.f 15“}: .L .7111}? :L I ,9; .4‘ “ I~ “W. h. . l ‘5 . t R. 55 . . \ “3 233: i... n v .. 1 m e aunt: we i; 2 c 5.155.. .. C .1 {affitifrflfé 45.2.. $.25? ‘erafitcizzlz If; . , e x :77: ‘ . ‘ ‘ , ‘ l 5...... . . . , C L — »v. I V f. ‘1; 4 l . x. 7:13;. wt ...T .5} 2.5. 4': 11.0.5}, I. ((1).. re (A?! (5 5a.... «I. . y ‘ . 2 £51. F37. £7; . » S),.D\I(l.(‘rau lllllllllllllllllllllllllllllllllllllllllllllllllllllllll 3 1293 007864 LIBRARY Michigan State University K , This is to certify that the thesis entitled ELK, DEER, AND SMALL MAMMAL RESPONSES TO THINNING RED PINE presented by Louis C. Bender has been accepted towards fulfillment of the requirements for M.S. Fisheries and degree In W Bio logy Major professor/If (2L uflfyfq/QQWZQ/c / Date 9/73 7f? 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution W ‘Y' L) 1:)“ 6.} A‘ PLACE IN RETURN BOX to rem ave this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE "1&42‘” " rm AAY 0 7 2012 h 0430 fé‘ MSU Is An Affirmative Action/Equal Opportunity Institution c:\clrc\dmd\n.nln3-p.' EIK, DEER, AND SMALL MRMMAL RESPONSES TO THINNING RED PINE Louis C. Bender A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1990 c05537o AETRACI' ELK, DEER, ANDSMAILMAMMALRESPONSES 'IO'IHINNINGREDPINE By Iouisc. Bender MatureredpinestarrlswerethinnedtoZdensitiestoassessthe effects of cverstory thirming on wildlife habitat attributes and utilization of red pine. Measured understory variables did not differ among thinning treatments or controls, although trends of increased woody stem densities were apparent. Overstory thinning had little effect on ungulate forage nutritional quality, except for an increase in forage moisture contents. Small mammal populations tended to increase with overstory thinning, but evaluation of small mammal data was complicated by population fluctuations not attributable to treatment. Deer and elk use stratified into 3 distinct levels of utilization; 16.1 mZ/ha thins > 25.3 mZ/ha thins > controls. Increased ungulate use is likely attributable to (1) increased forage quantity and quality and (2) increased hiding/security cover attributes of the thinned stands. 'Iheresultsofthisstudysuggestthatcverstorythinningcanbean effective means of enhancing wildlife habitat and increasing wildlife use of redpine. AW This project was funded entirely by McIntire—Stennis Federal funds. I would like to thank my major professor, Dr. Jonathan B. Haufler, for his advice and support throughcxrt this project. Appreciation is also given to the other members of my committee, Dr. Glen mdderar and Dr. Don Dickmann, for their contributions to this project. Special thanks is given to all the interns and lab assistants who slaved so diligently on all aspects of this project. Thanks also to the many graduate students who provided field assistance, including Jim Hirsch, Jerome Leonard, Petrus Gunarso, and especially Mark Otten, who probably put as much field time into this project as his own. My deepest appreciation is also due Jim Hirsch, Mark Otten, Jerome Ieonard, Paul Padding, and Chen Jian for their friendship and support; you guys made all those long nights in the office almost tolerable. To Jim, Mark, and Jerome in particular I owe too much to list; suffice it to say, it was a pleasure and a privilege knowing you gentlemen. My life is immeasurably richer for the experience. Many things went into this project, not the least of this was my undergraduate preparation. IIhanks to all the faculty at the University of Montana, especially Dr. P.L. Wright, who taught me, by example, to never put any limitations on just how much one man can learn. A special thanks also to all my friends at UM too, especially Heavy, Webb, Big ii Bob, the great Laye, and Spock of Calgary—this thesis is as much a product of you guys as me. My deepest appreciation also goes to a special person who died a long time ago and whom I never met. It was 3.1;. who drove me to science, and urged rte to seek origins, both of man and tine. Although this work falls short of those lofty goals, tine has taught He that if one would catpreheni the beginnings, he must first understand the present. Finally, the biggest thanks of all, to my parents and family. Without them, none of this would have been possible. I thank them all for their support, and instilling in me the desire to succeed. These things made Ire what I am, and in turn made this thesis. Thank you all. 'IABIEOFwNI'ENI’S Egg LISTOFTABIES....... ......... .......... vi LISTOF FIGURES" ............ ....... ......viii mmowcmom... ........... ......... 1 OBJECTIVES ........................ ..... 8 STUDY AREA DESCRIPTION ..... .. ................................ ...... 9 WILDLIFERESPONSE‘S ............. ................... 23 Elk and Deer....... ...................................... 23 Small Mammals................... .................... ..... 24 UREA ANAIMSIS.. ................ ........... ........... . ........ 25 VEGEIHTIVE COMPOSITION ............................ ...... ...... 27 WOody Densities ...... . ................................... 28 Herbaceous Frequencies.............. ....... . ........... .. 32 SMAII.MAMMAIS ..................................... ...... ...... 34 BIG GAME USE.............. ..... . .............................. 40 Pellet-Group Counts .......... . ............. ...... ........ 40 Browse Utilization...... ..... . ........................... 45 NUTRITIONAI.ANAIXSES... ...... . ...... .... ...................... 45 Moisture Content ........................... ... ........... 45 Ash Content .................................. .... ........ 54 Ether Extract..... .................... ........ ........... 57 Crude Protein.......................... ............ ...... 60 DISw$IwIOOOOOOIDIO ......... .0... ................................ 64 VEGEIHTIVE RESPONSES ....... ......... ......................... . 64 NUTRITIONAL ANALYSES.... ....................... . .............. 67 crude Protein .................. ...... ................. ... 68 Moisture Cbntent........... ...... ......... ............... 72 Ether Extract ..... ........... ......... . ..... .. ........... 74 Ash Cbntent.................................... .......... 75 SMAII.MAMMAIS ............... ..... ..... ........ ................ 77 iv ELKANDDEEIRRESPONSES..... ..... ..... ..... ............ 81 SUMMARYANDWW8...................... .................. 87 IITERATURECITED ......................... ... 91 APPENDH.................................................... ....... 98 Table 10 11 LIS'I‘OF'I‘ABLES Page Site and stand characteristics of the 18 red pine studyplotsinthePRcsr", MI ............. .. ...... 17 Mean (SE) densities of all woody stems present on pretreatment 25.3 mZ/ha (110 ftZ/ac) and control plotsinJuly, 1987.............. ................ . ...... 29 Mean (SE) densities of all woody stems greater than 1m tall and less than 10.2cm dbh in 1988 ................ 31 Mean (SE) densities of woody stems greater than 1m in height and less than 10.2cm dbh between the differentially aged 16.1 mZ/ha (70 ftZ/ac) study plots in l988....... .................................... 33 Mean (SE) minimum populations, species richness, and small mammal diversity between pretreatment 25.3 mZ/ha (110 ftz/ac) plots and controls in 1987 ........... 35 Mean (SE) minimum populations, species richness, and small mammal diversity among all thinning treatments and controls in 1988 ............... . .................... 37 Mean (SE) minimum populations, species richness, and small mammal diversity between the two age classes of 16.1 mz/ha (7o ftZ/ac) thins in 1988 ................. 39 Mean (SE) percent moisture content in forage samples collected in Winter and Smeer 1988 among all treatment combinations ..... . .............. . . ............ 52 Mean (SE) percent moisture content in forage samples collected in Winter and Summer 1988 from the differentially aged 16.1 mZ/ha (7o ftZ/ac) thins ........ 53 Mean (SE) percent ash (total minerals) in forage samples collected in Winter and Summer 1988 among all treatment combinations. ........ ....... ........ 55 Mean (SE) percent ash (total minerals) in forage samples collected in Winter and Summer 1988 from the differentially aged 16.1 mZ/ha (7o ftZ/ac) thins ........ 56 vi Table 12 13 14 15 Arl Page Mean (SE) percent ether extract (crude fat) in forage samples collected in Winter and Summer 1988 among all treattent combinatlons ...... 58 Mean (SE) percent ether extract (crude fat) in forage samples collected in Winter and Smeer 1988 from the differentially aged 16.1 mZ/ha (7o ftZ/ac) thins ............................ ....................... 59 Mean (SE) percent crude protein in forage samples collected in Winter and Simmer 1988 among all Mean (SE) percent crude protein in forage samples collected in Winter and Smmer 1988 from the differentially aged 16.1 mZ/ha (7o ftZ/ac) thins ........ 62 Species of vegetation found on the PRCSF study plots in 1987 and 1988 ........................................ 98 Absolute (AF) and relative (RF) frequencies of common herbaceous vegetation found on the pretreatrent 25.3 m2/ha (110 ftz/ac) and control plots in 1987.. ......................... . .................. ..... 102 Absolute (AF) and relative frequencies (RF) of common herbaceous vegetation found on the PRCSF study plotsduring1988............. ......................... 104 Absolute (AF) and relative frequencies (RF) of common herbaceous vegetation found on the differentially aged 16.1 mz/ha (7o ftZ/ac) thins during 1988... ....... 106 figure LIST OF FIGURES Page Location of the Pigeon River Country State Forest inMichigan.. ..... ................... 11 Mean monthly long term precipitation and mean monthly precipitation during the study period for Vanderbilt, MI.......... ..... 13 Mean monthly long term temperatures and mean monthly temperatures during the study period for Vanderbilt, MI. . . ....... . ................... . ........... 15 Mean (SE) species specific and total pellet-group counts per thinning (mZ/ha) treatment for Spring 1988 .................................... . ............... 42 Mean (SE) species specific and total pellet-group counts per thinning (mZ/ha) treatment for Spring l989.......................... ............. ... .......... 44 Mean (SE) species specific and total pellet-group counts per study plot for the two ages of 16.1 nP/ha (7o ftZ/ac) thins for 1988 and 1989 ....... . ....... 47 Mean (SE) percent utilization of selected ungulate forages for all thinning (mZ/ha) treatments for the PRCSF study plots in Spring 1988. ............... 49 Mean (SE) percent utilization of selected ungulate forages for the two age classes of 16.1 mz/ha (7o ftZ/ac) thins in Spring 1988 ...... . ............ ..... 51 viii Red pire (m resinosa) is the most intensively managed conifer type in the lake States (Benzie 1973) , currently covering approximately 567,000 ha (1.4 million acres) (Dickmann et a1. 1987). Red pine is an extremely valued timber species because it is long lived, comparatively free of insects and disease, rapid growing, and produces straight high quality timber (Eyre and Zehngraff 1948) . Additionally, red pine's rich color, attractive form, vigorous growth, and transplanting ease make it popular for ornamental plantings (Collingwood and Brush 1974) as well as plantations (Benzie and W 1983) . These combinations of characteristics make red pine the best silvicultural conifer in the Lake States (Eyre and Zehngraff 1948) . Red pine historically covered a much greater area than it presently does; up to 2.8-3.0 million ha (7-8 million acres) a century ago (Benzie 1977). Red pine grows in both pure and mixed stands (Fwells 1965), tending to be associated with jack pine (P. banksiana), aspen (Populus spp), white birch (Bet—mil mifera), and scrub oaks (mercus spp) on drier sites, and white pine (P. strobus), red maple (Air 2.13.111“), red oak (Q. rubra), balsam fir (Abies balsamea), and white spruce (Pioea g1_agc_a) on moister sites (Benzie 1977) . Although occurring on a variety of sites, red pine grows best on moist, well drained acidic sandy or loamy soils (Fowells 1965), such as Entisols, Spodosols, and Alfisols (Benzie and McCumber 1983). Red pine is shade intolerant, regenerating 1 2 best on mineral soils under an even-aged management scheme (Benzie and member 1983) . Mature unthirmed red pire stands are typically close-canopied, dense, and have little understory development (Fowells 1965) . Gysel (1966) found only 4.2% of the total ground area covered by woody vegetation less than 1.8m (5.9 ft) high under a mature red pine stand in Michigan. Hardwood seedling survival under a mature red pine overstory was almost nil in New York (Tobiessen and Werner 1980) , attributed to a lack of endomycorrbizal colonization resulting in insufficient phosphorous uptake. Kennedy and Wilson (1971) also noted the lack of undergrowth associated with dense, mature red pine stands, as did Dickrmann et a1. (1987) and Van Wagner (1963). The lack of red pine understory developrent compared to other plantation pines has been described by 'I‘appeiner and Alm (1975) and Grisez (1968) . Site quality strongly affects understory development in red pine stands, however; good sites tend to show greater development than do poor (D. Dickmann, pers. comm.) Due to this lack of understory development and resulting lack of structural diversity and wildlife food species, red pine stands are generally considered poor habitat for wildlife (Benzie 1977) . Gysel (1966) noted that animal and bird species camon in northern Michigan were scarce in red pine plantations. White-tailed deer (Odocoileus vig' inianus) numbers were approximately one-third that of surrounding areas, and while small mammals commonly famd in northern Michigan were found in the plantations, their populations were small. Ross et a1. (1970), however, reported a deer density high enough that individuals 3 were frequently observed and produced significant vegetative differences betveencpenareasandexcloslresinamatureredpinestandin Minnesota. Hare (% spp) browsing, waver, was almost absent in the same stand. Benzie and McCumber (1983) also reported that red pine stands were poor habitat for game birds and mammals. Mature red pine stands do provide