 . . . . . . . . . . Body Weight Loss of Deer in 1955 Feeding Experiment Analysis of Covariance of Weight Data From.l955 Deer Feeding Experiment . . . . . . . . . . . . . . ‘Weights of Deer and Food They Consumed in Feeding Experiment Canparing Deer Entering a Winter in Fair and Excellent Condition, January-March, 1956 . . . Body weight Loss of Deer on 1956 Feeding Experiment Number of Sound Acorns Caught in Mast Traps in Roscommon County Study Stand I 1953-1955 . . . . . Number of Sound Acorns Caught in Mast Traps in Oscoda County Study Stand 1953-1955 . . . . . . . . Number of Sound Acorns Caught in Mast Traps in Lake County Study Stand 1953-1955 . . . . . . . . . Number of Pounds of Acorns Produced Per Tree in Roscommon County Study Stand I 1953-1955 . . . . . Number of Pounds of Acorns Produced Per Tree in Oscoda County Study Stand 1953-1955 . . . . . . . . Number of Pounds of Acorns Produced Per Tree in Lake County Study Stand 1953-1955 . . . . . . . . . Average Number of Pounds of Acorns Produced Per Tree in Three Study Stands 1953-1955 . . . . . . . Page 11 12 13 15 17 18 21 22 35 37 to he Table Page 16 Insect-Damaged Acorns Caught in Mast Traps 1953-1955 . . . h6 17 Insect-Damaged Acorns from Samples of Acorns Collected for Controlled Deer Feeding Studies 1952-1955eoooooooooooooooooooooooh? 18 Animal Damaged Acorns Caught in Mast Traps 1953-1955 . . . 51 19 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 195h Red Oak Acorn Crop Through Its Two Year Development Period in Roscommon County Study Stand I . . . . . . . . . . . . . . . . . . . . . . . . . 56 20 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 195k Red Oak Acorn Crop Through Its Two Year Development Period in Oscoda County Study Stand . . . 57 21 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1955 Red Oak Acorn Crop Through Its Two Year Development Period in Roscommon County Study' Stand I . . . . . . . . . . . . . . . . . . . . . . . . . 58 22 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1955 Red Oak Acorn Crop Through Its Two Year DevelOpment Period in Oscoda County Study Stand . . . 59 23 Aborted Acorns Caught in Mast Traps 1953-1955 by Month O C O C O O . C . C C O C O O C C C 0 O C C . O C 61 2h Comparison of Prediction of Acorn Crop with Pounds of Acorns Produced on Trees in Four Study Stands in 195hand1955 oooooooooooooooooooooo7o Figure \OCDNJO'sU'I 10 12 LIST OF FIGURES Deer pens at Houghton Lake Wildlife Experiment Station Deer feeding rack and acorn feeding box 'Weight records of 60 deer that died of malnutrition at Cusino and Houghton Lake Wildlife Experiment Stations 0 O O O O C O O C I O O O O O O O O O O 0 weight loss of deer, Houghton Lake Wildlife Experiment Station, controlled feeding eXPeriments, 195h-57 o o o o o o o o o o The oak forests of northern Michigan . . Mast traps in oak study stand . . . . . Immature acorns damaged by squirrels . . Immature acorns damaged by porcupines . Maturing acorns on a defoliated northern red oak in 1952 . . . . . . . . . . . . . . . . Prediction of the red oak group acorn crop, l95h Prediction of the white oak group acorn crop, l95h Prediction of the 1955 acorn crop . . . . . . . . Page 20 27 33 he A9 63 67 68 69 INTRODUCTION A drastic reduction in carrying capacity of much of Michigan's winter range for white-tailed deer (Odocoileus virginianus) has resulted in periodic heavy starvation losses (Bartlett, 1951), but in some years deer in heavily overbrowsed areas survive the northern Rfichigan winter in good condition. Since it is possible that acorns may be a factor affecting the survival of deer, this study was undertaken to determine the value of acorns as food for deer, and to attempt to develop a method of predicting acorn crops. It also led to some suggestions for oak forest management for deer. The study started in 1952 at the Houghton Lake Wildlife Experiment Station of the Michigan Department of Conservation, located in Roscommon County, Michigan. The presentation is in two parts. The first deals with investigations of the food value of acorns to deer by controlled feeding of penned animals. The development and evaluation of a method of predicting acorn crops are presented in the second part. THE VALUE OF ACORNS IN THE DIET OF MICHIGAN DEER In good mast-crOp years the acorns produced in the oak forests in the northern half of the Lower Peninsula of Michigan provide a large amount of food for wildlife. Acorns are classed by Forbes _e__t_a_l. (191:1) as one of the most important natural deer foods. I fed penned deer browse food diets supplemented by acorns to investigate: (l) the amount of acorns needed to sustain deer through a winter in poor browse food areas, (2) the relative food value of acorns of the red oak group compared to white oak acorns, and (3) the differences between deer entering a winter in excellent condition compared with deer starting the winter in fair condition (simulating conditions that might be found in years of mast crop abundance as compared with years of crop failure). GENERAL.METHODS The experimental deer were kept in a battery of 23 pens made of 2 widths of snow fencing, one above the other, supported by cedar posts (Figure 1). Deer were penned individually, with one exception noted later. All food was weighed fresh (to the nearest 0.1 lb.) as it was put in the pens, and all uneaten material was weighed and removed. Browse foods used in the experiments were cut from the lower limbs of trees that would ordinarily be available to deer. Fresh browse was out two or three times weekly and never stored more than 2 or 3 days before feeding. Weight changes frequently occurred in browse foods between the time they were placed in the pens and the time the uneaten portions were removed, because of evaporation or absorption of moisture, or coating by snow and ice. Control feed racks were used as an aid in checking this variation and adjusting the computed daily food consumption. 'Wooden feed racks with Chicken wire tops held browse foods fully assessible to the deer. Acorns were put in small wooden boxes (Figure 2). Loss in deer body weight was used as the main criterion of diminishing physical condition. In controlled-feeding studies at Michigan's Cusino Wildlife Experiment Station during l936-hl, Davenport (l9h6) found that a body weight loss of around 30 per cent during the winter period was the maximum a deer could sustain. This percentage appears to be supported by recent records (Figure 3) of 60 penned deer dead from.malnutrition at the Cusino and Heughton Lake Wildlife Experiment Stations. In my experiments, when a deer Figure 1. Deer pens at Houghton Lake Wildlife Experiment Station. I“ 1 , " I . Ii" " 'I" k: HE» |':u' ‘u fli‘f’”:'; '1 Ill‘lllflflw- "WWII ';:LD ."" Figure 2. Deer feeding rack and acorn feeding box. Weight Loss At Death 70 60 T +— 0 Juvenile female A 0 Juvenile male ‘ Adult female *‘ A Adult male 111111111'111 4O 60 80 IOO I20 I40 |60 Initial Weight Figure 3. Weight records of 60 deer that died of malnutrition at Cusino and Houghton Lake Wildlife Experiment Stations. had lost between 25 and 30 per cent of its fall weight, I stopped controlled feeding and put the animal on a good food diet. Blood samples were taken periodically from deer on my experiments and analyzed for red and white cell counts, hemoglobin content, packed cell volume, non-protein nitrogen, and phosphorus, but no blood factor was discovered that indicated the physical condition of the deer. Deer used in the experiments were either wild-trapped stock or semitame deer raised from.fawns at the Ogemaw State Game Refuge, Ogemaw County, Nfichigan. Since semitame deer were easier to handle (thus less likely to injure themselves) than wild-trapped deer, and since preliminary experiments revealed no apparent differences in reaction to feeding between the two types of deer, I used semitaue animals exclusively in the remainder of my experiments. The controlled feeding experiments were designed to be conducted for 90 days during winters of l95h, 1955, and 1956. The controlled; feedings were designed as separate experiments and analyzed as such. No statistical comparison was attempted between years, as Snedecor (l9h6) points out difficulties encountered in attempting to compare statistically experiments conducted under different environmental or climatic conditions. In the analysis of results, 60-day, 71—day, and 73-day weights were used in l95h, 1955, and 1956, respectively, because many deer had to be taken off controlled feeding shortly after these periods. The procedures and results of each experiment will be discussed separately. In the fall of 1952 acorns were abundant. In most oak areas in the northern Lower Peninsula.more were available than deer could consume. Nichol (1938) suggested that when acorns are plentiful deer imay eat little else. Davenport (1937) reported that three deer fed balsam (flies balsamea) and oak (Quercus app.) browse plus all the acorns they would eat consumed between 1 and 1.5 lb of acorns daily. I examined the rumen contents of 15 wild deer killed during October, 1952 and found that they had eaten approximately 1.5 lb of acorns per 100 lb body weight. 'Eight penned deer that I fed a mixed browse diet plus a free choice of acorns ate approximately 1.5 lb of acorns per day per 100 lb body weight. My first feeding experiment was designed to determine whether deer could survive a winter on poor quality browse species plus an acorn supplement of less than 1.5 lb per day. RESULTS OF FEEDING EXPEEIRENTS Controlled Feeding--l95h Eighteen deer were split into two similar groups of nine deer each. One of these groups received, as a basic diet, free choice of jack pine (Pinus banksiana) and oak browse. The other group's basic diet included a free Choice of balsam fir, plus 10 per cent of the amount of white cedar (Thuja occidentalis) necessary to sustain a deer in good condition (approximately 8 oz per 100 lb body weight per day) as determined by Davenport's (19h6) feeding experiments. Each of these two main groups was further divided into three sub- groups, each containing an adult doe and two fawns, except one subgroup made up of three fawns. Pairs of subgroups representing both main groups each received a different level of acorn supplement. One pair was given 0.5 lb of acorns per day per 100 lb body weight of deer, the second pair was given 0.25 lb of acorns per day per 100 lb body weight, and the third pair was fed no acorns. The acorn supplement was fed as a mixture of three Species and consisted of approximately 50 per cent Hill's oak (Quercus ellipsoidalis), approximately h2 per cent northern red oak (Quercus rubra), and approximately 8 per cent white oak (Quercus alga). benard, gt_§l. (1935), Davenport (19b6), and Dahlberg and Guettinger (1956) demonstrated through controlled feeding that deer will starve on balsam alone. Since oak and jack pine browse are classified as starvation foods by Davenport (l9h6) and Dahlberg and Guettinger (1956), I believed that my deer would not be sustained on these basic diets, and any differences demonstrated between 9 10 groups would be the result of the acorn supplement. After 60 days on the above diets, half of the 18 deer had lost more than 25 per cent of their original weight (Table 1). Only' one of the deer receiving no acorn supplement could be carried beyond this point. After 66 days, only 7 of the original 18 deer could have continued. Five of these were receiving the 0.5-lb acorn supplement. An analysis of covariance (following Snedecor, l9h6) failed to demonstrate significant differences among adjusted mean weight losses in the three acorn-supplement categories (Table 2). However, since the weight-loss classes of groups of deer with different acorn supple- ments can logically be arranged in a sequence (i.e., O, 0.25, and 0.5 lb/cwt) without reference to the outcome of the experiment, it seems legitimate to compare the extreme cases (0 and 0.5 lb/cwt) here. A "t” test was used, wherein the difference between adjusted mean weight losses for the groups of deer on the O- and 0.5-lb acorn supple- ments (18.3 and 1h.0 1b respectively) was compared with a standard error computed in the manner described by Cochran (1957). The canputed "t" value (2.hb with 13 degrees of freedom) is significant at the 5-per cent level and falls just short of the 1-per cent level of significance ("one-tailed" test). The covariance adjustment of mean losses (to a common initial weight; 67.9 lb in this experiment) gives percentage losses (last line of Table 3) which.differ very little from the unadjusted rates. On the assumption that the deer body weight loss rates as indicated in Table 3 would continue over an additional 30 days, only those deer receiving 0.5-lb acorn supplement would have a chance of surviving. Therefore, an acorn supplement, of the species tested, less than .euofio ooHHouunoo no exeo om wouu< nee .eueuo oeHHouunoo no eheo mo wouw< «e .euowo ooaaouunoo no wheo co Heum< e A.nwu oesnmnoo ooom deuce I}, 'i TnU newnmzfi a.dm s.os q.