habitat for species such as red squirrels (Tamiasciurus hudsonicus) and pine martens (m americana) , as well as same songbirds (Benzie and McCumber 1983). Additionally, menagingredpirestardsrearthemdnimmnrecamendedstodtinglevels (13.8-23 mZ/ha (60-100 ftZ/acre) basal area), providing openings, and prescribed burning may help improve wildlife habitat quality by providing a more favorable understory (Benzie 1977, Dickmann et al. 1987). Under current high density management operations (32.1+ m2/ha (140+ ftZ/acre)), however, red pine stands are of little value to most wildlife species (Benzie and member 1983) . In order to meet projected softwood timber demanis, it has been recommended that red pine be restored on up to 4.0 million ha (10 million acres) of former pine forest types in the lake States (Benzie 1977). Red pine stands are also being forced onto sites capable of supporting northern hardwoods or other deciduous species, a practice both economically and ecologically questionable. Increased empesis is concurrently being placed on multiple uses of timberlands, however, with values such as recreation and wildlife becoming increasingly important on private-lands as well as legislatively mandated on public lands (Thomas 1979, Dickmann et al. 1987). As previously noted, red pine stands can present a conflict between timber production interests and 4 wildlife habitat quality. It is therefore important that practical methods of improving red pine stands as wildlife habitat be identified and evaluated to satisfy the diverse demaros for wildlife. However, any method of improving red pine as wildlife habitat should ideally cause minimal reductions in red pine timber production. Only wildlife enharoing activities which meet these prerequisites are likely to be adopted by land managers, who face the dilemma of having to maintain timber-generated revenues while silmlltaneously incorporating multiple- use demands into their management objectives. ‘Ihirmingarriprescribedburningofredpinestandshavelongbeen established silvicultural practices (EYre and Zehngraff 1948, Fowells 1965, Lundgren 1983). Von Althen et al. (1978) found that stands thinned from 55.1 mZ/ha (24o ftz/acre) to 23—32.l mZ/ha (100-140 ftZ/acre) produced greater quantities and higher quality timber than unthinned stands under 3 different management strategies; pulp, sawlogs and pulp, and poles, sawlogs, and pulp. Economically, thinned stands were either equal to or produced greater returns than did unthinned. Iothner and Bradley (1984) found that Wifl'l an initial stocking density of 988 trees/ha (400/acre) stands kept thinned to 18.4 mz/ha (80 ftz/acre) showed the highest financial returns. Other research noted thatredpinestandsthinnedtoaconstantbasalareaof27.6m2/ha (120 ftz/acre) every 10 years produced the maximum merchantable volume per acre over a wide range of sites (Lurxigren 1983). Additionally, Benzie (1977) noted that thinned stands showed the most rapid diameter growth, and that thinning allows the removal of smaller, slower growing, and poor quality or damaged trees in favor of larger high quality crop trees. 'mequalityoftimberproducedfronthimedstandscanalsobe maintained (Lothrer and Bradley 1984), as red pine prunes itself naturally better than any other native conifer of the lake States (Eyre and Zehngraff 1948). Benzie (1977), however, indicated that stands managed near the minimum recommended stocking densities may need artificial pruning to produce high quality knot-free lumber. Prescribedburningisalsoconductedinredpinestands, bothto prepare a mineral seedbed for regeneration (Benzie and Mcctnmber 1983) and to control understory shrubs to allow regeneration (Rickman 1964, Alban 1977). In general, mature red pine stands are very resistant to fire (Van Wagner 1970) , and probably depended on fire historically for regeneration (Collingwood and Brush 1974). Additionally, Alban (1977) noted that prescribed burning had beneficial effects on soil fertilization under red pine in Minnesota. Although thinning and prescribed burning are current management practices for red pine, little consideration has been given to the effects of these treatments on understory corposition and structure, especially as this relates to wildlife. Benzie and mm (1983) speculated that managing red pine at minimum recommended stocking levels would favor a greater diversity of understory species and increase the supply of wildlife foods, thereby improving the stands for wildlife. Alban (1977) found that spring burns and periodic summer burns increased themmberofwoodystemsmfleramatureredpirestarxi. Hazel (Coglus spp) stems have also been found to resprout vigorously after their aerial shoots had been killed by a spring burn (Buckman 1964). Anderson 6 et al. (1969) found that tmderstory herbaceon cover in a red and white pirestaniirrzreasedasthedensityofthepineoverstorydecreased. Dickmann et al. (1987) also described increased understory development in thinned red pire stands in northern Michigan, additionally reporting a high degree of wildlife use of these thinned stands. None of these studies, however, looked specifically at the effects of thinning and/or burning red pine stands on wildlife, or on the developtent of an understory favorable to wildlife. However, thinning and prescribed burning of mature pine stands has long been recognized as an effective means of improving wildlife conditions in southern pine forests (Lay 1956, 1957; Dills 1970, Halls 1973, Wolters and Schmidtling 1975, Wolters et al. 1982) and western pine forests (Reynolds 1969, Ieege and Hickey 1971, Kruse 1972, Lowe et al. 1978, Roppe and Hein 1978, Keay and Peek 1980, Oswald and Covington 1983, Roberts and Tiller 1985, Crouch 1986). AS these treatments are current silvicultural practices for red pine, their potential to increase the quality of red pine stands as wildlife habitat by providing greater urxierstory development and productivity while minimally impacting current management operations is significant. As noted above, thinning and prescribed burning have the potential to greatly increase the quality of wildlife habitat associated with red pine (Benzie 1977, Dickmann et al. 1987) while maintaining current levels of timber production and its associated revenues (von Althen et al. 1978, Dindgren 1983, Lothner and Bradley 1984). Thus, a combination of thinning and prescribed burning appears to represent the most feasible and minimally impacting means of enhancing wildlife habitats in red pine types. As 7 little is known of the responses of wildlife to these managorent operations in the red pine type, however, it is imperative that the beneficial impacts of these treatments be doomented before any land manager adopts, or indeed should be expected to adopt, these management activities as a means of enhancing the wildlife resource associated with red pine. OBJECTIVES The objective of this study is to evaluate the responses of elk (m elapgus), deer, and small mammals to thinning andburning of red pine stands, aswellastoevaluatetheeffects ofvarioisthii'mingard prescribed burning treatments on understory developxent in red pine stands. Specifically, the objectives of this study are: (1) Determine differences in elk and white-tailed deer use of red pine stands managed by thinning and controlled burning versus unthinned, unburned starris. (2) Determine the effects of various thinning and prescribed burning treatmems on small mammal populations in mature red pine stands. (3) Determine the influences of various thinning and prescribed burning treatments on understory cotposition and nutritional quality in mature red pine stands. S'IUDY SITE DESCRIPTION This study was corducted within the 33,590 ha Pigeon River Country State Forest (PRCSF) , located in the rorth—central lower peninsula of Michigan. The PRCSF is approximately 21 km east of Vanderbilt, MI, and occupies portions of dueboygan, Montmorency, and Otsego comties (Fig. 1). The PRCSF lies within the Presque Isle Rolling Plain, Emmet-Alcon Hill land, and the Huron lake-Border Plain physiographic regions (Sommers 1977). The watershed is drained by the Black, Pigeon, and Sturgeon Rivers, which originate in the coniferous swamps on the southern edge of the forest and flow northward. Soil types in the PRCSF include swampy highly fertile soils, medium-high fertility soils on till plains and moraines, and dry sandy soils on outwash plains (Moran 1973) . 'Ihe climate alternates between continental-type and semi-marine (Moran 1973). large daily, monthly, and seasonal temperature changes are typical. Fall and winter temperatures are moderated by lake-effect cloud cover and prevailing westerly winds. long term mean monthly precipitation and mean monthly precipitation during the study period are shown in Figure 2 (NCAA 1987, 1988) . long term mean monthly temperature andmeanmonthlytemperaturesduringthestudyperiodareshownin Figure 3 (NCAA 1987, 1988) Diversity in soil types, drainage, and exposure results in a 9 10 Figure 1 . location of the Pigeon River Country State Forest in Michigan. 11 oEncoo «now «no.5... 33m 3:300 helm coon—n. Figure 2 . Mean monthly long term precipitation and mean monthly precipitation during the study period for Vanderbilt, MI. l3 _ new.“ home anon anus need .+ la la.— la.— INu 1n.— 1: the (mo) uonqudloaid Klqguow neon 14 Figure 3. Mean monthly long term temperature and mean monthly temperatures during the study period for Vanderbilt, Ml. 15 eZ _O 4/) -<1 —2 r—F‘h ovum U Emu." 4. Head E3 used I (3 3 aameaadmam Altnuom mean l6 variety of vegetation types within the PRCSF. This natural diversity is further enhanced by logging, burning, plantations, and farming (Beyer 1987) . 'Ihe vegetative types present can be grouped into 5 categories (Spiegel et al. 1963): (1) riverbanks and bottomlands, (2) sandy outwash plains, (3) outwash plain-morainic ecotone, (4) steep morainic slopes, and (5) morainic uplands. Moran (1973) added coniferous swamps as an additional category. Vegetation on the study plots was dominated by dense 60-year-old stands of red pine. Occasional live and standing dead jack pine were interspersed in the canopy. Understory vegetation consisted of scattered stems of deciduous shrubs such as black cherry (Ms serotina), chokecherry (P. viginiana), red maple, serviceberry (Amelanchier spp), and beech (m gra_ndifolia). Ground cover was primarily absent. Soils on the study plots were high quality Emmet sandy loams, with site indexes ranging from 60 to 76 (N. Caveney pers. comm.) . Stand and site characteristies for the study plots are given in Table 1. 17 Table 1. Site and stand characteristics of the 18 red pine study plots in the Pigeon River Country State Forest, Ml. GROUP No. li'll'JI’Sl Year Year Site Basal Area PIDT 16.1 25.3 control Planted Thinned Index (HR/ha) Big 70 6 — — 1930 76* NB7OS x 1987 18.1 (3.0) NB7OE x 1987 18.0 (3.3) NB70W x 1987 18.7 (3.3) SB7OE x 1987 16.3 (4.0) SB7OS x 1987 17.3 (3.0) SB70N x 1987 18.3 (1.7) Small 70 3 - - 1931 61 S705 x 1986 20.0 (1.1) S70N x 1986 23.4 (3.1) S7OE x 1986 20.3 (3.6) Pine Ridge - 2 1 1931 60 PRC x -- 34.4 (3.4) rem x 1987 21.2 (4.7) PR8 x 1987 24.6 (4.3) Hill Prairie - 2 1 1932 66 HPC x -- 36.9 (7.0) HPS x 1987 20.1 (5.4) HPW x 1987 22.7 (2.3) ccc - - 1 1930 60 ccc x -- 33.1 (5.0) Milkjug - 1 — 1931 70 M3 x 1987 26.3 (5.7) County Line - 1 - 1928 62 CL x 1987 25.5 (3.6) * Actual site index much more variable within cut. 1 Number of plots per basal area (mZ/ha) treatment. MEI‘HOIE Eighteen 2 ha (5 ac) study plots were established in mature red pine stands of uniform site daracteristies, age, and stand corposition. 'Ihe18 plotsweredividedinto3experimental treatmentgroupsbasedon overstory basal area in the following proportions: 9 plots at 13.8-16.1 mZ/ha (60—70 ftZ/acre) , 6 plots at 23-25.3 mZ/ha (100-110 rtZ/acre), and 3 plots at 32.1+ mZ/ha (140+ ftz/acre). Thinning of plots to the indicated basal areas was done by the Forestry Division of the Michigan Department of Natural Resouroes under contract logging. Interval since thinning was to be kept constant for all 15 thinned study plots. Due to the nature of the thinning operations (commercial sales), however, three 16.1 mz/ha (70 ftZ/acre) plots were thinned approximately 9—12 months prior to the other 12 study plots. Since these plots had an additional season's growth, theywereusedtoassessany temporaltrendsthatmay be manifested with advancing time since thinning, and were each allocated to a different burn treatment. A firebreak was established around 6 of the experimental plots, but due to drought conditions in 1988, firebreaks were not corpleted around the rest of the study plots. 'Ihree plots thinned to 25.3 mZ/ha (110 ftZ/acre) and 16.1 mZ/ha (70 ftz/acre) respectively were randomly selected for prescribed burning on a 5 year rotation, with 3 additional 16.1 mz/ha (70 ftz/acre) plots randomly selected for a 10 year burn. The remaining 3 plots at 16.1 mz/ha (70 ftZ/acre) and 25.3 m2/ha (110 ftz/acre) served as controls for 18 19 the burn treatment. None of the unthinned plots (32.1+ mZ/ha (140+ ftz/acre)) were to be burned; they functioned as thinning and burn controls. Prescribed burns were to be conducted in Spring of 1988. The timingarrlintersityofmrningwastobekeptasconstantaspossible within the 3 replicates per treatment and among all treatments. Die to the drought conditions of 1988 (Fig. 2), however, the prescribed burning aspectofthisstudywasunabletobecotpletedinl988. Dietothe large number of wildfires throlghout the Spring of 1988 and extremely low fuel moisture contents, the risks of burning were considered excessive . Vegetative Sampling Growth rates and productivity of the overstory red pine under the various basal area and burning treatments will be monitored by the Department of Forestry, Michigan State University, East lensing, MI. Understory responses were monitored as part of this study. Overstory basal area prior to and after thinning was determined using a 10-factor basal area prism. 