~m seem mod on m mmm m.en m.am s.~a sane mm an a she use you ewe use «atone .sn m.HN m.nm o.sm news we so z own «.0 mafia omaoup ueoeo one Eeeaem m.m~ H.o¢ N.¢~H «was no o< m mmw m.a m.wm a.mm toe mm as a New use tea awe use echoes .nn w.n m.os n.Hs sac mm no m sow m~.o moan mesons ueoeo one seeHem o s.s¢ ¢.moa «as we o< m maukome o m.m¢ H.N- nos mm no m mow ennoue oz o m.m¢ “.mo~ sens 00 no m Cam eesoun neocu one seeaem mnuooe weooo EeeHem H.0N ~.om o.mo «or am o< m omm ~.~H o.noH c.0m sewn do so m Hum use you zeo non enuooe .nH o.- m.m- m.mm «ewm Om no 2 new m.o moan mesons xeo one onwn xoen ¢.AH o.m~ m.~m eeoo um o< m Nmiammq ¢.s o.m~ H.moa ems no no m mam use you zeo non enuooe .na m~.o e.nn H.o~H H.~w eene on as 2 mew mane masons xeo ecu mane some 0 o.m~ H.~N «seas mm so m com o o.sm m.w~ nos we so m Hum ennooe oz o m.~w m.ooH «on on no m mom mesons xeo one onwm xoeo mnuoo< xe0. onwm xoeo aenwm Henawwuo ow< xom .oz noun neon smmaidnouezihuennemi.eunoeoammnm nuoo< enHm illl I'll euean museum neaue>ueum no oosnenoo henH ooom onu one noon «0 eunwwez a «name 12 qw.qu N omo.mq meoH news oeuemfive mo eoneoawfiane mo ueou pom mm.w «H omm.N~H oo.mmN No.~¢~ on.moo.m ma moOa nqnuaz oo.¢H mo.m as.mmw N eunesofimmne nwoo< 0H oHo.NmH oo.o~m mm.mmm sm.¢¢¢.¢ ma deuce onenom soooenm eeuenwm NmMnmvm «MrmVAmuxvm NAwavm Booeeum noHueaue> new: we eeoumon we now mo eeeuwen mo mounom Qumawumm NO wHOHHM muonooum one emuenum mo warm unoawuomxmwnwoeem noon «mmH scum euen unwfie3 mo eonewue>oo mo manwwenn N ensue 13 .nH ca mead unwaos one unwges Heauflna Home a o.o~ o.q~ m.o~ auxou eemw mead eweunoauen oeuennoo m.o~ m.¢~ m.m~ eeoH aweuneouen meouu am one cod mmq mm omm eHeuoa s.- N.o~ H.5N eeoH eweuneouom emouu Hm moH mm um Hm we NH mm ma on ma mm «H no NH mm ma co Heooonaeeaem o.o~ m.- ¢.wN mead eweuneouea eeouo mm mm 0H mm «H mm m Ho oN mm mm mo an on mg on ~H es xeo-uaaa goes eeoq unwaea moo; unwwez ahv mead mxv unmwuz enema owmem name»: Managua agents neauncn teens: Henunan use Homing m.o use non pH m~.o onoz eunesefimmnm nuoo< a m wanea unenauomxmwnaoeem smaa no noon mo mead unwaoz moon 1h 0.5 lb per day per 100 1b body weight would not be sufficient to carry deer through a 90-day winter period. Controlled Feeding--l955 During the winter of l95h-55, a.feeding experiment was conducted to investigate nutritional differences between white oak and red oak acorns. Forbes, gt_al.(19hl) analyzed red oak (Quercus £2233) acorns and reported, "The values for composition, digestibility, and digestible nutrients indicate that the acorn is one of the most important of natural deer feeds." It appeared to me that deer in the wild preferred acorns of the white oak group over those from the red oaks. Given a choice, penned deer ate white oak acorns first. Sixteen deer were divided into four groups each containing an adult buck, an adult doe, a buck fawn, and a doe fawn. Three of the groups received jack pine and oak browse as a basic diet. In addition, one group got 0.5 lb of white oak acorns per day per 100 lb body weight, the second had 0.5 lb per 100 lb body weight of red oak acorns, and the third received no acorns. The fourth group was fed a "good-food" diet consisting of free choice of Kellogg deer pellets (a deer food concentrate produced by The Kellogg Company, Battle Creek, Michigan), shelled corn, apples, cedar, aspen (Populus tremuloides), juneberry (Amalanchier spp.), and cherry browse (232223 serotina). Table h presents deer weight and food consumption data collected in this experiment. A “t" test does not show evidence of significant differences between mean losses in the groups receiving red and white oak acorn supplements (t - 1.28, 10 degrees of freedom). Both types of acorns are apparently good deer foods. An analysis of covariance, including 15 .mumflo ooaaonunoo no m%eo Hm nouw< H H.ms H.on o.o w.c e.n o.No m.mo ea m.mc as a mmowm e.m H.N o.oH N.m o.a m.me o.wa so co es 2 omowm H.m w.wm H.Hn H.NN e.oH m.ae o.Na so ooH e< a see e.om m.n c.w N.eN e.a m.so m.ws awn wen o< 2 see oooe ooou humane munon new ueo .mwwm nuoo muoH nonno nm< too ua< iaem o o.eo e.meH os mm as e emmm o m.oe N.eNH em ea ee 2 amowm o N.sen o.NoN can m.omn es 2 ems o e.msn H.mHH em aoH es a ems omsoso two one cone sees mnuoo< oz s.o~ a.m~ o.cNH we m.mo em a Nocww m.om ¢.¢HH m.m¢ mm mm no 2 moowm uzo neg %eo pom mnuooe e.Ns N.Nm o.ae man use e< z oemm seo set on m.o mete e.Nm o.aan o.mw mm men es e wmmm mesons xso one cone some mnuoo< xeo omm m.m~ a.ooH w.0m mm m.eo so e Hocwm w.m~ s.me n.0a mm mm es 2 Neemm use use ass coo mosoos m.mm o.mmn H.sa me as ea a mom sso scans on m.o sens s.~e s.ma w.man not NNH e< z owe seeosc seo one cone some mucosa sso sons: see ease some Haesne Hsoswnso ewe xom .oz cone Aoav ooEnmnoo ooom deuce Anav unwflox noon mmen .oossziiasssosn .eouooe xso sons: one tom mo sons> ooom wnflnemEoo uneEHnemxm wnfloeem nH oeanmnoo %o£H ooom one neon mo munwwea s stoma 16 all four groups (Tables 5 and 6), shows highly significant differences among adjusted mean weight losses. Since the mean weight loss of the "good-food" group was so much less than the loss in the other groups, a ”t" test was run contrasting the 0.5-lb and no-acorn diets (thus excluding the "good-food" diet). The test falls short of significance at the 5-per cent level (t - 1.69, 10 degrees of freedom; "one-tailed" test). The adjusted percentage losses are again very nearly the same as those obtained from unadjusted data (Table 5). The deer receiving no acorns lost less weight than a similar group during the previous year's feeding experiment. From general observation of the deer on this experiment as compared to those from. the previous year, and because the original weights in 1955 were heavier than those in 1951:, I believed that the animals starting the 1955 feeding were in better condition than those starting the l95h experiment. This probability influenced the design of the 1956 feeding. Controlled Feeding--1956 During the winter of 1956, a controlled feeding experiment was designed to investigate differences between deer entering a winter period in excellent condition and deer starting the winter in fair condition. To determine whether a deer was in excellent or fair condition, the general appearance of the animal and the weight gain, or the lack of it, through the fall period were used. The terms "excellent" and "fair" are defined as follows: Excellent--having an excess accumulation of fat. Fair--only a.moderate amount of body fat. 17 .nH nw mmoH unwfloa one unwfieB Hefluwnfi noooa m.¢ A emu momv meoa oweuneouea oeumnmom N.o mmoH oweunoouom mmouu Aessev s. so N we H OOH Hm qu ueHo Hefioeom H.HN m.an e.mN Asses same each oweunoouoa ooumnmo< N.HN w.mH o.mm mood oweunouuoa mmono 0H so NH Om mH mm ma mm NH mm ON on mm moH ma am on omH an men Hm «NH mm not aso-oaee some mmOA unwaoz mmoq unwwea Aav mmoq Axv unwfloz unwfiee HeHanH unwfioz HewanH names: HefiuwnH mnnou< xeo oom mnnoo< xeo oufln3 onoz uewa semen use and La m.o uBo mom AH m.o munoEodmmnm nnou< H m sense unonwnooxm wnfioeoh mmma nw mean we mmoq unwwez xoom .Ho>eH ones “on H ue ooneofiMHnwflmH Ru 1 am .wom m mom.¢ow mmoH neeE oeuenmoe mo eoneoflmwnwfie mo umou pom H¢.oH HH moq.wma mm.wa mm.omm.m om.mmm.ma NH muoH nflnuwz ma.mow No.moH mm.0ma m muneEeHoone nu00< an wea.Nee ee.eNa.n em.sse.m ma.mam.mn me Hence onenwm Eooeoum meuenvm NA%n>vm .Mn%VAMnxvm NAmdxvm Eoooeum new: we moonwea mo 85m . mo nowuewue> oueEHumm mo muonum muoDfiOMm @fim mwkmfldm m0 mESm mmwuwom mo eupnom unoEHnemxm wnfioeom poem mmma Eoum euem uaneB mo monewne>oo mo mamhaenm o eaaefi 19 This test was designed to simulate conditions that might be found in years of mast crop abundance as compared with years of crop failure. Nichol (1938) reports records of deer going from poor to good condition in 15 days when acorns are abundant in the fall. Three groups of four deer each were used. Each group contained two adult does, a buck fawn, and a doe fawn. In October, the deer of Group 1 were placed on a half-ration diet so that they would not accumulate an excess of fat. The adult deer in this group lost weight between October and January (Table 7). At the same time the other two groups had an unlimited ration of good food. From January 9 through March 23, 1956, all three groups were fed a basic free-choice diet of jack pine and oak browse. Groups 1 and.2 got nothing else, while Group 3 was given an acorn supplement of 0.25 lb of acorns per hundredweight per day. Only those deer receiving an acorn supplement were penned individually. One adult doe in Group 2 was injured at the end of the first week and was removed from the experiment. The four deer in Group 1 were placed together in a pen, as were the deer in Group 2. This grouping of deer was necessary since we were con- ducting other feeding experiments at the same time and our number of holding pens was limited. Table 7 presents deer weight and food consumption data collected in this experiment. The results indicated a trend toward differences in weight loss percentages of deer entering a winter period in good condition as compared with deer entering in only fair condition (Table 8), but there is little evidence of significant differences. 'Weight loss records of 77 deer on experimental winter diets at Houghton Lake (Figurell) indicate, in general, that deer in good I50v O |4C)_ 0 I30*- ‘ A I20- 5 E3 0 60 Day Weights a) to o o ‘1 O O} O Figure h. DIETS Browse plus 32- lb. acorns Browse plus 0 log; lb. acorns Cedar browse Jackpine or aspen browse l I l I l l l l I 70 80 90 '00 ”0 I20 I30 I40 ISO Initial Weight Weight loss of deer, Houghton Lake Wildlife Experimmt Station, controlled feeding experiments, 19514-57. .ononw Scum oe>oaem .mnfioeom no m%eo m wouwe sen Hosea exoum est .muefio oeHHouunoo no emeo mm weave %* .mmnnauoH mnoaueu oeonoeu no oeoeao nens unwaoz e 21 m.mH m.om n.5sa mo Hm no 2 an N.m s.mHH ~.mHH cm so no m Nooo .noauqonoo unoHHooxo nH o.HN m.mmH m.mo NoH mHH on m HHoo neunfis aunH .mnuooe AH m~.o o.mH o.mo m.omH mm mHH o< m oHoo one mesons xeo one enfio xoeo o in mHH o< m «semaw o no we no 2 mmH o m.nmo ~.amm 00 on no m mooo .noHuHonoo uneHHooxo nH wounas 0 man Hen e< a new ooeH .omsoac see one send rose 0 0m emo\Ho no m sum 0 a.mwo a.moo mm tow\oo no 2 comm o moH em~H\NoH o< m sow .noHuHonoo uwem n“ wounds 0 mm eoHH\NNH o< m mom ounH .eesoun see one onao xoeo enuoo< xeo enum xoeo teaenam Henwwauo ew< new .02 NnHv oeanmnoo ooom Heuoa rdnav umwwoz neon sown onma .nouezwmuenneo .nowuaonoo unoHHooxm one when n“ Henna: e wnwueunm neon wnHueoEoo unosfiumnxm wnfioeom ng oeanenoo zany ooom one Home «0 eunwaes n wanes 22 .nH nH mmoH uanes one uanes HeHanH nmoQH H.HN mH we mm om om mNH HN oHH mnuooe on .xeoienHo xoeo uHem «.mH mH mm «H on mm HsH enuooe on .xeo-oan xoeo unoHHeoxm m.wH 0H Hm oH oo NH aHH use you om mHH enuooe AH mN.o .xeoioan xoeo uneHHooxm mmOH mmOH uanoz ueHn nOHuHonoo HeHanH eweunoouoo mmouu uanez HeHanH HuneEHueoxm wnHoeom ommH no Home mo meH uanoz hoom w oHneH 23 condition resist winter weight loss better than deer in fair condition. Food Preference Observations Other controlled feeding studies have indicated that deer'show marked preferences in their choice of browse species, and that in general preferred foods are the more nutritious foods (Maynard, §t_al., 1935; Davenport, 1937; Nichol, 1938; and Dahlberg and Guettinger, 1956). Palatability and browse utilization studies have led to lists of seasonal deer food preferences (Nichol, 1938; Petrides, l9hl; and Davenport, 19h6); but I can find scant reference to indicate that one deer might show a.marked preference for a specific browse species that is different from the preference shown by another deer. That such individual variations do exist was demonstrated by the deer I fed basic diets of jack pine and oak. Deer 6010 ate about three times as much jack pine as ask; in contrast deer 6011 ate about three times as much oak as jack pine (Table 7). Tables 1 and b show similar differences. As indicated by weight loss of these different animals (Tables 2, 5, and 8), there is little evidence that any one of the various combinations of these two browse Species chosen by the deer constitutes a better diet than any'other. The four deer fed a good-food diet in 1955 also showed differences in the amount of individual good-food species they ate (Table h). In one instance a deer (819, Table I) refused to eat acorns for over a.mcnth; when it did finally start, it would not eat all that were offered. This was the only deer that did not eat all of the sound acorns offered, shells and all. Occasionally insect-infested acorns were inadvertently offered; these the deer would not eat. 2h The fact that such individual differences did occur in deer choice of both poor and good browse food suggests that in a deer feeding experiment as many animals as possible should be used. Adequate numbers of animals increase the chance of bracketing variations that might lead to erroneous conclusions concerning deer in general. A SUGGESTED METHOD OF PREDICTING ACORN CROES FOR DEER The size of acorn crops interests game managers because acorns are a valuable wildlife food (Martin, 3131., 1951). Wildlife managers could use acorn crop predictions as an aid in recommending deer hunting regulations liberal enough to permit harvest of surplus deer that might not survive a winter without the supplemental food provided by an acorn crop (Duvendeck, 1957). Some Factors Affecting Acorn CTEBE A review of some of the factors affecting acorn crop production will aid in understanding some of the difficulties that may be encountered in attempting acorn crap prediction. Oak trees begin acorn production between'20 and h5 years of age and probably reach a.peak between hO and 80 years (Kittredge and Chittenden, 1929). Downs (19h9) states that some asks are probably good producers, and some poor, due to heredity, He also states that oaks produce good craps of acorns once in 3 or h years on the average. Davenport (l9h6) estimated that in Michigan oaks produced good acorn crops in 2 out of 5 years on the average. Reid and Goodrum (1957) found the black (or red) oak group had a larger percentage of acorn- producing trees than the white oak group. Christisen and Korschgen (1955), Downs and McQuilkin (19hh), Gysel (1956), and Kittredge and Chittenden (1929) point out that acorn production may vary widely' between years; between oak species; between similar trees of the same species; and between oak stands in the same year. Verme (1953) found that acorn production on trees was greatest in the part of the 25 26 crown exposed to sunlight. From the time oaks blossom until the time the mature crop falls (5 to 6 months for the white oak group and 16 to 18 months for the red, or black oak group) many factors can reduce the potential sound acorn crop size, and the portion of the crop that would become available to deer. Allen (l9h3) suggests that long wet periods in the spring may curtail fertilization of oak flowers. He also states that in Michigan late spring frosts that kill oak flowers are probably the most important factor reducing potential acorn crop size. Turkel, gt_§l, (1955) describe formation of an abscission layer between the developing ovary and the involucre in white cake that causes early abortion of immature acorns. Birds and squirrels may eat.more than 20 per cent of an acorn crap from oak trees before the crop would normally fall (Christisen and Korschgen, 1955). They also state that in Missouri insects damaged over 80 per cent of post oak (Quercus stellata) acorn crops from 19b? to 1952. Martin, et_al. (1951) list h5 species of animals found in Michigan that use acorns for food. I observed porcupines, not listed by them, feeding heavily on acorns in some areas. The Study Area The study'area included the 1.5 million acres of oak forests in the northern half of the lower peninsula of’Nfichigan for extensive surveys. I used portions of four oak stands, one in Lake County, one in Oscoda County, and two in Roscommon County for intensive studies (Figure 5). PRESQUE ISLE Q I g 9] A‘LPENA OTSEGO MONTMORENCY GLADWIN ISABELLA MIDLAND * Housman LAKE WILDLIFE EXPERIMENT salmon O INTENSIVE eruov OAK ammo FIGURE 5. OAK FORESTS ti IOITIERI, MICHIGAN , 28 sails. Kittredge and Chittenden (1929) described the soils found throughout the oak forests in northern Michigan as "a group of closely similar upland well-drained soils which may be described in general as sands or light sandy loans to depths of three to five feet or more. They are characterized by a very thin humus layer, low moisture content, low moisture-holding capacity, acid reaction to three feet or more, and low fertility.“ The soil type in my Lake County study stand is Roselawn fine sand. In the Oscoda County stand the type is Roselawn sand, and in the two Roscommon County study stands the soil type is Rubicon sand. T_0posraphx The topography in northern Michigan oak forests ranges from flat to rolling to hilly lands with steep slopes. The Lake and Oscoda County study stands are in rolling areas with moderate slopes. The Roscammon County study stands are in a smooth plane area with slopes less than six per cent. Climate Climate in the northern half of the Lower Peninsula of Michigan as described by Wills (191a) is of two distinct types. Narrow belts along the great lakes shoreline have a modified marine climate. The interior counties have a climate that alternates between continental and semimarine. When there is little wind there may be pronounced fluctuations in temperature with hot weather in summer and severe cold in winter. A strong wind from the Lakes may immediately 29 transform the weather into a seminarine type changing the temperature as much as 20 F. Precipitation is fairly well distributed through the year, averaging around 30 inches. Vegetation Hill's oak, northern red oak, and white oak are the three oak species most commonly found in the oak forests in northern Michigan. They occur as nearly pure stands of Hill's oak or northern red ask or as varying mixtures of two or all three species. Jack pine, red pine (Pinus resinosa), white pine (Pinus Strobus), red maple (Acer rubrum), trembling aspen (Pmlus tremuloides) and large-toothed aspen (EQPulus grandidentata), are the tree Species other than oak most frequently found in the oak forests. The shrubs I most commonly found in the oak forest understory are Juneberry (Amslanchier spp.), willows (Sill); spp.), hawthorns (Crataegg spp.), dogwoods (m spp.), sumac (El-TEE app.), witch-hazel (Hamamelis Virginians) and hazelnut (Corylus americana). In general the shrub understory in the oak forests is sparce and shrubs are found where Openings in the overstory occur. Ground cover plants I found most frequently were bracken fern (Pteridium spp.), sweet fern (Comptonia peregrine), blueberries (Vaccinium app.) and several species of grasses that I did not identify. Ground cover through the oak forests varies from Sparse to dense. In Roscommon County study stand I there are approximately 16 oaks per acre over 12 inches dbh. Composition by species is approximately 60 per cent Hill's oak, 35 per cent northern red oak, and 5 per cent white oak. In the understory I found some scattered Juneberry and willow. There are also scattered red pine 15 to 20 feet in height. 30 The dense ground cover consisted principally of sweet-fern and bracken fern. In Roscommon County study stand II there are approximately 8 oaks per acre over 12 inches dbh. Composition by species is approxi- mately 67 per cent Hill's oak, 31 per cent northern red oak and two per cent white oak. There are a few scattered red maple and aspen in this stand. The shrub understory has willow, juneberry and hawthorne scattered through the stand. Ground cover is relatively dense and consists principally of sweet-fern and bracken fern. In the Oscoda County study stand there are approximately 18 oaks per acre over 12 inches dbh. Composition by species is approximately 55 per cent Hill's oak, 25 per cent white oak and 20 per cent northern red oak. A few aspen and red.maple are scattered through.the stand. Shrubs are scarce. A few witch-hazel were noted. The moderate ground cover found consists principally of sweet-fern, bracken fern, and grasses. In the Lake County study stand there are approximately 20 oaks per acre over 12 inches dbh. Composition by Species is approximately' to per cent Hill's oak, 35 per cent white oak, and 25 per cent northern red oak. Shrubs are very scarce. Ground cover is moderate and consists principally of bracken fern, grasses and scattered sweet-fern. REVIEW AND DEVELOPMENT OF ACORN CROP PREDICTION METHODS To determine the accuracy of an acorn crop prediction.method a.measurement of the mature crOp is desirable, and in attempting to predict acorn craps, all of the factors that can reduce the potential crop size must be considered. Methods of Measuring Acorn Crops Various techniques have been.used to measure acorn crops on individual trees and in oak stands. Verme (1953) made total counts of the number of nearly mature acorns on individual trees. This method, though very accurate, he states is too time consuming to use over extensive areas. He also found that sample area counts on trees agreed closely with total counts. Petrides, gt_§1, (1953) counted the number of post oak acorns on the first 25 twigs encountered on each tree as an index of year to year acorn abundance in East Texas. This system is suitable to secure stateawide information on acorn crops where oaks have readily accessible