'I‘ransects were randomly placed throighthestudyplotsardbasalareareadingsweretakeneveryZOm. Only live overstory pine (red and an occasional jack) basal area was recorded. Basal area was determined by tallying all overlapping trees as viewed throigh the prism; exactly aligning trees were counted as one-half. Random quadrat sampling was employed to determine frequencies of herbaceois species and densities of woody species. Quadrat sizes were 20 Zn X 5m for herbaceous species and 2m X 30m for woody species. Nested quadrats were randomly placed in the study plots by establishing a coordinate system with 2 adjacent sides of each plot serving as coordinate axes. Pairs of random numbers were then selected and served as coordinates for quadrat locations. Frequelcies of all herbaceous species were tallied within the 2m X 5m quadrats. Baseline sampling in 1987 recorded stem densities of all woody species in the 2m X 30111 quadrats regardless of size. In 1988, only delsities of woody species greater than 1m in height and less than 10.2cm dbh were recorded in order to document the development of an understory considered available to ungulates. A 5m buffer plot around the edge of each study plot was not sampled to elimirate edge effects. Annual productivity for herbaceous and woody species was to be determined using the clip and weigh technique described by chisel and Lyon (1980) . However, due to the low densities and productivities observed on the control plots, as well as certain 25.3 mZ/ha (110 ftZ/acre) plots, it was decided that statistically adequate productivity sampling would result in the reloval of a disproportionately large amount of the annual productivity from these study plots, thus possibly biasing other vegetative measures in both the short and long term by severely decreasing plant vigor. Therefore, it was decided not to conduct productivity estimates in either 1987 or 1988. 21 NUTRITIONAL ANALYSES Nutritional analyses were conducted on selected wildlife forages. Six species— black cherry, red maple, beech, aspem, brambles (Ms spp), and bracken fern (Pteridimm agu_ilinum)- were selected for analysis on the basis of ungulate palatability (Rogers et al. 1981) and/or availability. Samples for all 6 species were collected for nutritional analysis during August 1988 (Sumner 1988), while samples of black cherry, red maple, and brambles were collected in January 1988 (Winter 1988). summer samples of black cherry, red maple, beech, and aspen were separated into twigs and leaves for analysis, as plant parts have been shom to differ in nutritional quality within a species (Nagy and Haufler 1980) . Aspen was present in sufficient quantities to sample only in the 16.1 mZ/ha (7o ftZ/acre) and 25.3 mZ/ha (110 ftZ/acre) study plots; thus, no nutritional analyses were performed on aspen from the controls. Sample collection for nutritional analyses involved randomly clipping samples from as large a number of individual plants per plot as possible (Campa 1982). This was done to cover as broad a range of individual and microsite variation as possible. Only current annual growth (CAG) was taken. Additionally, GAG samples were collected to a height of only 2m to allow the nutritional parameters analyzed to accurately reflect what is commonly available to browsing ungulates. Collected samples were placed in plastic bags and frozen as soon as possible following collection in Winter. Summer 1988 samples were immediately dried. Following collection, samples were oven dried to a constant weight 22 at 60°C. Samples were then gromd through a Wiley mill until they passedthroighalmmmesh. Gromdsampleswerestoredatroon tetperature in Whirl-pats (NASO) Inc, Ft. Atkinson, WI) until analyses were performed. Forage samples were analyzed for moisture content, total minerals (ash), crude fat (ether extract), and crude protein. Prior to chemical analyses, peroelt dry matter was determined for all samples. Dry matter was determined by drying 1.0-1.1 g subsamples at 100°C for 24 hours (Campa 1982). Dry matter determination assured that nutriticna analyses conducted would not be biased by failing to account for the weight of any relaining moisture in the dried vegetative samples. Percent moisture content and percent ash were determined following procedures outlined in A.O.A.C. (1975). Ether extract content was determined by the methods stated in A.O.A.C. (1975) with 2 modifications. First, vegetative samples were weighed into tared filter paper packets instead of thimbles to facilitate the extraction of a larger mmber of samples per run (Campa 1982). Secondly, a 3:1 mixture of anhydrous ethyl ether to methanol was used for extraction rather than pure ether to decrease volatility (H. Campa, pers. comm). Crude protein was calculated by first determining total nitrogen levels via a Kjeldahl digestion method utilizing a 'Iecator Block Digestor, Model 13-40 (Tecator Inc., Boulder, CD) and analyzing digested samples with a Technicon Autoanalyzer II (Technicon Industrial Systems, Tarrytown, NY). Percent crude protein was then calculated by multiplying total nitrogen values by 6.25 (Nagy and Haufler 1980) . 23 WIIDLIFE RESPONSES Elk and Deer Elkarddeerresponsestothevariolsthinningtreatmentswere determined by pellet-group coints and browse utilization surveys. Additionally, track counts were also to be used to assess ungulate utilization by dragging firebreaks constructed around each study plot; the inability to have firebreaks established around all the study plots due to the drought conditions of 1988 made track counts impossible, however. Browse utilization surveys were conducted on all study plots during early April 1988. ‘Ihree ungulate browse species of high, medium, and low prefereoe (red maple, black cherry, and beech, respectively) were chosen for utilization surveys in order to sample the full continuum of ungulate browsing pressure. Surveys were conducted using the 100 point intensive browse survey technique as described by Wyoming Game and Fish (1982) . Percent utilization was calculated as the percentage of available (less than 2 m high) twigs that were browsed. Pellet-group counts were conducted on 3 randomly placed 6m X 100m permanent transects on each study plot. Pellet-groups were tallied by species on eachtransect, andall 3transectswerecombinedtogiveone valueperplotforthe3parameters;deergro1ps, elkgroups, andtotal (combined deer and elk) groups. All transects were cleared of all groups during counting to minimize seasonal and double count biases (Neff 1968) , and all counts were made by the same individual to minimize observer biases (Neff 1968) . Pellet-group colnts were expressed as 24 groups counted per plot, and used for relative corparisons between treatments and controls. Small Mammals Stall mammal populations were monitored by live-trapping a 6 X 6 grid centered in each study plot. Trap spacing was 20m betwee1 trap locations, and trapping was conducted twice annually. In 1987, 2 Sherman live traps (H.B. Sherman Co., Tallahassee, FL) were placed at each trap location; due to the low capture rate encountered, this was changed to 1 trap per location in 1988. Traps were selectively placed at each grid location to maximize the likelihood of capture and covered with vegetation. Trapping was conducted in both July and August of 1987 and 1988. All study plots were trapped concurremtly with a 5 day trapping period. 'Ihe bait used consisted of a mixture of either rolled or crushed cats, peanut butter, dry commercial 6% protein dog food, raisins, and anise extract. Traps were checked early each morning of the trapping period. All stall mammal captures were identified to species and either ear tagged or toe clipped. Species, idemtification number, and location of capture were recorded for all captured individuals. 25 MA ANALYSIS Statistically adequate sample sizes for all measurements were determined using Freese's (1978) sample size determination formula: 11 = tzsz/Ez, where t = tabulated t value at the 90% probability level 52 = sample variance E = allowable error (mean X 0.10) OoIparisons of all data collected among all 3 treatment combinations (16.1 mZ/ha (7o ftZ/acre), 25.3 mZ/ha (110 ftZ/acre), and control) were made using the nonparametric Kruskal-Wallis one-way analysis of variance (Seigel 1956) . Corparisons involving only 2 of the treatment combinations, or corparisons between the 1 and 2 year old 16.1 mZ/ha (70 ftz/acre) thinnings, were made using the nonparametric Mann-Whitney U test (Seigel 1956). Small mammal diversity was determined by the Shannon-Wiener diversity index (Ricklefs 1979): H=- pilnpi, where Pi = the proportion of species i in a sample of N species we to the small number of small mammals captured, the enumeration technique of Krebs (1966) was used as an index of small mammal 26 N = A+P, where N = minimum mmmber of individuals of a species alive at time t A=actualmm1berofiniividualsofaspeciescapturedattimet P = number of previously marked individuals of a species caught aftertimet,butrotattimet The minimum mzmber of individuals of each species captured was determined for each study plot for each trapping period. Vegetative Corposition Analysis of baseline vegetative data from 1987 indicated that the 25.3 mz/ha (110 ftZ/acre) and control plots were similar in vegetative corposition. A list of all species found on the study plots is presented in the Appendix (Table A-1) . No baseline data were taken for the 16.1 mZ/ha (70 ftz/acre) plots, as thinning was already underway at this point. Of the 57 herbaceous species rworded on all study plots combined in 1987, all but mullein (Verbascum thapgs), enchanter's nightshade (Circaea alpina) , Virginia creeper (Parthenocissus gum efolia) , buttercup (Ranunculus spp) , and violet (Viola spp) were present on the 25.3 mz/ha (110 ftz/acre) plots. Honeysuckle (lonicera canadensis), pyrola (m seomda), bunchberry (Ms canadensis), pinegrass (Galmtis spp) , horsetail (Eggisetum arvense) , white avens (gym canadense), beggar's tick (megs frondosa), cleiatis (Cleratis viginiana) , and trillium (Trillium spp) were absent from the controls. Twenty-three woody species were recorded on the plots in 1987. All species recorded were present on both 25.2 mz/ha (110 ftz/acre) plots and controls. Vegetative corposition remained similar on all treatment and control plots in 1988 as well. Of the 43 herbaceous species recorded in 1988, 30 were present on the 16.1 mz/ha (70 ftZ/acre) plots, 31 on the 27 28 25.3 mZ/ha (110 ftZ/acre) plots, and 30 on the controls. Only 1 species, dock (m acetosella), was fond on the plots in 1988 that was rot present in 1987. Fifteen species present in 1987 were absent in 1988. Nineteen woody species were recorded on all plots in 1988. All species were present on all treatments and controls, with the exceptions of witchhazel (H. virg' iniana) and striped maple (A. mlvanicum) , which were absent from the controls, and white birch, which was absent from the 16.1 mZ/ha (70 ftZ/acre) plots. Woody Densities Density of woody shrubs ard sprolts was similar in 1987 (Table 2). Following treatments, no significant differences were present in total sters or number of species between 16.1 mZ/ha (7o ft2/acre) plots, 25.3 mZ/ha (110 ftZ/acre) plots, and controls, although 16.1 mZ/ha (7o ftZ/acre) thins averaged 700 and 1100 sters/ha more than the 25.3 mz/ha (110 ftZ/acre) thins and controls, respectively (Table 3) . No significant differences were fond in individual species densities in 1987 (Table 2). In 1988 (posttreatment), white ash (Fraxinus americana) was significantly more abundant in 16.1 mZ/ha (70 ftz/acre) thins than either 25.3 mZ/ha (110 ft2/acre) thins or controls. White birch was significantly more aburdant in 25.3 mZ/ha (110 ftZ/acre) thins than in 16.1 mZ/ha (7o ftZ/acre) thins, ard serviceberry was more abundant in controls than the 16.1 mz/ha (7o ftZ/acre) thins (Table 3). The most aburdant species in all plots were black cherry, red maple, white ash, beech, aspen, and beaked hazelnut (Coglus cornuta). Table 2. Mean (SE) densities of all woody stems preseit on pretreatment 25.3 mz/ha (110 ftZ/acre) ard control plots in July 1987. BASAL AREA (mZ/ha) SPECIES 25.3 Aspen 1325 (570) 377 (270) Basswood 357 (207) 107 (90) Beaked Hazelnut 215 (87) 57 (42) Beech 1097 (192) 845 (60) Black;Cherry 6862 (1050)a 3840 (907)b Brambles 7962 (2852) 5867 (5050) Chokecherry 2395 (435) 2235 (890) Currant 2067 (577) 955 (892) Dogwood 45 (32) 2 (2) Elm 927 (655) 307 (200) Fir 455 (145) 247 (130) Honeysuckle 22 (10) 47 (22) Ironwood 1725 (1027) 760 (497) Oak 187 (75) 47 (15) Red Maple 5425 (1182) 4062 (1292) Red Pine 10 (7) 205 (125) Russian Crab 300 (115) 927 (807) Serviceberry 2980 (875) 2325 (137) Spruce 7 (5) 2 (2) 30 Table 2. (cont'd) BASAL AREA (mz/ha) SPECIES 25.3 striped Maple 65 (40) 52 (42) Sugar Maple 6607 (3080) 3032 (812) White Ash 600 (497) 142 (62) witchhazel 125 (82) 147 (147) mm 41589 (6280) 25741 (7450) a,b Means in the same row with different superscripts are significantly different (P < 0.10) 31 Table 3. Mean (SE) densities of all woody sters greater than 1m tall and less than 10.2cm dbh in 1988. BASAI.AREA (mR/ha) 25.3 SPECIES 16.1 CONTROL Aspen 233 (42) 389 (244) 52 (52) Beaked Hazelnut 336 (285) 458 (348) 96 (86) Beech 207 (60) 211 (54) 222 (118) Black.Cherry 305 (91) 509 (106) 515 (193) Brambles 101 (63) 400 (357) 37 (32) Chokecherry 82 (29) 82 (21) 158 (76) Elm 15 (05) 48 (25) 37 (31) Fir 21 (10) 14 (07) 32 (22) Ironwood 36 (07) 65 (26) 150 (55) Oak 74 (37) 19 (11) 5 (05) Red Maple 576 (244) 232 (122) 158 (63) Russian Crab 5 (05) 48 (41) 274 (257) Serviceberry 53 (13)a 96 (18)ab 147 (52)b Striped Maple 4 (04) 5 (04) o (00) Sugar Maple 115 (39) 54 (30) 122 (116) White Ash 1095 (137)a 111 (67)b 30 (20)b White Birch o (00)a 25 (12)b 5 (05)ab witchhazel 17 (11) 33 (21) 0 (00) TOTAL 3116 (430) 2450 (452) 2062 (532) a,b Means in the same row with different superscripts are significantly different (P < 0.10) 32 Corparisons between the 1 ard 2-year—old 16.1 mz/ha (70 ftZ/acre) thins irdicated that total woody densities were significantly greater on 2-year-old thinnings (Table 4). Individually, black cherry, red maple, beaked hazelnut, and fir were all significantly more abundant in the 2- year-old thins. Sugar maple (A. saccharum) was the only species significantly more aburdant on the 1-year-old sites. Herbaceous Frequencies The absolute and relative frequencies of all herbaceous species fond on the study plots are presented in the Appedix (Tables A-2 through