branches. Mbst northern Michigan oaks do not. Allen and McGinley (19h?) and Cypert and webster (19h8) used ground counts of acorns falling on triangular shaped plots approximating 10 per cent of the crown area of oaks. Dalke (1953) counted seeds and acorns on mil-acre ground plots spaced at one chain intervals along surveyed lines. Downs and McQuilkin (l9hh) used box shaped traps with chicken wire tops to sample acorn production of individual trees. Easley and Chaiken (1951) described an inverted pyramid-shaped seed trap made of kraft paperboard mounted on a wooden frame. It samples an area 3.3 feet square. Seeds are held in a wire basket fastened to the cpen bottom. Though not 31 32 completely animal proof I believed (and as explained later found) there would be little loss of acorns caught in this type trap. I used this type trap to sample acom‘production of oaks in my four intensive study plots. Gysel (1956) later designed a similarly shaped trap made of waterproofed canvas mounted on a galvanized pipe frame. He covered the top with poultry netting to reduce animal entrance. This type trap is more durable than the paperboard type and more efficient for fewer acorns bounce out. Gysel (1956) tested both the canvas and paperboard traps by dropping 100 white oak and 100 black oak acorns on the traps from.a height of 30 feet. He found that 78 per cent of the white oak and 58 per cent of the black oak acorns were deflected from an uncovered paperboard trap while only 32 and 12 per cent respectively were deflected from an uncovered canvas trap. Since the mature cake I studied in northern Michigan are not as tall as those Gysel delt with in southern Michigan, and thus the unobstructed free fall of acorns was less, I tested uncovered paperboard traps from a height of 20 feet. I drapped 250 acorns of each northern red oak, white oak, and Hill's oak, and found that hl, 22, and 18 per cent respectively were deflected. I used these figures to adjust the number of acorns caught in.mast traps in my study areas. I used the following approach to measure acorn production of individual trees. I placed two mast traps under the crown area of each tree samples in three of my study stands, one in Lake County, one in Oscoda County, and in Roscommon County study stand I (Figure 6). Traps were placed in August, 1953 and data collected for the 1953, l95h, and 1955 acorn crops. Sample trees were chosen by entering the stand 33 Mast traps in oak study stand. Figure 6. 3h from a road and picking each mature full crowned oak tree encountered that was 12 inches in diameter, or over, at breast height. In Lake County 25 trees (nine white oak), ten Hill's oak, and six northern red oak were sampled. In Roscommon County study stand I 19 trees (one white, 11 Hill's, and seven northern red oak) were sampled. In Oscoda County 20 trees (five white, 11 Hill's, and four northern red oak) were sampled. Traps were placed under opposite sides of the crown area of the trees to collect acorns from two areas of the crown that appeared to be of average density and crown exposure to sunlight for the particular tree. In Roscommon County study stand II 50 mast traps were placed at random throughout a hO-acre oak stand. Here I could expand the proportion of the acorn crop caught by the traps directly to the hO-acre area, rather than to individual trees as in the other three study areas. The area was divided into 10.9 square foot blocks (the area of one mast trap). The blocks were numbered consecutively and numbers were drawn from a table of random numbers to determine the location of the 50 traps. The traps were placed in a chosen block only if it occupied an area on which acorns could fall. It was necessary to draw 99 random.numbers to have 50 plots under the crown area of oaks. The traps were not set in this stand until 195h. Data was collected here for the 195k and 1955 acorn crop. Acorn Crop Measurements The number of sound acorns caught in mast traps during 1953, l95h, and 1955 are presented in Tables 9, 10, and 11. These data 35 oe ch mN Hence 0 eN o e.N NH N em N m.m HH mH ems NH Hscoe e No N m.m oH H mcH m e.H e m eN o m.m eH m He e m.m m m ceH N e.H wH m He n e.e N m eHH c m.N NH O we H m.m o 0 so N H.m cH n eH o m.m m m cH H m.m mH m Nm N N.N s e Hm o e.N sH m eN N H.m m N em H H.m mH a me o m.N H mmsH emeH mmeH eoHeEsm .oz mmeH emeH mmeH eoHeasm .oz wnuoon mo uenEnz eou< nacho eons wnuoo< mo nooEnz eou< nacho cope mo unmouom mo unmouom m Ho m H.m N mmeH emeH mmeH eoHessm .oz mnuoo< mo uenEnz econ nacho come we unoopom aso sea #8 m. HHHZ rso ounce mmeHImmmH H sebum sebum eudaoo soesooeom oH «gene umez nH unwneo enuoo< onnom mo neonnz a oHneH HmH NqN N HeuOH aw om we 0 o.H mo Nm 0 H.m Nm mm o N.H 0N om H N.N mmmH ommH mmmH oonEem mnuoo< mo Honsnz een< nsouo mo uneuuom ON oH HH Ooz coma mms Nwo.H mm Hsooe on Na m m.m oH om No m H.o wH HN moH N H.m NH Nm cm 0 mom mH mm Nm N w.¢ «H No oNH N m.m OH Hm 00H 0 w.o m mm NHH N q.¢ w on HmH om w.m N on omH o w.H 0 mm on N ¢.N N mmmH qmmH mmmH oonEem .oz enuoo< mo uunEnz eeu< nsouu cone «0 uncouom Nm mum NoN MN ON w o mH mH MN co m NN um NmH oH NmH oN mmmH ommH mmmH mnuoo< mo neoEnz Hence m.m mH H.m NH m.m e H.H m a.N H eoHoasm .oz eou< nacho cone mo unmouom aso eta x8 m. Haw: aeo cases mmeH-mmeH eesam sebum sooeoo seooeo aH moeue use: nH unwneo mnnoo< onnom mo nonEnz oH oHneH 37 NHH wa N Heuoa No nHH NH Hence 0 MN 0 H.o NH m NNH o H.o «H m m N m.o «N om mmm 0 Hence 0 NoH o H.m oH N m o H.¢ MN on owH o N.N m m m o w.m mH N o o m.N mN m MN 0 w.m N H wH m m.m mH m owH o m.H NN N mH N o.o o N HH 0 o.o mH NH m o m.N HN mH mm o m.m o NH mN H q.o NH N no 0 o.N 0N m mm o m.m m N wH o w.m HH 0 mH o m.N 0H mH mm o H.m N m NN o o.H m 0H Hm o H.N mH wH moH o N.N H o oH H H.m m mmmH «mmH mmmH ooHoEem .oz mmmH ommH mmmH oeHoEem .oz mmmH omoH mmmH ooHosem .oz mnuoo< mo neoEnz eeu< n30uo cone mnuoo< mo uooEnz eou< nsowo some mnuoon mo nonEnz eeu< nsouo eons mo unconom mo unmouem mo unmouem xeo oom xeo m.HHHm xeo eanz mmmHINmmH ensue Noaum suoeoo oxen oH enema umez nH unwneo enuoo< onnom mo nonnnz HH oHneH 38 represent sampling of the mature acorn crop dropping from asks in the fall. The mature crop usually falls following heavy frosts that occurred during my three years of observations in the last week of September, or the first two weeks of October. I made weekly collections of acorns from mast traps beginning in mid-September and continuing through October. The bulk of the crops had fallen by the 31d of two weeks following frost. I checked possible loss of acorns from my mast traps through animal activity by marking 50 acorns and placing five in each of ten traps in Roscommon County study stand I. These acorns were left in the traps throughout the fall collection period in 1953. Only two were lost during the period. I believe that loss of acorns from the mast traps through animal activity is negligible. Expressing acorn production in numbers of sound acorns produced can be misleading because acorns of different species are not the same size. I believe the number of pounds of acorns produced by' individual trees, or in oak stands, lends itself better to inter- pretation of the value of an acorn crop to animals such as deer. I weighed ten samples of each of the three species of acorns (northern red, Hill's, and white oak) collected for controlled deer feeding experiments in 1952, 1953, 195b, and 1955. The average number of sound acorns per pound as collected was as follows: Northern red oak 86 Hill '3 ask 298 White oak 15h Downs and McQuilkin (19kb) reported 166 white oak acorns per pound and 90 Eastern red oak acorns per pound. Christisen and 39 Korschgen (1955) reported 10h and 169 white oak acorns per pound from samples taken in Missouri in two different years. I could find no mention in the literature of weights of Hill's oak acorns. To estimate the total number and weight of acorns produced by trees in my study plots I computed the ground area covered by the crown area of each study tree by dividing the closed plane curve ground area into 5 foot wide parallel strips and approximating each strip by'a rectangle as described by Griffin (1936). I could then compute the proportion of the area.sampled by the mast traps. An estimate of the total number of acorns produced by each tree was computed by adjusting the number caught in.mast traps, using the figures for per cent caught, and then expanding the adjusted number caught in the area sampled by the traps to the crown area of the tree. Numbers of acorns were then converted to pounds by dividing by the average number of acorns per pound for each species. Tables 12, 13, and 1h present my computed estimates of the number of pounds of acorns produced by' the trees I studied in three of the study stands. The figures certainly concur with previously published reports pointing outrthat large variations in acorn production occur between species, years, oak stands, and individual trees in the same stand. In Roscommon County study stand II where 50 mast traps were placed at random, the estimated acorn crop production was 879 1b in 195h (590 lb northern red oak and 289 1b Hill's oak). In 1955 an estimated h6 1b were produced in this stand (33 lb northern red oak and 13 1b Hill's oak). This represents a production of 22 lb per acre in l9§h and just over one lb per acre in 1955; or an to Table 12 Number of Pounds of Acorns Produced Per Tree in Roscommon County Study Stand I 1953-1955 White Oak Hill's Oak Red Oak Tree Pounds of Acorns Tree Pounds of Acorns Tree Pounds of Acorns No. 1953 1954 1955 No. 1953 1954 1955 No. 1953 1954 1955 2 0.8 15.6 2.3 1 0.0 14.7 1.6 13 0.6 21.6 1.3 3 0.3 3.4 0.4 14 0.0 34.7 2.7 4 0.3 8.5 0.8 15 0.6 9.6 1.8 5 0.0 2.2 0.7 16 1.3 40.7 0.0 6 0.1 6.1 0.0 17 4.7 93.0 2.4 7 0.5 6.1 0.3 18 2.1 151.4 3.1 8 0.7 4.8 0.3 19 0.0 16.3 1.7 9 0.7 39.2 0.2 Total 9.3 367.3 13.0 10 0.3 8.3 1.2 Av. Per Tree 1.3 52.5 1.9 11 0.3 4.3 0.3 12 0.0 0.4 0.0 Total 3.2 98.0 5.8 Av. Per Tree 0.3 8.9 0.5 Table 13 Number of Pounds of Acorns Produced Per Tree in Oscoda County Study Stand 1953-1955 White Oak Hill's Oak Red Oak Tree Pounds of Acorns Tree Pounds of Acorns Tree Pounds of Acorns No. 1953 1954 1955 No. 1953 1954 1955 No. 1953 1954 1955 1 6.6 52.4 3.8 2 0.4 14.6 7.0 5 0.7 39.4 19.0 3 109.4 26.6 15.8 6 0.0 48.3 8.4 11 0.0 107.8 95.1 4 1.8 14.5 5.2 7 4.0 15.3 8.9 16 0.0 20.3 31.2 12 3.8 3.8 0.0 8 0.7 11.8 7.8 20 7.4 83.7 41.9 13 1.9 16.8 5.5 9 0.0 9.8 2.9 Total 8.1 251.2 187.2 Total 123.5 114.2 30.3 10 0.3 16.0 5.3 Av. Per Tree 2.0 37.8 46.8 Av. Per 14 0.2 3.0 3.0 Tree 24.7 22.8 6.1 15 0.0 7.1 4.1 17 0.3 15.0 3.0 18 0.3 4.5 3.2 19 0.4 12.4 5.2 Total 6.6 157.8 58.8 Av. Per Tree 0.6 14.3 5.3 1:2 Table 14 Number of Pounds of Acorns Produced Per Tree in Lake County Study Stand 1953-1955 White Oak Hill's Oak Red Oak Tree Pounds of Acorns Tree Pounds of Acorns Tree Pounds of Acorns No. 1953 1954 1955 No. 1953 1954 1955 No. 1953 1954 1955 5 0.3 2.6 1.5 1 0.0 16.9 3.0 13 0.0 76.0 9.4 8 2.0 10.9 4.4 2 0.0 13.7 2.7 16 0.0 11.8 4.7 11 0.8 3.8 0.4 3 0.0 4.4 0.6 20 0.0 43.4 3.0 12 0.2 4.1 2.2 4 0.0 7.4 2.0 21 0.0 7.1 9.5 15 0.0 2.0 0.4 6 0.2 1.3 0.7 22 0.0 192.9 3.1 18 1.1 4.1 0.2 7 0.0 8.5 0.9 25 0.0 2.7 1.4 19 0.0 1.0 1.0 9 0.0 34.7 5.8 Total 0.0 333.9 31.1 23 0.0 0.6 0.4 10 0.0 20.3 0.9 Av. Per Tree 0.0 55.7 5.2 24 0.2 0.6 0.3 14 0.0 13.7 0.3 Total 4.6 29.7 10.8 17 0.0 7.9 0.6 Av. Per Total 0.2 128.8 17.5 Tree 0.5 3.3 1.2 Av. Per Tree 0.02 12.9 1.8 h3 average production of 2.8 lb per tree in 195h and 0.13 lb per tree in 1955. These figures are considerably lower than production figures for trees in Roscommon County study stand I which is immediately north of a road bordering the north side of this stand. I feel it is highly' probable that 50 traps set at random are not adequate to sample acarn production on hO acres. Further data would have to be gathered using more mast traps to confirm.this probability. For the data gathered in the three stands where production of individual trees was sampled I added and averaged the production of all of the trees by species (Table 15). It is interesting to note that the average production in pounds per tree for northern red oaks was considerably higher each year than the average production of Hill's oaks, and in two out of the three years higher than white oak pro- duction. The differences were greater than could be accounted for by relative weights of the three species of acorns. A greater number of northern red oak acorns per tree was produced. Martin, EELEE: (1951) lists northern red oak first in mentioning species of oak considered of particular importance to wildlife in the northeast region of the United States. The number of the large northern red oak acorns necessary to provide a given amount of food is less than the numbers needed in other oak species producing smaller acorns. This fact undoubtedly contributes considerably to the importance of northern red oak as a.wildlife food producer. It appears that this species may also produce a larger crop more consistently. moouu NH «as mmmhu NM ¥¥ mocha ma % a.mH o.em o.n e.~ o.~n m.o m.~ e.oH e.m meta tag mecsoa omstm>< mmsn «was mmsn mmmn «man mama mean. «man mass «taxmo pom asxmo m.HH«: #xmo moanz anal - mmsn museum upsum mouse aw done now wounpoum mcuoo< mo mpcsom mo nonasz ma mHQMH mwmuo>< hS Extent of Damage to Acorn Crops by Various Factors Several factors can have a marked effect on quantity and quality' of acorns available to deer. Insects I examined all of the acorns caught in mast traps in.my four study areas in 1953, 195k, and 1955 and computed the percentage, by oak species, infested by insects. Table 16 presents these data. I also took samples from acorns that were collected in 1952, 1953, 195h, and 1955, for con- trolled deer feeding studies and computed the per cent insect infested. Table 17 presents these data. I did not differentiate between acorns completely or only partially damaged by insects because I found in feeding studies that deer would not eat any insect infested acorns. In the four years for which I have data on the per cent of acorn crops infested by insects, two of the crops were large in northern Michigan (1952 and 19510 and two were small (1953 and 1955). It appears that in years of small crops (Tables 16 and 17) insects may infest a larger percentage of the crop than in years when acorns are more abundant. Based on my samples, insects may be expected on the average to infest 20 per cent or less of a good acorn crop and from 20 to 30 per cent of a poor crop in northern Michigan. was An accurate estimate of the per cent of an acorn crop taken by birds and arborial animals such as squirrels from oak trees before a crap falls, I believe is next to impossible to obtain. The evidence of damage to acorns by specific animals can sometimes be identified. Figure 7 shows the remains of immature acorns eaten by grey squirrels (Sciurus carolinensis). Figure 8 shows immature acorns eaten by porcupines (Erethizon dorsatum dorsatum). By collecting and counting h6 mm mm ma OH «H «a Hm NH ON ma mH «a ma ma 0H «H OH NH MH «mmfi .— cunawmm douu xmo mm mm ma xmo nu xmo mm no mm xmo mmmH muommcHJNn pmmema ocuood mo unmoumm use «has: xmo m.HHHm emu cumsuuoz xmo muw£2 xwo a.dfiwm Um“ CHUSUHOZ xmo s.H~H: won cumsuuoz xmo moan: xmo m.HHH: won cumsuuoz moHoomm HH H .00 mean .00 smoomo .oo coEEoomom .oo coEEoomom wou< mmmfiummmfi manna ummz cw unwsmo mauoo< mommamnuuoomcH 0H manma 47 mm me am NH xmo ouwsz om 0H am oh xmo a.dfihm ma ma HN o xmo mop cuonuuoz mmmH dmma . mmmH NmmH muocmcH hp commemn mcuoo< mo unmouom mmmHuNmmH moflvaum magmomm noon moaaouunoo How pcuooaaoo mcuoom mo moHdEMm Scum mauoo< mowmfimn-uoomcH NH anmH he Figure 7. Inna tare acorns damaged by squir rel: L9 emu:Aell-wwHallwaywJim-.2!#Wwwww CAT. NO. 7303! STUDENT! HULK "-(hlzml C‘NTIAL SCIENTIFIC COMPANY We «HM .m w 71"..“eran '0 O ‘7 Figure 8. Immature acorns damaged by porcupines. 50 such acorn remains in mast traps a minimum figure can be computed. This I did from acorn remains collected in mast traps (Table 18). Only a rough estimate can be made however of the total number of acorns eaten whole by ruffed grouse (1322523 umbellus), or carried away by blue jays (Cyanocitta cristata), grackels (Quiscalus spp.), woodpeckers or other birds, or cashed or hurried by squirrels. By watching birds for periods Cypest and Webster (l9h8) figured they took an estimated 13.8 per cent of a crop. I attempted to count the number of acorns taken by squirrels and birds during 20 minute periods in 1953 in Roscommon County. ‘When a flock of grackles, or even 3 or b blue jays are feeding on acorns in an oak, it is impossible to watch all of them at the same time, and therefore I believe impractical to attempt to count in this way the number they are taking. Even when I watched just one squirrel I could not be at all confident that I was getting an accurate count of the acorns it took. I could not keep the animal in view at all times, and when I moved to try and locate it my'movement would sometimes send it into hiding. Animal-damaged acorns caught in mast traps gave some data on the per cent of a crop taken by arborial animals. These figures (Table 18) I believe can only be considered as representing a part of the total taken by arborial animals. I think it is safe to state that, assuming similar arborial animal population levels, in years of poor and good acorn crops the amount of damage to a poor crop can be much greater than that to a good crop. At times not all of an acorn is eaten by an arborial feeding animal and sound acorn parts are dropped. The amount of food 51 o m u .00 oxmq m m N .00 mmoomo n m . HH .oo coeaoomom m m N H .00 cosaoomom mmwfl .mmwfl .mmww mou< mamaaa< en powmama mohoo< mo udoouom mmmaumnmfi wacky ude a“ uswsmo mcuou< vowmsmn HmEHG< ma magma . 52 available to deer in this form however I believe would be insignificant compared to that provided by the sound acorns thatfall. Disease Disease, such as fungi or bacteria, can reduce the acorn crop size. The embryo and cotyledon of disease infested acorns may be partially or wholly destroyed. Diseased acorns usually have a foul smell. I do not believe deer use them.to any extent. I measured the degree of damage to acorn crops by examining all of the acorns caught in mast traps and counting the number classed as affected by disease. The number of diseased acorns in mast traps was very small. I classified very few as diseased when I examined samples of acorns collected for deer feeding experiments to deter- mine the per cent insect infested. Gysel (1956) pointed out that it is not always possible in the field to tell whether a malformation of the cotyledon or embryo is the result of disease or some other kind of physiological change. Because I found so few acorns that could be classed as diseased, I believe that in northern Michigan disease is not an import- ant descimating factor in acorn crops. Methods Investigated for Predicting Acorn Crops Gysel (1958) suggested counting flowers and immature acorns with the aid of a spotting scope to predict the potential size of a crop. He felt later observations and check counts were needed to determine the actual size. Sharp (1958) described a method of estimating acorn crops by count- ing acorns on terminal branches 2h inches in length and rating the crop as a percentage produced in relation to the potential maximum yield. To be practical for use by game managers in Michigan, an acorn crop prediction method must be one that does not require too much time or man- power. It must also be a method that will yield a prediction by mid-July, 53 for recommendations for deer hunting regulations are considered at the August Conservation Commission meeting. Observation of Oak Flowers The earliest indication of the potential acorn crop size may come at the time oaks flower in the spring. In 1953 I attempted to count the number of female oak flowers on sample stems of sample trees in three of my study stands to investigate the possibility that oaks produced different numbers of flowers in different years that might lead to an early crop prediction. Gysel (1958) counted flowers and.1mmature acorns on oaks in southern Michigan using a spotting scope with a 30 x eye piece. He found this method time consuming. I tried 7 x 35 binoculars and found them inadequate. I climbed the trees to get close enough to the twigs of annual growth that held the flowers to make my counts. Counting female oak flowers by climbing trees to get near sample twigs, I feel, is an impractical method of attempting to gather data for acorn cr0p predictions over large areas. The technique is too time consuming. I also found that flowers were lost as a result of my climb- ing activity. This would invalidate such counts. Estimating Frost Damage to Oak Flowers When killing frosts occurred after oaks were in blossom (and some did each spring during the study), I surveyed the oak forests to estimate the percentage of trees in the stands and the percentage of the crown areas of the trees that had blossoms killed by the frost. Seven to ten days following a killing frost, damaged portions of oak stands have turned brown, while unaffected portions are pale yellow-green. The percentage of the "potential" crop destroyed can be estimated by such surveys. Because many factors can reduce acorn crop size following blossoming St I believe that the only definite estimate of an acorn crop size resulting ;from frost damage surveys would be one stating complete crop failure when frost kills all of the flowers. Counts of Fertilized Female Flowers In talking with orchard owners in Michigan's Grand Traverse region I learned they predicted cherry crop size by examining flowers and figuring the percentage fertilized. I attempted to count fertilized female red oak flowers on marked sample twigs on sample trees in two of my study plots in 1953, 19514, and 1955. A twig is a single stem of annual growth bearing flowers. These counts were made by climbing five trees in each stand to examine five marked sample twigs on each tree. I thought that the ovary of fertilized flowers would increase in size enough to distinguish them from those not fertilized. My fertilized flower counts were frequently higher than maturing acorn counts made later on the sane twigs. This was apparently because I could not distinguish the difference between an oak flower that had been fertilized and one that had not. Turkel, gt_al, (1955) state that white oak flowers are pollinated in late May, with fertilization occurring during late June or early July. They also describe abortion of immature acorns caused by formation of an abscission layer between the developing ovary and the involucre. The reason for this abscission layer formation is not known. I believe that this and other'mortality factors following fertilization.may vary'from year to year to such an extent that knowing only the percentage of fertilized oak flowers has little meaning. Counts of Maturing Acorns on.Sample Twigs Through the summers of 1953, 195h, and 1955, I made counts of maturing acorns on the same selected sample twigs used for fertilized 55 flower counts in an effort to determine a mortality rate that could be applied to acorn crop prediction. My'sample consisted of five twigs on each of ten northern red oak trees. Each twig was a single stem of annual growth that held maturing acorns. I marked each twig so that I could identify it at each count. Periodic counts were made from the time the acorns began development through two growing seasons to maturity; Tables 19, 20, 21, and 22 presents these data. Through the three-year study this method provided data on two different year's northern red oak crops. Apparently loss of immature red oak acorns from trees over the two-year development period is not similar for different crops or different areas. My counts of maturing acorns on sample twigs indicated a h3.3 per cent loss of'the l95h crop in August, 1953 in Roscommon County study area I, while only 25.7 per cent of the l95h crop in Oscoda County was lost in the same month. During the 1953 growing season 52.2 per cent of the maturing acorns in Roscommon County study stand I were lost, while only 35.8 per cent of those in Oscoda County disappeared during the same period. For the 1955 crop the percentage loss during the first growing season was nearly reversed for the two areas. In Oscoda County 50 per cent, and in Roscommon County 35.5 per cent were lost. I found no acorn abortion rate through.my counts of maturing acorns on sample stems that could be applied to acorn crop prediction. A larger sample might reveal more consistent figures, but the time involved in obtaining a much larger sample would then become pro- hibitive. 