A-4). Controls ard treatments were similar in species cotposition ard frequency in 1987 (Table A—2) . Only fescue (Festuca spp) differed significantly in 1987, with higher absolute ard relative frequencies on treatments as opposed to controls. In 1988, 5 species differed in both absolute ard relative frequencies between treatments ard/or controls (Table A—3) . Mullein, brote grasses (mus spp) , ard honeysuckle were all significantly more common on 16.1 mZ/ha (70 ftZ/acre) thins as opposed to 25.3 m2/ha (110 ftZ/acre) thins or controls, which did not differ. Moss was significantly more cormon on both 25.3 mZ/ha (110 ftZ/acre) thins and controls relative to 16.1 mZ/ha (7o ftZ/acre) thins, and western wheatgrass (W smithii) was significantly more comon on both 16.1 m2/ha (70 ftZ/acre) and 25.3 mZ/ha (110 ftZ/acre) thins than on controls. Only 1 species, dock, was present in 1988 ard not in 1987; however, both absolute and relative frequencies of dock were low. Colparisons of the 1 ard 2-year—old 16.1 mZ/ha (70 ftZ/acre) thins 33 Table 4. Mean (SE) densities of woody stems greater than 11m in height ard less than 10.2om dbh between the differentially aged 16.1 mZ/ha (70 ftZ/acre) study plots in 1988. TIMESINCE'IHINNING SPECIES 2 Yrs 1 Yr Aspen 218 (10) 241 (65) Beaked Hazelnut 1004 (800)a 2 (2)b Beech 97 (15) 161 (22) Black Cherry 645 (90)a 150 (43)b Brambles 70 (70) 154 (88) Chokecherry 126 (33) 6O (39) Elm 14 (7) 15 (7) Fir 59 (4)a 2 (2)b Ironwood 37 (10) 35 (11) Oak 78 (56) 22 (4) Red Maple 1367 (491)a 181 (39)b Red Pine 7 (7) 2 (2) Serviceberry 85 (29) 32 (6) Sugar Maple 0 (0)a 172 (42)b White Ash 611 (257)a 1187 (199)b witchhazel 37 (31) 7 (4) TOTAL 4500 (797)a 2424 (172)b a,b Means in the same row with different superscripts are significantly different (P < 0.10) 34 indicated that 5 species (strawberry (Fraggia vm iniana) , hawkweed (Hieracium aurantiacum) , starflower (Trientalis borealis) , Canada mayflower (Maianthemum canadelse) , ard trout lily (w americanum)) had significantly higher absolute frequencies in the 2-year—old thins (Table A—4). Bedstraw (Galium boreale) was more common in the 1-year-old thins. Hawkweed ard starflower were not significantly different in relative frequencies; the other 4 species were. 'Iotal herbaceous composition was similar between the differently aged 16.1 mZ/ha (70 ftZ/acre) thins. SMAILI‘W/JMAIS Atotal of9species ofsmallmammalwerecapturedduringthe colrse of this study. Of these 9 species, 6 were commonly captured on all plots: eastern chipmunk (Ellis striatus) , red-backed vole (Clethrioms mi), deer mouse (m spp), masked shrew (m cinereus) , jumping mouse (Nameozaw insignus) , ard eastern mole (Scalopus agu_aticus) . The remaining 3 species, red squirrel, flying squirrel (Glaums sabrinus), ard short-tailed shrew (Blarina brevicauda) , were captured only in 1987 from a limited number of sites. Small mammal community corposition differed among the study plots in 1987. Four of 6 species trapped during the July 1987 trapping period occurred only on the 25.3 mZ/ha (110 ftz/acre plots); red-backed vole, masked shrew, short-tailed shrew, ard red squirrel (Table 5). None of these species, however, had more than 2 individuals trapped. Of the 8 species rworded in August 1987, 5 were trapped only in 35 Table 5. Mean (SE) minimum populations, species richness, and small mammal diversity between pretreatment 25.3 mZ/ha (110 ftz/acre) and controls in 1987. BASALAREA (mZ/ha) PERIOD SPECIES 25.3 JULY Eastern oiipmmk 3.00 (0.76) 1.33 (0.27) Red-backed Vole 0.17 (0.17) 0 (0.00) DeerMouse 2.83 (0.81) 3.33 (1.21) Masked Shrew 0.33 (0.33) 0 (0.00) Short-tailed Shrew 0.33 (0.21) o (0.00) Red Squirrel 0.33 (0.33) 0 (0.00) TOTAL INDIVIDUAIS 7.00 (1.43) 4.67 (0.98) ’IUI‘ALSPECIFS 3.17 (0.50) 2.00 (0.27) DIVERSITY 2.43 (0.29) 2.10 (0.45) AUGUST Eastern Chipmunk 4.00 (1.18)a o (0.00)b Red-backed Vole 0.33 (0.21) 0 (0.00) JumpingMouse 6.33 (1.56) 2.33 (0.72) Deermouse 3.50 (2.12) 2.00 (0.82) Masked Shrew 0.83 (0.40) 0.33 (0.33) Short-tailed Shrew 0.33 (0.21) 0 (0.00) Red Squirrel 0.83 (0.40) 0 (0.00) Flying Squirrel 0.50 (0.34) 0 (0.00) m INDIVIDUALS 16.67 (3.27)3 4.67 (0.27)]0 TOTALSPECJTS 4.50 (0.57)a 2.33 (0.27)b DIVERSITY 3.63 (0.95) 1.87 (0.40) a,b Means in the same row with different superscripts are different (P < 0.10) 36 the pretreatment 25.3 mZ/ha (110 ftZ/acre plots); eastern chipmumk, red- backed vole, short-tailed shrew, flying squirrel, ard red squirrel (Table 5). Of these species, only the eastern chipmunk was captured often elolgh to be considered a common inhabitant. Cormmity corposition among treatments ard control was identical during both 1988 trapping periods. Of the 6 species captured, all were present in all treatrents and control in both trapping periods. No significant differences were present in total number of individuals known to be alive (all species), nlmmber of species captured, or small mammal diversity in July 1987 (Table 5). No significant differences were present among minimum known population sizes of individual species. Total mmbers (all species) ard number of species captured were significantly higher in the pretreatment 25.3 mZ/ha (110 ftZ/acre) thins as opposed to the controls in August 1987. Among individual species, only eastern chipmunks significantly differed in population numbers, with higher densities rworded in the pretreatment 25.3 mZ/ha (110 ftZ/acre) plots. Small mammal diversity was greater on pretreatment 25.3 m2/ha (110 ftz/acre) thins relative to controls, but not significantly so. Total small mammal mmmbers and species mmnbers differed significantly in July 1988 (Table 6): 16.1 mZ/ha (70 ftZ/acre) thins showed greater total numbers and species richness than 25.3 mZ/ha (110 ftZ/acre) thins or controls. Twenty five-point—three mZ/ha (110 ftZ/acre) thins and controls did not differ significantly in either totalsrallmammalmmbersornumberofspeciescaptured. smallmammal 37 Table 6. Mean (SE) minimum populations, species richness, ard small ‘mgggal diversity among all thinning treatments and controls in BASALAREA (mZ/ha) PERIOD SPECIES 16.1 25.3 CONTROL JULY Eastern diipmnk 3.56 (1.61) 1.00 (0.33) 0.33 (0.27) Red-backed Vole 3.89 (0.74)3 0.50 (0.21)b 0.33 (0.27)b DeerMouse 0.33 (0.25) 0.33 (0.21) 0.67 (0.67) Jumping Mouse 2.11 (0.46) 1.83 (0.50) 1.67 (0.27) Eastern Mole 0.11 (0.11) 0.33 (0.33) 0.33 (0.33) Masked Shrew 0.33 (0.16) 0.17 (0.17) 0.33 (0.33) TOTAL INDIVIwAIS 10.33 (2.04)a 4.17 (0.55)b 3.67 (0.27)b 'IOI‘ALSPECIIS 3.56 (0.32)a 2.67 (0.30)b 2.67 (0.27)b DIVERSITY 2.59 (0.29) 2.37 (0.31) 2.56 (0.29) AUGUST Eastern Chipmunk 2.00 (0.61) 1.83 (0.72) 0 (0.00) Red-backed Vole 4.44 (0.96) 2.00 (0.66) 1.33 (0.67) Deer Mouse 1.89 (0.57) 0.83 (0.48) 0.67 (0.67) Jumping Mouse 0.22 (0.15) 0.67 (0.33) 2.33 (1.21) Eastern Mole 0.33 (0.17) 1.17 (0.60) 1.00 (0.58) Masked Shrew 1.11 (0.35) 1.00 (0.63) 0.33 (0.33) mm INDIVIwAIs 9.00 (0.89) 7.50 (2.00) 5.67 (1.52) TOTAL SPECIES 3.67 (0.35) 3.33 (0.61) 2.67 (0.72) DIVEEGITY 2.55 (0.28) 2.88 (0.51) 2.30 (0.74) a,b Means in the same row with different superscripts are different (P < 0.10) 38 diversity did not differ among treatments or controls. Among irdividual species, only red-backed voles differed significantly in July 1988, showing greater mmmbers on 16.1 mZ/ha (70 ftZ/acre) thins as opposed to either 25.3 mZ/ha (110 ftZ/acre) thins or controls. In August 1988, totalirdividualmmbersardmmmberofspeciescaptlredincreased progressively with degree of thinning, but the differences were not significant (Table 6). Small mammal diversity did not differ among treatments or control. Altholgh all irdividual species mmmbers (with the exception of ijping mice) were greater on either 16.1 mZ/ha (70 ftZ/acre) or 25.3 mZ/ha (110 ftZ/acre) thins relative to the control, the differences were not significant. July 1988 corparisons between 1 ard 2-year-old 16.1 mz/ha (70 ftZ/acre) thins indicated that small mammal diversity was significantly higher on 1-year-old thins (Table 7) . Total individual numbers were higher on the 2-year-old thins, but not significantly so. Nmmber of species captured did not differ. Among individual species, deer mice were significantly more abindant on the l—year—old thinnings. No other species differed significantly in numbers. Eastern moles ard masked shrews were present only on the 1-year—old thinnings, but not in significant mmmbers. One and 2—year—old 16.1 mZ/ha (7o ftZ/acre) thins did not differ significantly in total numbers, total species captured, or small mammal diversity in August 1988 (Table 7). Among individual species, eastern chipmunks showed significantly greater numbers in the 2-year—old 16.1 mz/ha (70 ftZ/acre) thins, while red—backed voles showed significantly greater mmmbers in the 1—year-old thins. 39 Table 7. Mean (SE) minimum populations, species richness, and small mammal diversity between the two age classes of 16.1 mZ/ha (70 ftZ/acre) thins in 1988. TIME SINCE 'IHINNING PERIOD SPECIES 2 Yr 1 Yr JULY Eastern Chipmunk 6.00 (4.50) 2.33 (0.39) Red-backed Vole 3.33 (0.98) 4.17 (0.98) DeerMouse 1.00 (0.47)a 2.67 (0.51)b Jumping Mouse 0.33 (0.33) 0.17 (0.17) Eastern Mole 0 (0.00) 0.17 (0.17) Masked Shrew o (0.00) 0.50 (0.22) TOTAL INDIVIDUALS 11.00 (5.56) 10.00 (1.25) TOTAL SPECIES 3.00 (0.82) 3.83 (0.15) DIVERSITY 1.72 (0.38)a 3.02 (0.24)b AUGUST Eastern Chipmunk 4.33 (0.67)a 0.83 (0.31)b Red-backed Vole 1.00 (0.58)a 6.17 (0.60)b DeerMouse 1.33 (0.88) 1.83 (0.48) Jumping Mouse 0.67 (0.33) 0 (0.00) Eastern Mole 0.33 (0.33) 0.33 (0.21) Masked Shrew 2.00 (0.00) 0.67 (0.42) TOTAL INDIVIDUALS 9.67 (0.72) 10.17 (1.28) TOTAL SPECIES 4.33 (0.54) 3.33 (0.39) DIVERSITY 3.25 (0.60) 2.20 (0.20) a,b Means in the same row with different superscripts are different (P < 0.10) 40 BIGGAMEUSE Pellet-Group Oonrts Deer ard combined pellet-grolp comts were significantly greater in 16.1 mZ/ha (7o ftz/acre) thins as opposed to controls in Spring 1988 (Fig. 4). Neither 16.1 mz/ha (70 ftZ/acre) thins nor controls differed significantly from 25.3 mZ/ha (110 ftZ/acre) thins, which were intermediate in pellet-groups colnted. Although elk pellet-groups showedatrerdtoincreasewithamolmtofthinning, theincreaseswere rot significant. Spring 1989 pellet-group counts revealed significant differences in numbers between both thinnings ard control for all 3 categories (Fig. 5). Both 16.1 mZ/ha (70 ftZ/acre) and 25.3 mZ/ha (110 ftz/acre) thins showed significantly higher elk pellet group numbers than did the controls; 16.1 mz/ha (70 ftz/acre) thins were greater than the 25.3 mZ/ha (110 ftz/acre) thins as well, but the differences were not significant. Deer and combined pellet-group counts segregated into 3 significantly distinct groupings; 16.1 mZ/ha (70 ftz/acre) thinnings > 25.3 mz/ha (110 ftZ/acre) thinnings > controls (Fig. 5). Compared to controls, deer group counts were 510 and 281 percent higher in 16.1 mZ/ha (7o ftZ/acre) and 25.3 mZ/ha (110 ft2/acre) thins, respectively. Total combined group counts were 484 ard 291 percent greater in 16.1 mZ/ha (7o ftZ/acre) and 25.3 mZ/ha (110 ftZ/acre) thins, respectively, corpared to controls. No significant differences were present among elk, deer, or 41 Figure 4. Mean (SE) species specific ard total pellet-group counts per thinning (mZ/ha) treatment for Spring 1988. 42 A480,“. ”~me ad th D I AOMHZOU now an: an ON J.O']d 33d SdflOHf) 43 Figure 5. Mean (SE) species specific and total pellet-group counts per thinning (mZ/ha) treatment for Spring 1989. 44 Adhbh. MHHQ MAN E D I AOEOU o.— H an. cm on .. an .I.O'ld 83d SdflOHD 45 ccmbined pellet-group comts between the 1 ad 2-year—old 16.1 mZ/ha (70 ftz/acre) thins in 1988 or 1989 (Fig. 6) . Browse Utilization The percentage of browsed stets for black cherry, red maple, or beech did not differ significantly between 16.1 mZ/ha (70 ftZ/acre) thins, 25.3 mz/ha (110 ftZ/acre) thins, and controls in 1988 (Fig. 7). Corparisons of the 1 ard 2-year-old 16.1 mz/ha (70 ftz/acre) thins revealed black cherry to be browsed significantly more on the 2-year-old thinnings (Fig. 8) . Red maple and beech did not differ between the two ages of 16.1 mZ/ha (70 ftZ/acre) thins. Browse ratios based on preference (red maplezblack cherry:beech) did not differ among treatrents or controls, nor between the differently-aged 16.1 mZ/ha (70 ftz/acre) thins. NUTRITIONAL ANALYSES Moisture Contelt Moisture contents of red maple, black cherry, ard brambles did not differ significantly in samples collected during January 1988 between 16.1 mz/ha (7o ftz/acre) thins, 25.3 mZ/ha (110 ftZ/acre) thins, and controls (Table 8). In cotparisons between the 1 and 2-year-old 16.1 mZ/ha (70 ftZ/acre) thins, only brambles differed significantly in percent moisture content; brambles from l-year-cld thinnings showed a 42% higher moisture content (Table 9). For woody and 1 herbaceols species were analyzed for percelt 46 Figure 6. Mean (SE) species specific ard total pellet-group counts per study plot for the two ages of 16.1 m /ha (70 ftz/acre) thins for 1988 and 1989. moms essay meme 47 Mam moms stack memo mom hmmw owccEe. D mmmr ooccE... I .ou .oN .. on .LO'Id 83d SdflIOHS 48 Figure 7. Mean (SE) percent utilization of selected ungulate forages for all thinning (mZ/ha) treattents for the PRCSF study plots in Spring 1988. 49 an; 9mm womwm FMMWMU Mojm ID AOMHZOU now as: L I mu om SDIMJ. 033M088 % 50 Figure 8. Mean (SE) percent utilization of selected 2ungulate forages for the two age classes of 16.1 m 2/ha (70 ft:2 /acre) thins in Spring 1988. 51 mum; 9mm Wmmwmnu Momma Mofim D hood tonnage. I wood ooflfldfl. J I mN cm SDIMJ. 