56 Table 19 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1954 Red Oak Acorn Crop Through Its Two Year Develooment Period in Roscommon Countxr§tudy Stand I Tree TWig Number of Maturing Acorns No. No. July August September July August September 1953 1953 1953 1954 1954 1954 13 1 4 2 1 l l 1 2 4 2 2 2 2 1 3 3 2 2 1 1 1 4 5 2 2 2 2 2 5 1 1 1 1 l 1 14 6 6 5 4 3 3 3 7 3 0 0 0 0 0 8 3 3 3 3 2 2 9 2 2 1 1 1 0 10 1 1 1 1 1 1 16 11 2 0 0 0 0 O 12 2 2 2 O 0 0 13 l 0 0 0 0 0 14 1 1 1 1 0 0 15 3 1 l 1 1 1 17 16 3 0 0 0 0 0 17 5 0 0 0 0 O 18 7 3 3 2 2 2 19 5 5 4 4 4 3 20 6 5 4 1 1 0 18 21 6 4 3 3 3 3 22 6 4 2 2 1 1 23 8 4 4 3 2 2 24 2 2 2 1 1 0 25 1 O O O O 0 Total 90 51 43 33 29 24 Percent loss between counts 43.3 b b U1 0‘ 11.1 00 o No Total percent loss 73.3 57 Table 20 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1954 Red Oak Acorn Crop Through Its Two Year Development Period in Oscoda County Study Stand 11 Tree Twig Number of Maturing Acorns No. No. July August September July August September 1953 1953 1953 1954 1954 1954 5 1 2 2 2 2 2 1 2 4 2 l 1 1 1 3 3 2 2 0 O 0 4 5 5 4 2 2 1 5 3 2 2 1 l 1 11 6 6 3 3 3 3 3 7 7 5 4 3 3 2 8 6 6 6 5 5 4 9 6 4 4 3 2 2 10 5 4 3 3 3 3 16 11 5 4 4 2 2 1 12 5 3 3 l 1 0 13 5 2 1 1 1 0 l4 4 4 4 1 O O 15 4 1 1 l l 1 20 16 4 3 3 3 3 2 l7 4 4 2 2 2 2 18 6 6 5 3 3 2 l9 4 4 3 3 3 3 20 4 2 1 1 1 1 21 21 l O O 0 O O 22 4 4 3 2 2 1 23 3 1 1 1 1 0 24 5 4 4 2 2 2 25 4 4 4 4 4 3 Total 109 81 70 50 48 36 Percent loss between counts 25.7 10.1 18.3 H o (I) 11.0 “Total percent loss 66.9 58 Table 21 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1955 Red Oak Acorn Crop Through Its Two Year Develgpment Period in Roscommon Coungy Study_Stand I Tree Twig Number of Maturing Acorns No. No. July August September July August September 1954 1954 1954 1955 1955 1955 13 l 1 O 0 0 0 0 2 l O 0 0 0 O 3 2 2 2 O O 0 4 1 l 1 1 1 1 5 1 1 1 1 l 0 14 6 1 0 0 0 0 0 7 1 O O 0 0 0 8 1 0 O 0 0 0 9 3 2 2 2 1 1 10 2 l l l 1 1 16 ll 5 4 4 3 3 3 12 4 4 3 3 2 2 l3 1 1 1 1 1 0 l4 3 2 2 2 1 1 15 3 3 3 3 1 1 17 16 2 2 2 1 l 1 17 2 2 2 l 0 0 18 l 0 0 0 O 0 19 2 1 1 O 0 O 20 2 l 1 l l 1 18 21 l 0 O O 0 0 22 1 0 0 0 O 0 23 1 0 0 0 0 O 24 l 1 1 1 1 1 25 2 2 2 2 2 1 Total 45 30 29 23 17 14 Percent loss between counts 33.3 N o N 13.3 13.3 0" c V Total percent loss 68.9 59 Table 22 Counts of Maturing Acorns on Sample Twigs on Trees Bearing the 1955 Red Oak Acorn Crop Through Its Two Year Development Period in Oscoda County Study Stand Number of Maturing Acorns Tree Twig July August September July August September No. No. 1954 1954 1954 1955 1955 1955 5 1 1 1 0 0 O O 2 2 2 0 0 0 O 3 1 1 1 1 l 1 4 1 l l 1 1 O 5 1 1 O O O 0 11 6 2 1 O O 0 0 7 1 1 O O 0 0 8 1 1 0 O 0 0 9 1 l 1 1 l l 10 1 0 0 O 0 0 16 11 1 0 0 O 0 0 12 1 1 O O 0 0 l3 1 0 O 0 0 0 14 2 2 2 O 0 0 15 1 O O O 0 O 20 16 2 2 2 2 l 0 17 l 1 l 1 1 1 l8 3 2 2 2 1 0 19 1 1 0 0 0 0 20 1 1 1 l l 1 21 21 4 4 4 0 0 0 22 1 O 0 0 0 0 23 1 1 1 1 1 1 24 2 l 1 l 1 1 25 2 1 1 0 O 0 Total 36 27 18 11 9 6 Percent loss between counts 25.0 25.0 19.4 U1 0 0‘ m o W Total percent loss 83.3 60 Collection of Aborting Acorns in mast Traps I investigated the possibility of finding an acorn abortion rate by collecting aborting acorns in mast traps throughout their develop- ment period. The same traps used to measure acorn production in my four study stands were used for collection of aborting acorns. Periodic collections from the mast traps gave data on the periods and degree of acorn mortality caused by abortion. Table 23 presents these data. Aborted acorns collected in mast traps gave data for the complete two-year development period of the 1955 red oak crop, and for the 195k and 1955 white oak crOps. If the data for these three crops are typical it appears that a high per cent of acorns that abort do so early in their development period. I do not feel a definite con— clusion should be reached on the basis of one red oak and two white oak acorn crops, however. The number of aborted red oak acorns of the l95h acorn crop caught in mast traps in July and August, 195k (a total of 66b in the four study stands) is more than three times the number of aborted acorns of the 1955 crop caught in its same development period. This could be the result either of the fact that the 19514 crop was larger than the 1955 crop, or it might represent a higher per cent of acorns aborting late in the development period of the l95h crap. Though I made no mast trap collections in 1956 I did observe a large number of acorns on the ground in June and July that had aborted from the 1956 red oak crop in Roscommon County. I believe that loss of acorns through abortion may vary enough from year to year to prevent any one abortion rate from being applicable to acorn crap prediction. .sm>aw cum mHmuou hHoo mauoom xmo ouwsa uom .xmo to» now one woman usom can you mounwwm H «m o.¢H a.mw oo o.mH o.mm xmo ouanz Hobos omm m.o o.m H.Hm a.mm mmma H.m o.¢ m.o n.H n.0m N.¢m Nee m.mm H.0m w.¢ m.m xmo pom HmuOH wwH o.o N.N «.0m m.oq mum ¢.m o.¢ 0.0 0.0 m.no w.mH mm n.¢~ H.¢m N.H 0.0 .00 oxmq aqa m.o w.m N.Ho m.¢m moo o.m m.m 0.0 H.H “.05 m.w on a.ms o.o~ o.om H.5H .oo mmoomo 1. mNN o.o m.¢ o.mw m.m oos N.o m.~ o.o m.¢ n.0N m.mo . sump 02 u HH .00 ,0 ooEEoomom sum H.H N.N H.¢n o.- 0mm N.N m.w 0.0 m.H 0.0H o.Nm nwm 0.0m o.mm H.N n.H H .00 aoBEoomom mauoo< mmaa mmma mmma mmma manoo< mmmH mmma omma «mma qmma emma ocuoo< «mmfi enmfi mmma mmma pouuon< uoo udom wa< hash wouuon< ws< hash uoo udom ws¢ %th touuon< ws< hHSH uoo udom HmuOH HmuOH HouOH mohoom couuon< Hooch mo udoohom sowo swooa omen mchoo< pouuon< Hmuoa mo udoouom sowo swoo< mmsfi mouoom coupon< Hmuoa mo udooumm uouo cocoa omen arose: as moss-mmmu meets use: as ohmsso ecuoo< ooowoea mu enema 62 Acorn CropiPrediction Through Visual Estimates In 1952 and 1953 I looked at many oak stands throughout northern Michigan to determine the extent of the oak forests, and to pick those stands I used for intensive study. Observations I made during this time led to my attempt to predict acorn crops through.visual estimates. I noted that acorns of a given year's crop were large enough by late June of that year to be easily seen with the aid of 7 x 35 binoculars. In 1952 when there was a bumper acorn crop in northern Michigan most of the mature oaks I looked at produced acorns over a large part of their crown area. The per cent of twigs bearing acorns within.the producing portion of the crown was also high, and.the number of acorns per twig was frequently five or more. A picture of a defoliated northern red oak taken in 1952 (Figure 9) shows acorn production as it appeared on many of the mature trees I looked at that year. During the fall and throughout the winter of 1952-53 more sound acorns were available than were used by all mast eaters. Nhny acorns were still available in the spring. In 1953 the acorn crap was poor. In this year only a small part of the crown area of mature oaks produced acorns. The per cent of twigs bearing acorns within the producing portion of the crown was small, and the number of acorns per twig was rarely more than one. In 1953 acorns were eaten shortly after they dropped and none were left by winter. Based on these observations I felt it might be possible to predict an acorn crop in terms of availability of acorns to deer by looking at oaks in June and observing the per cent of crown area producing acorns, the per cent of twigs within the producing portion bearing acorns, and the average number of acorns per twig. 63 Figure 9. Maturing acorns on a defoliated northern red oak in 1952. 6h Based on my observations of one excellent and one poor acorn crop I set up a five category classification of mast crop size in terms of availability of acorns to deer by arbitrarily adding three intermediate crop classes. It facilitates understanding of the terms I use to describe the quality of the crOp and is as follows: Quality of crop Availability to deer Poor Few acorns early in the fall; none thereafter. Fair Acorns available in the fall; few or. none throughout the winter. Medium Acorns in the fall and through part of the winter; none left by spring breakup. Good Acorns in the fall and winter; sane left in spring. Excellent Mbre acorns throughout the fall, winter, and spring than all mast eaters combined, at high population levels, can consume. To convert Observations of the per cent of the crown area of trees producing acorns, the per cent of twigs bearing acorns, and the average number of acorns per twig to acorn crop quality I set up the following formula. 