038M038 x 52 Table 8. Mean (SE) percent moisture content in forage samples collected in Winter ard Summer 1988 among all treattent combinations. BASAL AREA (mR/ha) PERIOD SPECIES 16.1 25.3 (IEVTROL WINTER Black Cherry 47.05 (1.66) 44.85 (1.39) 46.44 (0.98) Red Maple 49.05 (0.61) 48.63 (0.48) 49.25 (0.13) Brambles 37.36 (3.17) 35.67 (3.06) 35.63 (5.42) SUMMER Black Cherry — T 43.53 (0.97)a 38.74 (1.03)b 43.43 (1.86)a Black Cherry - L 34.64 (0.40)a 29.35 (0.80)b 32.24 (1.96)a Red Maple — T 45.07 (0.82) 43.91 (1.30) 47.14 (0.13) Red Maple - L 42.35 (0.76)a 38.35 (2.08)b 37.62 (1.48)b Beech — T 47.04 (0.53) 48.07 (0.30) 47.48 (0.67) Beech - L 47.05 (0.87) 48.03 (0.61) 45.99 (0.43) Aspen - T 34.74 (1.69)a 28.75 (0.88)b --- Aspen - L 32.14 (1.13)a 27.79 (0.78)b --—- Brambles 33.09 (1.14)a 29.26 (0.83)b 28.09 (1.19)b Bracken Fern 30.99 (1.51) 28.12 (1.49) 24.13 (3.43) a,b Means in the same row with different superscripts are differe1t (P < 0.10) T = ’Iwig samples L = leaf samples 53 Table 9. MEan (SE) percent moisture content in forage samples collected in Winter and Summer 1988 from the differentially aged 16.1 mZ/ha (7o ftZ/acre) thins. TmE SINCE 'IHINNING PERIOD SPECIES 2 Yr 1 Yr WINTER Black Cherry 43.22 (4.38) 48.96 (0.74) Red Maple 50.21 (0.16) 48.47 (0.82) Brambles 29.26 (4.72)a 41.41 (3.13)b SUMMER Black cherry - T 45.19 (0.70) 42.70 (1.31) Black Cherry - L 35.54 (0.22) 34.19 (0.49) Red Maple - T 46.40 (0.43) 44.41 (1.14) Red Maple — L 42.55 (0.66) 42.26 (1.14) mob - T 46.60 (0.51) 47.26 (0.77) Beech - L 46.79 (0.67) 47.18 (1.30) Aspen - T 37.64 (3.06) 33.29 (1.93) Aspen - L 35.08 (1.81) 30.67 (1.05) Brambles 35.97 (0.88)a 31.65 (1.30)b Bracken Fern 32.87 (1.22) 30.05 (2.16) a,b Means in the same row with different superscripts are different (P < 0.10) T = Twig samples L = leaf samples 54 moisture content in samples collected during August 1988. With the exception of brambles, woody species were analyzed separately for leaves ard twigs. For the Summer 1988 samples, percent moisture content was significantly higher in black cherry leaves ard twigs collected from 25.3 mZ/ha (110 ftZ/acre) thins as compared to 16.1 mZ/ha (7o ft2/acre) thins or controls, which did not differ (Table 8). Red maple leaves, aspen twigs ard leaves, and brambles showed significantly higher moisture contents when collected from 16.1 mZ/ha (7o ft2/acre) thins corpared to 25.3 mZ/ha (110 ftZ/acre) thins or controls, which did not differ. No other species differed in moisture content. Comparisons between 1 and 2-year-old 16.1 mZ/ha (70 ftZ/acre) thins revealed a significant difference in percent moisture content only for brambles, which, in contrast to the Winter 1988 samples, showed higher moisture content in the 2-year-old thinnings (Table 9). Ash Content Total mineral contents did not differ significantly among either treatments or controls for the 3 woody species collected in Winter 1988 (Table 10). No trends were visible in the data; black cherry showed highest mineral content in the 16.1 mz/ha (70 ftZ/acre) thins, red maple in the 25.3 mZ/ha (110 ftZ/acre) thins, and bramble mineral content was highest in the controls. Red maple twigs collected in the 1-year—old 16.1 mZ/ha (70 ftZ/acre) thins showed significantly higher total mineral contents than those collected in the 2—year-old thinnings (Table 11) . Neither black 55 Beech-T Beech—L ASpen-T ASpen-L Brambles BrackenFern 1.62 (0.14)a 4.23 (0.31) 2.58 (0.36) 5.47 (0.34) 5.04 (0.22) 6.01 (0.39) 2.20 (0.16)b 3.95 (0.54) 2.93 (0.30) 5.21 (0.28) 5.26 (0.30) 6.12 (0.27) Table 10. Mean (SE) percent ash (total minerals) in forage samples collected in Winter and Summer 1988 among all treatment combinations. BASAL AREA (mz/ha) PERIOD SPECIES 16.1 25.3 mNTROL WINTER Black Cherry 2.98 (0.24) 3.66 (0.19) 3.76 (0.35) RedMaple 3.08 (0.21) 3.22 (0.32) 2.76 (0.43) Brambles 3.30 (0.11) 3.12 (0.13) 2.96 (0.24) SUMMER Black Cherry - T 2.24 (0.21) 2.05 (0.12) 2.36 (0.59) Black Cherry — L 5.11 (0.19)al 5.49 (0.69)ab 6.79 (0.17)b Red Maple - T 1.99 (0.16) 2.32 (0.24) 2.28 (0.61) Red Maple — L 4.35 (0.13) 4.39 (0.35) 4.21 (0.31) 2.04 (0.31)ab 3.41 (0.25) 5.04 (0.65) 6.42 (0.50) a,b Means in the same row with different superscripts are different (P < 0.10) T = 'Iwig samples L = leaf samples 56 Table 11. Mean (SE) percent ash (total minerals) in forage samples collected in Winter and Summer 1988 from the differentially aged 16.1 mZ/ha (70 ftZ/acre) thins. TIME SINCE THD‘IND‘IG PERIOD SPECIES 2 Yr 1 Yr WINTER Black Cherry 3.14 (0.50) 2.90 (0.29) Red Maple 2.53 (0.14)a 3.35 (0.24)b Brambles 3.18 (0.15) 3.37 (0.16) SUMMER Black Cherry - T 2.62 (0.19) 2.05 (0.28) Black Cherry - L 5.25 (0.36) 5.04 (0.23) Red Maple - T 1.78 (0.29) 2.10 (0.20) Red Maple - L 4.21 (0.09) 4.42 (0.19) Beech - T 1.54 (0.06) 1.66 (0.21) Beech - L 4.29 (1.02) 4.20 (0.12) Aspen - T 3.02 (0.93) 2.36 (0.33) Aspen - L 6.18 (0.92) 5.12 (0.19) Brambles 5.34 (0.25) 4.96 (0.31) Bracken Fern 6.29 (0.44) 5.87 (0.56) a,b Means in the same row with different superscripts are different (P < 0.10) T = Twig samples L = leaf samples 57 cherry nor brambles differed significantly between 1 or 2-year—old 16.1 mZ/ha (70 ftZ/acre) thins for Winter 1988. 'Ibtal mireral contents for the 6 species sampled in Simmer 1988 revealed only beech twigs and black cherry leaves to differ significantly between treatments ard comtrols (Table 10). Beech twig samples collected from 25.3 mZ/ha (110 ftz/acre) thins showed significantly higher mineral contents than did those collected from either 16.1 mz/ha (70 ftZ/acre) thins or controls, which did not differ significantly. Black cherry leaf mineral content was significantly higher in samples from the controls as compared to 16.1 mZ/ha (70 ftZ/acre) thins,- 25.3 mZ/ha (110 ftZ/acre) thins did not differ from either controls or 16.1 mz/ha (70 ftz/acre) thins. Corparisons between the 1 ard 2-year-old 16.1 mz/ha (70 ftz/acre) thins revealed no significant differences within a species. Seventy percent of the forages analyzed, however, showed higher mineral contents in the 2—year—old thinnings (Table 11) . Ether Extracts Black cherry crude fat levels were significantly higher in samples from 25.3 mZ/ha (110 ftZ/acre) thins than either 16.1 mZ/ha (70 ftZ/acre) thins or controls for Winter 1988 samples (Table 12) . No other species differed among treatments or control. No significant differences in crude fat levels were revealed in colparisons of 1 ard 2—year—old 16.1 mz/ha (70 ftZ/acre) thins for Winter 1988 samples (Table 13). For Summer 1988 samples, only aspen twigs differed significantly in 58 Table 12. Mean (SE) percent ether extract (crude fat) in forage samples collected in Winter and Summer 1988 among all treatment combinations. BASAL AREA (mQ/ha) PERIOD SPECIES 16.1 25.3 CONTROL WINTER Black.Cherry 11.27 (0.27)a 12.57 (0.26)b 12.02 (0.20)ab Red Maple 13.03 (1.08) 12.62 (1.00) 12.28 (0.67) Brambles 11.27 (0.14) 10.74 (0.73) 11.68 (0.18) SUMMER Black Cherry - T 9.58 (0.20) 9.50 (0.31) 10.95 (0.98) Black.crrmry - L 15.15 (0.76) 14.08 (0.72) 13.18 (0.29) Red Maple - T 9.75 (0.36) 9.37 (0.20) 10.17 (0.97) Red Maple - L 20.93 (1.64) 20.04 (1.39) 16.84 (0.91) Beech - T 8.39 (0.23) 8.93 (0.29) 8.40 (0.32) Beech - I. 11.37 (0.80) 10.04 (0.31) 10.46 (0.26) Aspen - T 13.12 (1.00)a 10.48 (0.34)b --- Aspen - L 16.69 (0.84) 16.89 (0.75) --- Brambles 11.45 (0.34) 11.23 (0.52) 12.27 (0.60) Bracken Fern 10.73 (0.27) 11.05 (0.54) 13.39 (1.57) a,b Means in the same row with different superscripts are different (P < 0.10) T = Twig samples L = leaf samples 59 Black Cherry - L Red Maple - T Red Maple — L Beech - T Beech - L Aspen ' T Aspen 7 L Brambles Bracken Fern 13.49 (0.49)a 12.35 14.85 14.08 11.90 11. 08 (0.18) (0.88) (0.37) (1.20) (1.61) (0.04) (0.57) (0.31) Table 13. Mean (SE) perce'lt ether extract (crude fat) in forage samples collected in Winter ard Summer 1988 from the differeltially aged 16.1 mZ/ha (7o ftZ/acre) thins. TIME SINCE 'I‘I-IleNG PERIOD SPECIES 2 Yr 1 Yr WINTER Black Cherry 10.90 (0.39) 11.45 (0.35) Red Maple 14.91 (0.97) 12.09 (1.43) Brambles 10.92 (0.29) 11.45 (0.11) SUMMER Black Cherry - T 9.88 (0.45) 9.43 (0.21) 16.14 (0.94)b 17.55 11.23 10.45 (0.54) (2.36) (0.22) (1.05) (1.20) (0.94) (0.43) (0.32) a,b Means in the same row differe1t (P < 0.10) T = 'Iwig samples L = leaf samples with different superscripts are 60 percent ether extract (Table 12) . Aspen twigs showed significantly higher crude fat contents in samples collected from 16.1 mZ/ha (70 ftZ/acre) thins as compared with 25.3 mZ/ha (110 ftz/acre) thins (no aspe'i was collected on the controls due to absence or insufficie'lt quantities). Black cherry leaves showed significantly higher crude fat levels in samples collected from 2—year—old 16.1 mz/ha (7o ftZ/acre) thins compared with l-year-old 16.1 mZ/ha (7o ftZ/acre) thins (Table 13). No other species differed between the differentially aged 16.1 mZ/ha (70 ftZ/acre) thins. Crude Protein Protein analysis of the 3 species collected Winter 1988 revealed that only black cherry differed in crude protein levels among treatments and controls (Table 14). Black cherry samples from 25.3 mZ/ha (110 ft2/acre) thins and controls had significantly higher protein levels than samples from 16.1 mZ/ha (70 ft2/acre) thins. Crude protein levels for red maple ard brambles tended to increase with increased degree of thinning, but the differences were not significant. Comparisons of 1 and 2-year—old 16.1 mZ/ha (7o ftZ/acre) thins revealed no significant differeoes in crude protein levels among treatments ard controls for any species (Table 15). Samples collected from the 2-year—old thinnings tended to be consistently higher in crude protein concentration, but not significantly so. Black cherry leaves, red maple leaves ard twigs, beech leaves, ard aspen leaves ard twigs all showed significant differelces in crude 61 Table 14. Mean (SE) percent crude protein in forage samples collected in Winter and Summer 1988 among all treatment combinations. BASAL AREA (mP/ha) PERIOD SPECIES 16 . 1 25 . 3 CDNTROL WINTER Black Cherry 5.05 (0.34)a 6.54 (0.24)b 6.67 (0.27)b Red Maple 4.93 (0.40) 4.99 (0.55) 4.73 (0.31) Brambles 4.97 (0.30) 4.61 (0.42) 3.46 (0.61) SUMMER Black Cherry - T 4.99 (0.25) 4.67 (0.32) 4.96 (0.56) Black.Cherry - L 14.66 (0.54)ab Red Maple - T Red Maple - L Beech-T Beech-L Aspen ’ T Aspen-L Brambles Bracken Fern 2.80 (0.21)ab 9.97 (0.22)a 3.77 (0.31) 11.79 (0.36)a 4.97 (0.14)a 15.21 (0.53)a 11.58 (0.72) 10.56 (0.82) 16.14 (0.39)a 2.55 (0.22)b 8.97 (0.18)b 4.22 (0.17) 10.11 (0.60)b 5.65 (0.58)b 17.71 (0.77)b 12.56 (0.29) 10.93 (0.79) 13.35 (0.38)b 3.52 (0.15)a 9.06 (0.17)b 3.83 (0.18) 9.73 (0.15)b 11.48 (0.49) 11.81 (3.61) a,b Means in the same row with different superscripts are different (P < 0.10) T = Twig samples L = Leaf samples 62 Table 15. Mean (SE) percent czude protein in forage samples collected in Winter and Summer 1988 from the differentially aged 16.1 mZ/ha (70 ftZ/acre) thins. TIME SINCE 'IHIINNDJG PERIOD SPECIES 2 Yr 1 Yr WINTER Black Cherry 5.17 (0.26) 4.98 (0.56) Red Maple 5.54 (0.52) 4.63 (0.52) Brambles 5.40 (0.32) 4.71 (0.41) SUMMER Black Cherry - T 5.06 (0.11) 4.95 (0.42) Black cherry — L 13.54 (0.48) 15.22 (0.68) Red Maple - T 2.69 (0.20) 2.85 (0.31) Red Maple — L 9.63 (0.38) 10.15 (0.26) Beech - T 3.42 (0.16) 3.98 (0.49) Beech - L 12.10 (0.49) 11.64 (0.50) Aspen - T 4.88 (0.25) 5.03 (0.18) Aspen - L 14.25 (0.63) 15.69 (0.67) Brambles 12.58 (0.67) 10.83 (1.08) Bracken Fern 8.86 (0.44)81 11.41 (1.06)b a,b Means in the same row with different superscripts are different (P < 0.10) T = Twig samples L = leaf samples 63 protein levels among treatments for Summer 1988 (Table 14) . Black cherry leaves and red maple twigs collected from 25.3 mZ/ha (110 ftZ/acre) mins were significantly higher in crude protein than samples collected from the controls; neither 25.3 mZ/ha (110 ftZ/acre) thins nor controls differed significantly from 16.1 mZ/ha (7o ftZ/acre) thins. Red maple leaves and beech leaves showed significantly higher crude protein levels in 16.1 mZ/ha (70 ftZ/acre) thins compared with 25.3 mz/ha (110 ftZ/acre) thins or controls, which did not differ. In contrast, both aspen leaves and twigs showed significantly higher crude protein levels in 25.3 mZ/ha (110 ftZ/acre) thins as opposed to 16.1 mZ/ha (70 ftZ/acre) thins. Neither black cherry twigs, beech twigs, brambles, nor bracken fern differed significantly in crude protein level amcmg treatments or controls. Bracken fern samples collected from l-year—old 16.1 mZ/ha (7o ftz/acre) thins showed significantly higher protein levels than those collected from 2-year-old 16.1 mZ/ha (7o ftz/acre) thins (Table 15). No other species differed among comparisons of 1 and 2-year-old 16.1 mz/ha (70 ftZ/acre) thins. DISCUSSION VEXIEI‘ATIVE RESPONSES Baseline vegetation data from 1987 indicated that vegetative carposition was similar among the 25.3 mz/ha (110 ftz/acre) plots and controls. No baseline data were available for 16.1 mZ/ha (70 ftZ/acre) treatments due to the timing of thinning operations. Following thinning, vegetative composition remained similar among all study plots. However, the nlmlber of plant species present on both treatments and controls decreased considerably from 1987 to 1988. The decline in species richness on all plots is probably attributable to the drought conditions of 1988. Although it was hypothesized that the diversity of understory species would increase with the opening of the red pine overstory and subsequent greater resource availability (MacArthur 1972, Dickmann et al. 1987), the lack of precipitation may have mediated this effect. Additionally, all study plots were located on high quality sites; in spite of the tendency of conifer stands to tie up nutrients (Miller et al. 1979), the inherent quality of these sites may have diminished any thinning induced compositional changes. These high quality sites may have already supported close to their maximum compliment of species, with responses to thinning manifesting themselves as changes in frequency, density, or productivity of the species already present. The tenuousness of the 64 65 foothold of many of these species was indicated by the prdaably drought-induced loss of 15 species from the study plots between 1987 and 1988. Finally, all results of this study are responses to only 1 or 2 growing seasons following overstory thinning; thinning responses may not be fully apparent until 3-5 years post-thin (D. Dickmann, pers. comm.) . 