65 Numerical value Criterion 0* If 2 .3 Per cent of crown O-h 5-33 3h-66 67-100 producing acorns. Per cent of twigs O-h 5-33 3h-66 67-100 bearing acorns. Average number of Less than 1-2 3-4 5-6 acorns per twig. one When the numerical values for each of the three criteria are added the quality of the acorn crop is rated as follows: poor 0-2, fair 3-h, medium 5-6, good 7-8, excellent 9. The formula is based on.my observations of the excellent 1952 acorn crop when all three criteria on oaks I looked at wouLi receive a numerical value of 3, and the poor 1953 acorn crop when the numerical value for the oaks observed was zero for per cent of crown producing acorns and average number of acorns per twig, and one for per cent of twigs bearing acorns. It seemed logical thatlwhen the criteria were classed between these extremes acorn crop production would have to be in one of the three intermediate categories. In 195b, 1955, and 1956, I attempted to predict the size of the subsequent fall's acorn crop by Spot checking the oak forests throughout northern Michigan in late June and early July and applying my formula to observations. In each stand checked I looked at at least 20 mature trees 12 inches or over dbh that appeared to represent a cross-section of the mature portion of the stand. On each tree I counted the number of acorns per’twig on at least 15 twigs scattered throughout the pro- ducing portion of the crown. I made predictions for the red and.white 66 oak groups, not for oak species. The predictions for l95h are given in figures 10 and 11. The white oak and red oak acorn crop predictions differed in many areas. Acorn crop prediction for both red and white oak groups was similar throughout the oak forests in 1955 (Figure 12). In 1956 the prediction was for a poor crop for both groups throughout the oak forests with the exception of a few scattered areas too small to be indicated on a map. Each year that I predicted a crop I checked areas throughout the oak forests periodically through the following fall, winter and Spring to check the accuracy of my estimates. Before snowfall, and after the spring thaw, an estimate of acorn availability was obtained by walking through portions of oak stands and looking for acorns on the ground. During the winter when snow covered the ground deer would paw through as much as 16 inches of snow irlareas where acorns were available to feed on them. I also found that squirrels digging in the snow indicated that acorns were available. When digging activity of deer and squirrels ceases I believe very few acorns are left. I checked this by clearing the snow from several four to five foot square areas under oaks in stands where acorns had been available and looked for acorns in the cleared areas. I found acorns that had been insect infested, but no sound acorns in these cleared areas. Subsequent checks of these oak stands following the spring thaw indicated that no acorns were avail- able then. In.195h and 1955 I used the visual estimate technique on trees in the four study stands to compare the predictions with the average number of pounds of acorns produced by the trees as computed from mast trap collections. Table 2b presents this comparison. I also 67 EMMET CHARLEVOM CHEBOYGAN PRESQUEISLE Q 0 LEELANAU BENZE AVER MANIS ”E WE x'rono ANTNM GRAN :KALKAS. SAUKEEROSC MASON OSCEOLA CLARE MONT 015560 I 'RE.1ALPENA [MAW ARENAC GLAPWIN OCEANA , nccosu NEWAYGO ISABELLA am mouno .MUSKEGON This represents the bulk of the oak areas In the northern half of the Lower Peninsula. FIGURE l0. ESTIMATE OF THE I951 RED OAK IGOR! CROP - IHCIST I05! 9 68 . -' b a DO EMMET. ‘6 CHEBOYGAN PRESQUEISLE CHARLEVOIX 0 mm“, OTSEGO MONTMORENCY ALPENA scum GRAN” --==-‘-'"- AVE as: iig OSCEOLA CLARE GLADWIN _# OCEANA NEWAYGO MECOSTA ISABELLA MIDLAND ‘ wusntoon This represents the hulk of the oak areas In the northern half of the Laser Pullman. FIGURE II. ESTIMATE 0F TIE l9“ WHITE 0“ Am CROP ‘- W I”! Q '0 IIIUMMU BENZIE GRAND TRAVERSE CHARLEVOIX 69' E MMET CHEBOYGAN PRESQUE ISLE ANT RIM OTSEGO MONTMORENCY WEXFORD “I." OSCEOLA CLARE .......... ....................... ......... ....... OCEANA NEWAYGO MECOSTA ISABELLA MUSKEGON '1 ._ _' .1, ' ..- [MAW J ALPENA .75” éZALCONA ._. 5’ . A9,. 1 ARENAC FIGURE 12. ESTIMATE 0F TIIE IDES ACORN 0M - WT I955 ' 70 o.m N.H uoom noom a.mm m.m ooou uHom hwosoo ome o.H H.o uoom uHmm 0.0m a.mm coou mooo hudsoo oncomo H.o u: uoom nu w.~ u: Echoz u: HH mucsoo sossoomom o.H m.~ uoom uoom a.mm o.mH coco enHooz H hucsoo coasoomom xoo soc xoo xoo xoo xoo xmo xoo ccmum mosum mom ouH£3 com ouH£3 com ouH£3 mom . ouH£3 mmmH oopa pom mmmH ammo auoo< emmH cope pom emmH douo snoo< moosom owouo>< new ooHuoHcoum moosom ownuo>¢ wow oOHuoHcoum 33 one $3 5 nocsom Nessa uses 3 $an so couscoum monoo< mo mccsom :uHa mono snoo< mo :oHuonoum mo acmHuodsoo 4N 0..“nt 71 used the technique on each individual tree in each of the three study' stands where individual tree production was measured. I used these Observations, and the pounds of acorns produced per tree to add a figure for average number of pounds of acorns produced per tree to my classification of acorn crop size. It thus becomes: Average no. Quality of of pounds of crop acorns per tree Availability to deer Poor 0-5 Few acorns early in the fall; none thereafter. Fair 6-lO Acorns in the fall; few or none throughout the winter. Medium 11-20 Acorns through the fall and part of the winter. None left by Spring breakup. Good 21-30 Acorns through the fall and winter; some left in the spring. Excellent Over 30 More acorns throughout the fall, winter and spring than all mast eaters combined, at high pop- ulation levels can consume. EVALUATION OF PREDICTION METHODS Counts of Oak Flowers As I pointed out previously I believe attempts to count fenale oak flowers to gather data for acorn crop predictions over large areas is impractical for the technique requires too much time to use over extensive areas. Frost Damage Surveys Surveys of oak stands following frost can lead to an estimate of the per cent of the potential crop destroyed. Because other factors following flowering in oaks can reduce acorn crOp size the only definite acorn crop prediction based on frost damage surveys would be one of complete crop failure when all oak flowers are killed. Fertilized Flower Counts I found no way to tell the difference between an oak flower that had been fertilized and one that had not. Even if this were possible the technique of counting fertilized flowers to find a percentage to use in acorn crop predictions would be impractical because it would require too much time to be used over extensive areas, and I could find no rate cf loss of acorns caused by factors following fertilization that would apply to all years. Counts of Maturing Acorns on Sample Twigs and Collection of Aborting Acorns in Mast Traps Loss of acorns caused by abortion apparently is not similar for different craps or different areas. My counts of maturing acorns 72 73 on sample twigs and collection of aborting acorns in mast traps did not lead to an acorn abortion rate that could be used for early acorn crop prediction. From the data I collected it appears that a high per cent of the acorns that do abort may do so early in the development period. The data also indicate that the number of acorns that abort may be quite high compared to the sound acorns that mature (Tables 9, 10, 11, and 23). It also appears that in the two or three months immediately preceding the drOp of a.mature acorn crop few acorns abort. Observations of maturing acorn crops made in late June and early' July might therefore agree closely with the actual crop that matures. It should also be pointed out that sound acorns that abort in August and September of the year of a crap are developed to a point where they are half to three-fourths the size of mature acorns and would therefore provide considerably more food than acorns aborting earlier. Acorn Crop Prediction Through Visual Estimates In l95h and 1955 when I used the visual estimate technique to predict acorn crops in oak stands where I had subsequent.measurements of the crop my predictions of the quality of the crop and the average number of pounds of round acorns produced per tree agreed with.my classification of acorn crap size in 12 out of 1h instances. I believe that in Roscommon County study stand II the disagreement could be due to the previously mentioned probability that 50 mast traps placed at random were not enough to accurately sample, and thus compute, acorn production on to acres. Deer did frequent and dig for acorns in this stand through February, 1955, a fact that supports the prediction of a medium.crop in this area. 7h I cannot explain the prediction for a fair white oak acorn crop in the Lake County stand in l95h and the computed production of 3.3 pounds per tree which would place the crop in the poor classification. Following my predictions of the 1954, 1955, and 1956 acorn creps I checked oak stands throughout the oak forests in the fall and winter to see if my prediction of crOp quality agreed with the availability- to deer part of my classification of acorn crop size. In all oak stands I checked where my prediction was for a poor acorn crop I found that acorns were gone before snowfall. None were apparently left by winter. In areas I checked where I predicted a fair aconn crop I usually found a few sould acorns left by snowfall, but I saw very little digging through the snow by deer and squirrels in these areas during the winter. In oak stands where my predictions were for medium crOps I found plenty of sound acorns available through the fall for deer use. In these areas in winter deer did dig for acorns well into the winter when snow depths and snow crust conditions did not become severe enough to make deer stay in winter yards. In areas of predicted good acorn crops I found an abundance of acorns available throughout the fall and it appeared that a lot were left for winter use. When snow conditions did not prevent it deer dug for acorns in these areas throughout the winter, and I usually found some sound acorns still available in the spring. The only difference I could see between areas where predictions were for good and excellent crops was that in the latter areas there appeared to be quite a few more sound acorns left in the spring. In all of the stands I checked my prediction of the quality of the crop apparently agreed with the classification of the crOp in 75 terms of availability of acorns to deer. Since my acorn crop predictions in areas where I had subsequent measurements of the crop agreed with the amount produced in most instances I believe that this technique, when used in an area where the number of mature oaks is known, can provide a general estimate of the number of pounds of sound acorns that will be produced in the area. It will also, I believe, provide a fairly accurate estimate of the length of time through the fall and winter that sound acorns might be available to deer. The visual estimate technique of predicting acorn crops has been used by game managers in Michigan, with apparent success, since my- study ended, and it has been accepted as a useful game management tool. SUMMARY AND RECOMMENDATIONS Summary of Field Studies A study of the value of acorns as a winter food for deer was conducted over a three-year period at the Houghton Lake Wildlife Experiment Station in Michigan by controlled feeding of penned deer. The daily food intake of deer receiving a diet of good food was approximately h lb fresh weight per hundredweight. Deer ate approximately 1.5 lb of acorns per day per hundredweight, when they" were available. Of deer fed a basic starvation ration of jack pine and oak browse, or balsam plus a small amount of cedar, only those given a 0.5-lb supplement of acorns per hundredweight would have had a chance of surviving a 90-day winter period with a.weight loss less than the critical 30 per cent. 