'I‘rerrisofincreasedwoodystemdensitieswereapparentinthis study, although the differences were statistically significant only on a temporal basis, i.e. between 1 and 2-year—old 16.1 mZ/ha (70 ftZ/acre) thinnings. Despite differences of up to 50% in the total number of stems per hectare between the two thinning densities and controls, large variances within a treatment, particularly between 1 and 2-year-old 16.1 mZ/ha (7o ftZ/acre) thins, prevented statistical significance. The statistically significant differences between the differentially aged 16.1 mZ/ha (70 ftZ/acre) plots, especially the magnitude of these differences (2-year-old thins had almost double the stem densities of 1 year-old thins), suggested that the trend towards increased stem densities on thinned plots will become more pronounced with time, at least until the red pine canopy closes. The increased stem densities werebroughtaboutnotbyinvasionofnewspecies, butbyanincreasein the densities of species already present; particularly important were the increased densities of important wildlife forage species, especially red maple, white ash, aspen, and beaked hazelnut. The increased presence of these species enhances the quality of the thinned red pine stands as wildlife habitat beyond the level produced simply by increased understory densities alone. 'Ihus, it appears that thinning of the red pine overstory is resulting in a more abundant and structurally complex, 66 if not more diverse, understory which is of increasing attractiveness to wildlife. Similar responses to thinning pine overstories have been reported elsewhere. Dickmann et al. (1987) found that the lighter the overstory red pine stocking in Michigan, the heavier the understory growth. Increased understory stem densities and productivity have been long established results of thiming southern pines (Halls 1970, Blair and Brunnett 1977, Wolters 1982, Wolters et al. 1982, Halls and Boyd 1982), and thinning is considered essential to maintaining adequate production of wildlife forages on these forest types (Blair 1967, Halls 1973, Blair and Enghardt 1976, Hurst and Warren 1982). In the West, understory development was found to increase with pine overstory thinning in numerous studies covering a variety of different pine types (Clary and Pfolliott 1966, Barrett 1968, Reynolds 1969, McConnell and Smith 1970, 'Ihill et al. 1983, Crouch 1986). Krefting and Phillips (1970) also found similar understory increases following thinning mixed conifer stands in Michigan. The understory density responses observed in this study are probably a result of greater resource capture following opening of the red pine overstory. Thinning allows a greater radiant energy supply to reach the understory, resulting in greater assimilation rates and levels of photosynthesis (Barbour et al. 1980) . Moisture regimes also improve as a result of overstory thinning, as more rainfall reaches the understory due to less overstory interception; additionally, soil moisture regimes are improved due to decreased evapotranspiration in the overstory and less root competition. The significance of this increased 67 moisture availability may be demonstrated when considered in light of a species water requiretent, or ratio of water transpired to herbage produced. For most native vegetation of the United States, this ratio is approximately 700:1; therefore, "700 kg of water are needed to produce 1 kg of vegetation (Stoddart et al. 1975) . A third factor that may have affected understory development involves the freeing of nutrients, especially nitrogen, that tend to become immobilized in the poor quality conifer litter; the higher quality litter produced by the more productive deciduous understory tends to stimulate decomposition, resulting in greater nutrient cycling and hence availability (K. Pregitzer, pers. comm.) . Relative nutrient availability is also increased due to less root competition following overstory mortality. Therefore, it is likely that a combination of the above factors is responsible for the trend toward increased woody densities of understory species observed in this study. Additionally, although not evaluated in this study, it was visibly apparent that total herbage production was much greater on the thinned plots, especially the 16.1 m2/ha (70 ftZ/acre) thins, even though the composition and frequencies of herbaceous species did not differ among treatments and controls. This increase in total herbage production is also attributable to the above hypotheses. NUTRITIONAL ANALYSES Foraging wildlife species often preferentially select certain individual plants over others. A frequently given reason is that 68 certain plants are more palatable than others due to higher nutritional quality (Stoddart et al. 1975, Robbins 1983). Many factors have been implicated to influence palatability and hence animal selection, including nutrient content, especially protein, moisture content, mineral content, fiber or lignin content, texture, essential oils, and taste (Stoddart et al. 1975) . Analysis of certain of these nutritional factors may therefore help to indicate my certain sites receive preferential use by wildlife, as well as to indicate means of increasing wildlife use of sites. Certain complications exist in the interpretation of nutritional data, however. One important problem is that many of these nutritional parameters are interrelated. For example, highest protein levels tend to occur in plants early in the growing season when moisture content is also high, but lignin and fiber levels are apt to be low (Stoddart et al. 1975). Thus, it is not certain whether an animal selects for the high protein levels, high moisture content, or against the high fiber levels . Crude Protein Proteins are major constituents of an animals body, performing important roles in tissue construction as well as enzymes, hormones, fat transport, immune responses, and many other factors (Robbins 1983). Thus, it is important that a continuous adequate supply of protein be available to meet an animal's growth, maintenance, and reproductive requirements. Additionally, protein levels play an important role in plant palatability and thus animal selection (Stoddart et al. 1975) . 69 Protein responses in wildlife forages to thinning pine overstories have not been well documented. What work has been dore, however, suggested that understory forage protein levels may be enhanced by overstory pire thinning (Wolters 1973 , Blair and Brunrett 1977, Crouch 1986). Results from this study tended to be inconsistent, not showing the trends indicated by the studies noted above. Only 3 of 13 forages analyzed, summer black cherry leaves, summer red maple leaves, and summer beech leaves, showed significantly higher crude protein levels in either the 16.1 mZ/ha (7o ftZ/acre) or the 25.3 m2/ha (110 ftZ/acre) thins compared to the controls. In contrast to this, winter black cherry samples and summer red maple twigs showed significantly higher crude protein levels in the controls, and both aspen twigs and leaves had highest protein levels in the 25.3 mZ/ha (110 ftZ/acre) thins compared to the 16.1 mz/ha (70 ftz/acre). 'Ihus, protein levels did not reveal any increasing trends with red pine overstory thinning in this study. All winter forage samples and all summer twig samples in this study were well below the 7% crude protein level generally accepted as the minimum level necessary for normal growth and maintenance of white-tailed deer (Murphy and Coates 1966, Verme and Ullrey 1972). All summer leaf and herbaceous samples were well beyond this minimum, although 15 of the 17 summer forage/treatment combinations were below the 16% crude protein level reported for optilmmm growth and development (Verme and Ullrey 1972). Thus, thinning red pine stands did not allow for the development of sufficient protein levels to meet the minimal 70 requirements for ungulates for any winter forage samples, nor for optimum growth requirements with the summer forages sampled. Additionally, it should be noted that the protein levels recorded in this study are prdaably in excess of the actual protein levels present in the forages; recorded protein levels were based on total nitrogen levels, of which true proteins typically account for only 75-85% (Robbins 1983) . Although ruminants microbially synthesize their own proteins and amino acids from the nitrogen available in their diet (Short 1981), and thus receive high quality and easily digestible protein regardless of the quality of the forage protein, microbial productivity often can be improved by providing specific amino acids in the diet (Fontenot 1971) . In such cases where microbial protein synthesis does not completely satisfy amino acid requirements, forages with high protein levels are more likely to provide the needed amino acids. Die to the low crude protein levels found in this study, the understories in both thinned and unthinned red pine stands appear to inadequately meet the protein needs of ungulates, especially in winter. Additionally, specific amino acid needs, if present, are less likely to be met due to the low protein levels observed. Mature conifer forests are known to become extremely nitrogen deficient in the soils and organic matter layer as they develop, particularly on poorer soils (Miller et al. 1979) . This is due to the acidic, high carbon to nitrogen ratio litter of conifers, which limits the effectiveness of decomposers in breaking the litter down and freeing nitrogen; thus, nitrogen becomes immobilized in the organic matter layer and is effectively lost to the system. This negative nitrogen feedback 71 loop characteristic of conifer forests may be responsible for the low protein levels found in most wildlife forages in this study, as plants are unable to synthesize high protein levels in a nitrogen poor environment. Althoughnotcurrentlyseeninthisstuiy,theincreasedcrude protein levels associated with thinning overstory pire by others (Walters 1973, Blair and Brunnett 1977, Crouch 1986) might be the result of breaking this negative feedback due to the invasion of herbaceous and deciduous woody species following the opening of the overstory. This increase in understory species could have two primary effects. First, lmderstory herbaceous and deciduous woody species would provide a higher quality litterfall (lower carbonznitrogen ratio) into the system, stimulating decomposers and increasing rates of decomposition and nitrogen cycling. Secondly, if some of these understory species are legumes, they can further stimulate decomposition rates and understory growth by introducing more nitrogen into the system via fixation. The net result of these processes, both likely results of overstory thinning, would be to encourage even greater understory production and to make nitrogen increasingly available, creating a positive feedback loop in terms of litter quality and nitrogen availability. With increased nitrogen availability, it is likely that crude protein levels of forages waild increase, resulting in more palatable forages due to enhanwd nutritional quality (Stoddart et al. 1975) . Thus, it is likely that if the currently observed trend of increased understory development with thinning continues, positive protein responses may yet manifest themselves. However, since nitrogen 72 is typically the most limiting nutrient in forest communities (K. Pregitzer, pers. comm), its dynamics tend to be cyclically tight and complex. The dynamics recessary to increase forage protein levels therefore may in actuality be far more corplex than simply establishing a positive nitrogen feedback loop via enhanced understory development and increased rates of decomposition. Additionally, increased nitrogen availability and uptake may not recessarily result in higher forage protein levels. The increasingly available nitrogen may be diluted due to greater forage production, resulting in increased vegetative biomass but unchanged or even decreased forage protein levels. Moisture Content Water is one of the most essential nutrients for wildlife (Robbins 1983). Water has often been described as the basis for life; it is crucial in the animal body for a variety of activities including hydrolytic reactions, thermoregulation, transport and excretion of metabolic products/wastes, and many other solvent functions (Robbins 1983). Functioning in such vital metabolic roles, water is typically needed in large quantities by wildlife, which have 3 potential sources: free water, metabolic water, and water from foods (Robbins 1983). me to the importance of water in an animal's diet, moisture content of a forage is often a major determinant of plant palatability and therefore animal selection (Reynolds 1967, Radwan and Crouch 1974, Stoddart et a1. 