'White oak acorns were apparently more palatable than red oak acorns, but weight loss was not significantly' different between two groups of deer, one receiving white oak acorns, the other red oak acorns, as a supplement to a poor food browse diet. Deer entering a winter in excellent condition may be better able to resist a critical winter weight loss than animals that are in poorer, lighter condition. Acorn production of the oak forests in northern Michigan was studied from 1952 to 1956 in an attempt to find a practical method of predicting acorn crops for deer. Measurements of acorn production on individual trees using mast traps to collect samples revealed wide variations in acorn crops produced between years, between different oak stands, and between individual trees in the same year. 76 77 Investigations of factors that can reduce acorn crop size, quality, and availability to deer revealed that insects may be expected to infest from 20 to 30 per cent of small acorn crops, and 20 per cent or less of larger acorn crops in northern Michigan. Frost usually reduces the potential acorn crop size by killing some oak flowers. It may in some years greatly reduce the potential acorn crop size. Abortion of immature acorns early in their development period may considerably reduce the potential acorn.crop size. The per cent of the sound econ: crop taken by various animals is extremely difficult to accurately estimate, but it appears that in years of medium or better acorn crop production sound acorns may be available to deer through the fall and part or all of the winter. Disease is apparently not an important descimating factor in acorn crops in northern Michigan. Evaluation of methods investigated for predicting acorn crops led to these conclusions. Climbing trees and using a spotting scope to count oak flowers to get an early indication of acorn crOp size is too time consuming to be used over extensive areas. Frost damage surveys can lead to an early estimate of the per cent of the potential acorn crop destroyed. Attempts to count fertilized oak flowers to determine a per cent of the number of female flowers fertilized as an early indication of acorn crop size is impractical because of the difficulty in distinguishing between flowers that are fertilized and those that are not. None of the three above mentioned techniques would lead to an accurate early acorn crop prediction because factors reducing acorn crop size, such as acorn abortion, that follow these observations apparently vary from year to year. I found this apparently true in my attempts to figure an acorn abortion rate by counting maturing acorns on sample 78 twigs, and collecting aborting acorns in mast traps. Observations of mature oaks, made with the aid of binoculars in late June and early July of the year an acorn crop will mature, to visually estimate the per cent of the crown area of oaks producing acorns, the per cent of twigs within the producing portion of the crown bearing acorns, and the average number of acorns per twig can lead, I believe, to a general prediction of’the number of pounds of acorns that will be produced in an oak stand, and a fairly accurate estimate of acorn availability to deer through the fall and winter. This visual estimate technique of predicting acorn crops has been used with success by game managers in Michigan and.it has been accepted as a useful game management tool. Recommendations for Oak Management for Deer Based on.my study and observations, and published reports of other workers, previously cited, I compiled some recommendations for cutting practices in oak forests that would leave oaks to produce adequate quantities of acorns for deer in years of medium.or better acorn crop production. Good acorn craps may be produced in northern Michigan in two out of five years on the average. In a good.mast crop year one mature oak may produce over 100 pounds of sound acorns but the average sound acorn production per tree that reaches the ground is probably closer to 20 pounds. Only mature full-crowned oaks over 12 inches dbh and b0 years of age may be expected to produce this much. To determine the number of oak trees necessary to produce enough acorns for deer use through the fall and winter, the amount a deer 79 will eat, the portion of the crop that might be available to deer, and the number of deer using the area should be known. I found in controlled feeding experiments that deer will eat approximately one and one-half pounds of acorns daily when they are available. At this rate one deer would eat approximately 200 pounds of acorns between the middle of October when the bulk of an acorn crop has fallen and the end of February. My feeding studies indicated that a deer could survive a winter when it had one-half pound of acorns daily as a supplement to poor browse foods, so I believe it is reasonable to assume that a deer getting three times that much through February would come through the winter in fairly good condition. Only a part of the sound acorn crop that falls is available to deer. Other mast eaters take some. Based on my observations of the amount of acorns on the ground during autumn periods when acorn pro- duction per tree averaged approximately 20 pounds, I believe it is safe to assume that at least half the crop and possibly more is available to deer. But if only half this crop were available too. pounds of acorns would have to be produced to provide 200 pounds for one deer to use in fall and winter. Thus at an average production of 20 pounds per tree 20 mature oaks would be needed for each deer. Estimates of deer pOpulation are made each year by the Michigan Department of Conservation. The average number per square mile in the fall has fluctuated around 20 in the northern half of the Lower Peninsula. In areas with 20 deer per square mile it would take too oaks producing an average of 20 pounds of acorns per tree to provide an adequate amount of acorns for deer use in fall and winter. There is no way of being sure acorns will be available through 80 a winter, for weather conditions may become severe enough so deer ‘will yard in conifer swamps. I observed deer in winter'leaving deer yards and frequently traveling a half mile to dig through as much as 16 inches of snow for acorns. Leaving oaks within a half mile of deer yards will better insure acorn availability; Deer start feeding on acorns as soon as they drOp in the fall, and during this period deer are scattered through their summer range. Leaving oaks in summer range further than a.half mile from deer’yards may tend to distribute use of acorns and thus possibly prevent heavy use in the fall in areas adjacent to yards that might reduce the amount left there for winter use. Frost, by killing blossoms in the spring, often reduces the potential acorn crop size in northern Midhigan, particularly in low areas. In areas where oaks are to be cut, leaving trees in higher elevations will better insure that there could be good acorn production. A mixture of red and white oak groups will better insure a crop each year for one spring frost will not affect the same year's crop on both oak groups. Deer prefer white oak acorns over those of the red oak group and will eat them first. White oak acorns usually sprout shortly" after they drop in the fall and thus are less available through the winter than red oak acorns that do not sprout until the following spring. By leaving a higher ratio of white oaks to red oaks more than a half mile from deer yards, than in areas near yards, it is possible that deer might tend to use the acorns away from wintering areas first and thus better insure that acorns might be left in areas where they could be available in the winter. 81 A summary of my suggestions for management of oak for deer in areas where oak is to be cut follows: Leave a total of 20 mature full-crowned oaks over to years of age and 12 inches dbh for each deer using the area. Leave ten of the oaks within one-half mile of deer yards and ten in summer deer range further than a half-mile from deer yards. Leave oaks at a rate of three red to one white oak near yards and two red to one white oak away from winter range. When there are variations in elevations in an area where oak is to be cut, leave the oaks in higher elevations. I believe that in years of good mast crOp production in areas where there are as many oaks as I recommend, enough acorns should be produced to provide deer with acorns as food through the fall and most of the winter. My recommendations have been incorporated into the policy of the Game Division of the Michigan Conservation Department on timber cutting practices. They have been accepted and are being followed by State and Federal Foresters in Michigan. LITERATURE CITED Allen, Durward L. l9h3. Michigan fox squirrel management. Nfich. Dept. Cons. Game Div. Pub. 100, 1:01; pp. and Howard D. MCGinley; 19h7. A method for the ar-to-year measurement of mast yields. J.'Wildl. Mgmt.,ll (2 :18h-l85. Bartlett, I. H. 1951. Michigan winter deer starvation. Game Div. Rep. No. 1117, Mich. Dept. Cons., 3 pp. (mimeo.). Christisen, D. M. and L. J. Korschgen. 1955. Acorn yields and wildlife usage in Missouri. Trans. 20th N. A. Wildl. Conf., 337-357. Cochran, William G. 1957. Analysis of covariance: its nature and uses. Biometrics, 13 (3):26l-28l. Cypert, E. and B. S. Webster. 19h8. Yield and use by wildlife of acorns of water and willow oaks. J. Wildl. Mgmt., 12 (3):227-231. Dahlberg, Burton L. and Ralph C. Guettinger. 1956. The white-tailed deer in Wisconsin. ‘Wisc. Cons. Dept., Game Mgmt. Div. Tech. Wildl. Bull. No. 1h, 282 pp. Dalke, Paul D. 1953. Yields of seeds and mast in second growth hardwood forests, south-central Missouri. J. Wildl. Mgmt., 17 (3):378-380. Davenport, L. A. 1937. Find deer have marked food preferences. MiCh. ConSo, 7 (b):h‘6o . l9h6. Cusino browse survey, 1936-h1. Mich. Dept. Cons., Unpubl. Game Div. Rep. Downs, Albert A. 19h9. Trees and food from acorns. U. S. Dept. Agr. Pub. Trees. The yearbook of Agr., 571-573. and W} E. McQuilkin. l9hh. Seed production of southern Appalachian oaks. Jo Fore, 1‘23913-920. Duvendeck, Jerry P. 1957. Oak management and acorn crap forecast for deer in Michigan's lower peninsula. Game Div. Rep. No. 2156, Mich. Dept. Cons., 5 pp. (mimeo.). Easley, L. T. and L. E. Chaiken. 1951. An expendable seed trap. J. For. h9z652-653. 82 83 Forbes, E. B., L. F. Marcy, A. LeRoy Voris, and C. E. Franch. 19hl. The digestive capacities of the white-tailed deer. J. Wildl. Mgmt., 5 (1):108-11h. Griffin, Frank L. 1936. An introduction to mathematical analysis. Houghton Mifflin Co., Boston, Mass., x + 5L6 pp. Gysel, Leslie W. 1956. Measurement of acorn crops. For. Sci., 2 (h):305-313. . 1958. Prediction of acorn crops. For. Sci., h (3):239-2h5. Kittredge, Joseph and A. K. Chittenden. 1929. Oak forests of northern Michigan. Mich. Agr. Exp. Sta. Spec. Bull. 190, 117 pp. Martin, A. 0., Herbert S. Zim and Arnold L. Nelson. 1951. American wildlife and plants. McGraw Hill Book Co., Inc., New York, ix *- 500 pp. Maynard, L. A., Gardiner Bump, Robert Darrow, and J. C. Woodward. 1935. Food preferences and requirements of the white-tailed deer in New York State. New York Cons. Dept. and New York State Coll. Agr., Bull. No. 1, 35 pp. Nichol, A. A. 1938. Experimental feeding of deer. Univ. Ariz., Agr. Exp. Sta., Tech. Bull. No. 75, 39 pp. Petrides, George A. 19141. Observations on the relative importance of winter deer browse species in central New York. J. Wildl. Mgmt. , 5 (h):hl6-h22. , P. Parmalee, and J. E. Wood. 1953. Acorn production in East Texas. J. Wildl ngnt., 17 (3):380-382. Reid, Vincent H. and Phil D. Goodrum. 1957. Factors influencing the yield and wildlife use of acorns. Mimeo. paper presented at the Sixth La. State Univ. For. Sym., Baton Rouge, La., 143 pp. Sharp, Ward M. 1958. Evaluating mast yields in the oaks. Pa. State Univ. Agr. Exp. Sta. Bull. 635, 22 pp. Snedecor, George W. 19116. Statistical methods. Iowa State Coll. Press, hth Edition, xiii + has pp. Turkel, H. S., A. L. Rebuck, and A. R. Grove, Jr. 1955. Floral morphology of white oak. Pa. State Univ. Agr. Exp. Sta. 3111. 593,11; pp. Verme, L. J. 1953. Production and utilization of acorns in Clinton County, Michigan. Unpub. M. S. Thesis. Mich. State Univ. , 77 pp. Wills, H. Merrill. 19h1. Climate of Michigan. U. S. Dept. Agr. Pub. Climate and Man. The yearbook of Agr., 9lh-92h. "‘MMMMTT