1975). Thinning of overstory red pine appeared to have a generally beneficial effect on moisture contents of the forages sampled in this 73 study, particularly during summer. Five (black cherry leaves, red maple leaves, aspen twigs and leaves, and brambles) of the 10 wildlife forages sampled in Summer 1988 had higher moisture contents in thinned plots versus controls or in 16.1 mz/ha (7O ft2/acre) thins versus 25.3 mZ/ha (110 ft2/acre) thins. mm (1986), similarly, found moisture contents of forages to be increased by overstory lodgepole pine (P. contorta) thinning. Among Winter 1988 forage samples, however, overstory thinning had no significant effect on forage moisture content, nor did time since thinning appear to have an impact (with the exception of brambles). The increased moisture contents observed are likely a result of the individual plant supplying increased water quantities, particularly to the leaves, to corpensate for increased evapotranspiration and to facilitate increased metabolic activity in response to the increased radiant erergy availability (Barbour et al. 1980) . Modification of the overall stand hydrologic regime due to the removal of a percentage of the overstory and subsequent increased moisture availability (due to decreased overstory uptake and foliar interception) also undoubtedly played a role in increased understory moisture contents. Hydrologically, the proximate effect of overstory red pine thinning therefore appears to be increased moisture contents of understory species, particularly herbawous species and herbaceous parts of woody species. The ultimate result is a contribution towards increased wildlife use of the thinned stands due to more palatable forages, since moisture content plays a major role in forage selection (Stoddart et al. 1975) . The significantly greater deer and elk use of the thinned stands documented in this study may be at least partially attributable to 74 greater forage palatability due to increased moisture contents. Ether Extract Fatsareessentialelergysmroesinthedietofananimal, contributing 2.25X the metabolizable erergy of proteins or carbohydrates (Nagy and Haufler 1980). In addition to this energy factor, fats are also important in affecting the palatability of forages, supplying essential fatty acids, and functioning in the transport of fat soluble vitamins (campa 1982) . A complicating factor in the analysis of ether extract (EE) data is the lack of specificity of ether, or ether- methanol, as a solvent. In addition to lipids, numerous other common plant substances are soluble in ether and thus are extracted during distillation; among these are glycerides, phospholipids, sterols, pigments, waxes, volatile oils, and resins (Robbins 1983). Thus, the extracted EE values consist of not only lipids, but numerous other substances as well; hence, the common referral to E values as crude fat. me to these nonhomogeneous mixtures, however, nonspecific lipid extractions often have little nutritional meaning (Robbins 1983) . Thinning treatments had little effect on crude fat levels in this study. In the 2 instances of significant differences among treatments and controls, one (winter black cherry samples) showed highest crude fat levels in the controls, while the other (summer aspen twigs) showed highest crude fat content in the 16.1 mZ/ha (70 ftz/acre) thins. Similarly, age from thinning had no significant effect on crude fat values. other studies that have investigated nutritional responses to thinning overstory pines have not reported changes in crude fat levels 75 either (Wolters 1973, Blair and Brunrett 1977, Crouch 1986) . Thinning of the overstory red pire appeared to have very little effect on the ash content of the sampled understory forages. Only summer beech twigs and summer black cherry leaves differed significantly in ash content; for beech twigs, greatest ash contents were found in samples from the 25.3 mZ/ha (110 ftZ/acre) thins (2.20%), which were significantly higher than those from 16.1 mZ/ha (70 ftz/acre) thins (1.62%) but not from controls (2.04%). Ash content of black cherry leaves from the controls were significantly greater than from either of the 2 thinning densities. Ash as a whole comprised very little of any forage from any study plot on a dry weight basis; the maximum ash content of any forage was 6.79% from black cherry leaves from the controls. Other studies which commented on nutritional response to thinning overstory pines (Wolters 1973, Blair and Brunnett 1977, Crouch 1986) similarly did not report changes in ash contents, although ash content frequently increases with burning of pine stands (lay 1957, Leege 1969) . The utility of ash determination by itself is questionable, as the ash content of a forage tells nothing about what specific elements are present and in what quantities (Maynard et al. 1979, Nagy and Haufler 1980, Robbins 1983) . Although high ash values are generally considered to increase nutritional quality and palatability (Dietz 1958 as cited by Nagy and Haufler 1980) , the specific makeup of the ash could prove just the opposite; for example, if a significant portion of an increased ash 76 content is composed of a mineral such as silica, such as frequently happens in grasses following fire (Barbour et al. 1980) , then a high ash content would be an indication of lower nutritional quality, considering silica's role in herbivory defenses. As previously noted, the ash content is carprised of all inorganic elements in a sample. Since any one mineral comprises only a minute fraction of a plant's carposition, even highly significant increases in one or several minerals may not appreciably affect total ash content (C‘ampa 1982) . rIherefore, the lack of response of ash levels to red pine overstory thinning should not be unexpected, and to properly assess any nutritional benefits on a mineral basis as a result of overstory thinning would require individual micro and macro nutrient analysis, combined with a comparison to individual wildlife species elemental reeds and existing range deficiencies. Based on the nutritional analyses conducted, it appears that the understory nutritional quality responses to thinning the overstory red pine were not dramatic. 'Ihis relative lack of response is not surprising, however. The original objective of this facet of this study was to evaluate the forage nutritional responses to both thinning and prescribed burning. Based on the literature from the South (Lay 1957, 1967; Halls 1970, Dills 1970) and West (Leege 1969, Leege and Hickey 1971) , it was originally hypothesized that prescribed burning would have the most dramatic impact on forage nutritional quality. The undramatic nutritional responses to thinning alone then should not be taken as an indication that the understories of red pine stands 77 cannot be enhanced in terms of forage nutritional quality for elk and deer. Additionally, nutritional benefits from overstory thinning (or any other management) may be severely underestimated if nutritional evaluation is based solely on chemical nutritional parameters and not on changes in wildlife diets and forage accessability as well (Hobbs and Spowart 1984). Management such as overstory thinning may result in nutritionally enhancing effects such as earlier plant green-up, increased productivity of preferred forages, and removal of standing dead herbage and obstructions to browsing. While none of these effects may change any chemical nutritional parameters of a forage at all, they do improve the nutritional quality of the stand as a whole. Any future nutritional evaluation of the red pine stands should take into account these additional enhancements, and not just focus on the chemical constituents of the forages. SMAILDW’JMAIS Smallmammalnumbersardspecies richness tendedtobelcwonthe study plots, due primarily to both small plot size (2 ha) and the tendency of red pine to support low populations (Gysel 1966). Initial populations and diversity did not significantly differ between pretreatment 25.3 mZ/ha (110 ftz/acre) plots and controls in July 1987, although 6 species were captured on the treatment plots compared to only 2 on the controls. Of the 6 species captured on the 25.3 mZ/ha (110 ftz/acre) plots, however, 3 were represented only by 1 individual; red-backed vole, masked shrew, and red squirrel. Eastern chipmmks and 78 Perumyscus mice were the predamirant species on both sites. The 16.1 mZ/ha (7o ftZ/acre) plots were undergoing thinning during 1987, and were thus unable to be evaluated for baseline data. The August 1987 trapping period showed the pretreatment 25.3 mZ/ha (110 ft2/acre) plots to be significantly higher in small mammal diversity and total populations. Jmeing mice, not present in July, appearedontheshfiyplotsandwerethemestcamonspeciescapturedon both pretreatment 25.3 m2/ha (110 ftZ/acre) and control plots. The differencesinsmallmammalmmmberswereduetoincreasedabtmdanceof the July community plus the addition of flying squirrels and jtmping mice to the pretreatment 25.3 mz/ha (110 ftz/acre) plots. The addition of jumping mice and the masked shrew (1 individual) to the controls did not cverocme file loss of eastern dlipmurms; thus total small mammal numbers for the controls did not change, while the pretreatment 25.3 mZ/ha (110 ftz/acre) numbers more than doubled. Following thinning, plots thinned to 16.1 m2/ha (7o ftZ/acre) tended to have higher total small mammal numbers and species richness, although these differences were statistically significant only in July 1988. Increased total numbers were primarily due to greater numbers of red-backed voles and eastern chipmmms on the 16.1 mZ/ha (70 ftZ/acre) plots. Vole populations in general are highly related to the presence of subclimax cammmities and the abundance of herbaceous grasslike cover, especially in moist areas (DeGraaf and Rudis 1986). Opening of the red pine canopy resulted in increased understory development, mimicking subclimax situations, and increased frequency for many grasses in the thinned plots relative to the controls. Bromegrass, western 79 wheatgrass, and feswegrassallincreasedinabundancewiththirming, creating more favorable vole habitat. Additionally, increased slash due to the thinning operations also helped meet sore of the special habitat requirements of red—backed voles (DeGraaf and Rudis 1986) . It islikelythattheincreaseinvolepqlulations inthethinnedstandsis at least partially attributable to these factors. 'Iheirxzreasedmmbersofeasterndlipmlnksmthimedplots relative to controls is less easily understood. Increased slash levels as a result of the thinning operations may have provided chipmunks with increased cover, decreasing vulnerability to predation. Studies on clearcutting (Kirkland 1977, Beyer 1983) showed chipmunks to be excluded from clearcuts despite the increased slash levels. This exclusion, however, is likely due to the total loss of aerial cover, which is important to chipmunk habitat, probably to decrease aerial predation and provide mast (DeGraaf and Rudis 1986). Both 16.1 mZ/ha (7o ftZ/acre) and 25.3 mZ/ha (110 ftZ/acre) thins, however, maintained abundant overstories. 'Ihinning, therefore, is unlikely to expose chipmunks to increased aerial predation or loss of mast. It is likely, then, that increased cover and increased food supplies provided by the more abundant understories associated with the thinned red pine plots was at least partially responsible for the increase in eastern chipmunk numbers. Comparisons of 1 and 2-year-old 16.1 n12/l'1a (70 ftZ/acre) thinnings allwedatemporal factortobeevaluatedinregardtosmallmammal populations, i.e. would population parameters change with increased time since overstory thinning? Cotparisons of these different-aged plots 80 made the assumption that all environmental factors affecting the plots are similar except for age since thinning. lacking baseline data for the 16.1 mZ/ha (7o ftZ/acre) thins, it is impossible to know whether initial floral and faunal corpositions were similar. 'Ihus, although it mustbeassumedthat all factors otherfllanagesincethimingwere constant among the 16.1 mZ/ha (7o ftZ/acre) plots, the validity of this assmtption is questionable, especially in light of the different small mammal communities shown between pretreatment 25.3 mZ/ha (110 ftZ/acre) and control plots in August 1987. One and 2-year—old 16.1 m12/ha (7o ftz/acre) thins tended to be similar in small mammal numbers and corposition. The 2-year-old thins tended to have higher numbers of eastern chipmunks, probably as a result of the significantly increased understory development. One-year—old 16.1 m2/ha (7o ftZ/acre) thins tended to have higher red-backed vole populations. Since the differentially aged 16.1 mz/ha (7O ftZ/acre) thins did not differ in grassy cover nor in the apparent amount of slash present, both habitat keys for red-backed voles, the reasons for this difference are obscure. It is possible that slrrounding cover, as it influences colonization ability, may have played a role. Additionally, without baseline data to indicate otherwise, it is possible that red-backed vole populations differed initially among the 16.1 m2/ha (7o ftz/acre) plots. One factor that made evaluation of the stall mammal data more difficult was the 1988 drought. ‘Ibtal small mammal numbers increased in both treatments and controls from July to August 1987; as many of the 81 individuals captured in August were juveniles, this presumably VBS the result of a reproductive flush into the community. These overall increases on all plots were not present in 1988; 25.3 mZ/ha (110 ftZ/acre) thins and controls increased slightly in overall small mammal numbers, while 16.1 mZ/ha (7o ftZ/acre) plots dropped substantially from July to August. Additionally, total stall mammal numbers in both July and August 1988 were down substantially from 1987 levels for both 25.3 mZ/ha (110 ftZ/acre) thins and controls (16.1 m2/ha (7o ftZ/acre) plots were not trapped in 1987) . What effect the drought had on this decrease is speculative, although it is reasonable to assume that the record drought conditions contributed to the small mammal declines. Reproductive recruitment, an important factor in the population increases observed in 1987, is especially vulnerable to resource limitations (Ricklefs 1979) . 'Ihus, it is possible that charges in stall mammal numbers, diversity, or species richness may have been masked due to drought-induced low population numbers for 1988. ELK AND DEER RESPONSES Elk and deer use increased progressively with overstory red pine thinning, eventually resulting in 3 significantly distinct levels of usage; 16.1 mZ/ha (7o ftZ/acre) thins > 25.3 mZ/ha (110 ftZ/acre) thins > controls in 1989. Although Spring 1988 browse surveys did not show the significantly different levels of utilization shown by the Spring 1988 and 1989 pellet-group counts, corparisons of the l and 2-year—old 16.1 mz/ha (70 ftZ/acre) thirmings suggested that differences in browse 82 utilization may become more apparent with increasing time since thinning. Additionally, the differences in utilization indicated by pellet group-counts also became more prommced with time. A potential problem in evaluating elk and deer use of the study plots involves the biases and limitations associated with each technique used to assess use in this study. 'Ihe pellet-group technique especially has core under close scrutiny lately, primarily as a result of the questionable validity of the necessary assmrptions of this method (Collins and Unless 1981). Necessary assumptions of pellet-group analysis include (Neff 1968): (1) No daserver bias in seeing groups (2) No observer bias in accurately aging or species identification of groups ( 3) No loss of groups Additionally, if pellet-groups are to be used for absolute density estimates or to estimate absolute use of an area, the following assmrptions are also necessary: (4) Pellet-group mmbers linearly related to time spent in an area (5) Pellet-group density linearly related to population size (6) Constant species specific defecation rates The first two assmmptions above can be easily accounted for with stringent sampling methodology such as was employed in this study; the same observer for all transects and permanent regularly cleared transects. The third assmption can be modified into the assumption of constantlossofgroupsamongstudyplots ifgroupcountsaretobeused only for relative comparisons. This was the case in this study, and the 83 modified assumption is not nearly as biologically ugly as the original. Controversy exists over the validity of the final 3 assumptions. Collins and Urness (1981) concluded that pellet-group counts did not accurately reflect mule deer habitat use in Utah as determined by visual observation. Ieopold et al. (1984), however, noted that the evaluation of Collins and Urness (1981) was based on calculating and cotparing absolute pellet-group densities with habitat use rather than using the relative magnitude of groups in each habitat unit to simply rank use. Leopold et al. (1984) reevaluated the Utah data and showed that relative ranking of the pellet-group counts did accurately rank habitat use as determined by visual observation. The authors felt that pellet-group counts could accurately reflect relative habitat use as long as biologists stayed within the relative ranking boundary; any attempts to determine absolute use by combining group density determinations with species specific defecation rates go beyond the capabilities of the technique. Siumilar results have subsequently been reported by loft and Kie (1988), who determined that pellet-group coints were accurate in ranking relative use of habitats as determined by radio telstetry. This study attempted to use pellet-group counts only to rank relative habitat use. For this degree of resolution, the pellet-group technique appears to be valid. The only questionable assumption made for this purpose is that group counts are linearly related to time spent in an area. Although questionable, the studies mentioned above appear to validate this assumption, at least to the degree of resolution necessary for simple relative habitat use rankings. Biases are also present in the use of browse utilization to assess 84 relative habitat use. Chief among these are the selection of key browse species (does high use of a highly preferred or "ice-cream species" indicate an overall high level of use for the entire study area?) and, if browse utilization is expressed as an absolute density (browsed stems per hectare, etc), the tendency of utilization surveys to merely reflect the relative abundaroe of a species rather than comparable degree of use (E. Willard, pers. com.). The methodology used in this study accounted for these common biases by, firstly, selecting 3 key species for utilization determination of high, medium, and low preference (red maple, black cherry, and beech, respectively) and, secondly, by assessing utilization as relative use per individual, rather than absoluteuse perarea. Basedon the above, it appears thatthe increased elkanddeeruse documented in this study is valid, and not the result of methodological biases. Increased elk and deer utilization as seen in this study was probably the result of 3 factors: (1) Increased quantity of forage (2) Increased quality of forages (3) Increased hiding cover Increasedungulateuseofthinnedpinestandshasbeenwell documented int-heliterature. Increaseddeeruseofthinnedsouthernpinestands is due primarily to increased quantity and quality of forages (Blair 1967, Halls 1970, Halls 1973, Blair and Enghardt 1976, Blair and Brunnett 1977, Halls and Boyd 1982, Hurst and Warren 1982). Other cotplimentary results to this study have been documented in the West for 85 elk (Pearson 1968, Clary and Iarson 1971, Croudl 1986) and deer (Reynolds 1969, 'Ihill et al. 1983), with the increased use also attributed to increased forage production. Philleo et al. (1978) have presented complimentary results from New Hampshire pire forests, as have Krefting and Phillips (1970) and Telfer (1978) for midwestern swamp conifer stands. Dickmann et al. (1987) has reported heavy ungulate use forthirmedredpinestandsinthesameMichiganstate forestthatthis study was conducted in. 'Ihus, it appears well documented that deer and elk show increased use of thinned pine stands, a conclusion supported by the results of this study. Increased woody stem densities, particularly of preferred browse species such as red maple, white ash, aspen, and beaked hazelnut, undoubtedly contributed to the increased ungulate use by providing increased quantities of food. Additionally, altholgh not documented in this study, herbaceous production on the thinned study plots, especially the 16.1 mZ/ha (70 ftZ/acre) plots, was markedly visibly greater than on the controls. These increases in preferred browse species and overall production were at least partially responsible for the significantly greaterelkarodeeruseofthethiumedredpinestands. In addition to increased understory densities, certain nutritional characteristics were influenced by overstory red pine thiming, primarily the moisture content of forages. Moisture content has been shown to be a significant factor in determining plant palatability (Reynolds 1967, Radwan and Crouch 1974, Stoddart et al. 1975). ’Iherefore it is likely that the increased palatability of the forages in thinned stands also contributed to increased elk and deer utilization, 86 as ungulates are more likely to select areas of increased forage nutritional quality. Finally, the increased understory development associated with the thinned study plots likely contributed to greater hiding cover for elk and especially deer. Although cover parameters were not evaluated as partofthisstudy,theunderstoryo1thethinredstudyplots, especially the 2-year-old 16.1 mZ/ha (70 ftZ/acre) thins, was extremely dense in patches. Deer were frequently flushed out of these stands at close distances without being previously observed. In contrast, despite the denser overstory red pine stocking, elk and deer were far more conspicuous in the control plots, due to the absence of understory development. Adequate hiding cover is critical in the ecology of deer and elk, perhaps because it provides the animals with a heightened sense of security ('Ihoras et al. 1979, Lyon et al. 1985). Although probably not contributing to the increased deer and elk utilization of the thinned plots to the magnitude as the increased quantity and quality of forage, the increased quality of the thinned stands as hiding cover still likely contributed to these stands attractiveness to ungulates. Working in antithesis of increased hiding cover quality, however, is a decreaseinthequalityofthethenmalcoverassociatedwiththestaros withtheopeningofthecanopy. ElkanddeerintheHKBFtendto select mixed conifer or cedar swamps as thermal cover during critical periods such as hard winters (Beyer 1987). Therefore, little elk or deer use is probably sacrifiwd due to the decreased thermal cover attributes associated with the thinned red pine stands. SUMMARY AND WW5 Although preliminary, the results of this study indicated that thirmingredpineresultedintrsrisofiroreaseduroerstorywoodystsm densities, which became more pronounced with time. The increased stem densities observed were not the result of invasion of new species, but rather an increase in the densities of species already present on the study plots. Understory species which contributed the most to this overall increase were red maple, white ash, aspen, and beaked hazelnut; all are important wildlife forage species. This increase in understory density, particularly of preferred browses, contributed to the thinned stands increased attractiveness to wildlife. Overstory red pine thinning did not result in any significant species shifts in the understory. The inherent high quality of the sites (red pine site indices ranged from 60 to 76) may have mediated any corpositional shifts, as higher quality sites tend to support a richer and more diverse flora than corparable lower quality sites. Additionally, the drought conditions of 1988 may have masked any coxpositional shifts. Finally, insufficient time since thinning may have passed to produce any compositional shifts. Vegetative changes in response to overstory red pine thinning thus manifested themselves as the aforementioned increased woody stem densities, increased woody and herbaceous productivities, and slightly enhanced nutritional qualities of ungulate browses, particularly forage moisture contents. All of the 87 88 above results are at least partially attributable to enhanced understory water regimes, increased radiant energy availability, and enhanced nutrient availability and cycling rates. The lack of response of certain nutritional parameters, particularly protein levels, may be due to low overall nitrogen availability; with increasing time from thinning, however, nitrogen may becote increasingly available in the stands as high quality deciduous understory litter stimulates rates of decomposition and thus frees nitrogen immobilized in the poor quality (high carbonznitrogen ratio) conifer litter. Stallmammalmmmberstendedtoincreasewithincreasingdegreeof overstory thinning, although differences were statistically significant only during the July 1988 trap period. The drought conditions of 1988 may have lessened the stall mammal (as well as the vegetative) responses. No significant reproductive flush, evident in August 1987, was seen in August 1988. Individually, red-backed voles and eastern chipmunks showed the greatest responses, with both becoming significantly more abundant on the thinned plots. Both presumably responded to increased understory development and dead-and-down woody material. Elkanddeeruseincreasedwithoverstoryredpinethinning, eventually resulting in three distinct levels of utilization: 16.1 mz/ha (7o ftZ/acre) thins > 25.3 mz/ha (110 ftZ/acre) thins > controls. Increased elk and deer utilization was likely the result of increased quantity and quality of preferred forages, as well as enhanced quality of the thinned plots in terms of hiding cover. With continued understory development, the thinned stands are likely to bwote even 89 more favorable for elk and deer in the future. Additionally, if the prescribed burning aspect of this study, urable to be cotpleted initially due to the 1988 drought, is acootplished in the future, increased production and forage nutritional quality associated with prescribed bunting will likely result in even greater elk and deer utilization. It is thus evident that thinning overstory red pine can result in enhanced habitat quality and use of red pine stands by wildlife, particularly big game. 'Ihe degree of utilization increased with the degree of overstory thinning; 16.1 mz/ha (7O ftZ/acre) thins are significantly more favorable than 25.3 mZ/ha (no ftZ/acre) thins. Mature red pine stands are currently considered "biological deserts" with regard to wildlife, especially game animals, and often left unmanaged for wildlife due to this reputation. ‘Ihe results of this study, although preliuminary, provide quantitative evidence that red pine stands can be improved as wildlife habitat, especially for significant featured species such as elk and white-tailed deer. Thinning responses, however, are highly related to site quality. This study was conducted on high quality sites where thinning responses should be optimal; lower quality sites likely will not show the degree of response seen in this study. In conclusion, although this study did not cotpare levels of utilization of the thinned red pine stands with utilization rates of other forest types in the area, any degree of habitat improvement in red pine should be considered significant in lieu of the large acreages currently in red pine that are essentially unmanaged for wildlife, as 90 well as the significant acreages either currently being converted to red pine or scheduled for conversion. Additionally, prescribed burning of thinned stands will likely result in even more enhanced wildlife attributes. Finally, timber production levels associated with the thinned stands will be monitored by the Michigan State University Department of Forestry. If timber production can be maintained on the thinnedplots, aspastresearohsuggests, thenthinningofoverstoryred pine represents a practical and economically satisfactory means of improving the wildlife habitat quality of red pine stands. LITERATURECITED Alban, D.H. 1977. Influence on soil properties of prescribed burning under mature red pine. U.S.D.A. For. Ser. Res. Pap. NC-139. 8pp. Anderson, R.C., O.L. Ioucks, and A.M. Swain. 1969. Herbaceous response to canopy cover, light intensity, and throughfall precipitation in coniferous forests. Ecology 50:255-263. A.O.A.C. 1975. Official methods of analysis of the association of official analytical chemists. 12th ed. Assoc. Off. Aral. Chem. 1094pp. Barbour, M.G., J.H. Burk, and W.D. Pitts. 1980. Terrestrial plant ecology. The Benjamin/Cummings Publishing Co., Menlo Park, CA. 604pp. Barrett, J.W. 1968. Response of ponderosa pine pole stands to thinning. U.S.D.A. For. Ser. Res. Note PNW-77. llpp. Benzie, J.W. 1973. Red pine. Pages 58-61 Q1 Silvicultural systems for the major forest types in the United States. U.S.D.A. Agric. Handb. No. 445. ll4pp. 1977. Manager's handbook for red pine in the north central states. U.S.D.A. For. Ser. Gen. Tech. Rept. NC-33. 22pp. , and J.E. warmer. 1983. Red pine. Pages 89-91 i_n R.M. Burns (coup) . Silvicultural systets for the major forest types in the United States. U.S.D.A. Agric. Hardb. No. 445. Beyer, D.E. 1983. Wildlife response to whole-tree harvesting of aspen. M.S. Thesis. Michigan State Univ., East Iansing. 76pp. . 1987. Population and habitat management of elk in Michigan. PhD Dissertation. Michigan State Univ. , East Iansing. 148pp. Blair, R.M. 1967. Deer forage in a loblolly pine plantation. J. Wildl. Manage. 31:432-437. , and LE. Brunett. 1977. Deer habitat potential of pine-hardwood forests in Louisiana. U.S.D.A. For. Ser. Res. Pap. SO-136. llpp. , and H.G. Enghardt. 1976. Deer forage and overstory dynamics in a loblolly pine plantation. J. Range Manage. 29:104-108. 91 92 Buckttan, R.E. 1964. Effects of prescribed burning on hazel in Minnesota. Ecology 45:626-629. Campa, H., III. 1982. Nutritional response of wildlife forages to municipal sludge application. M.S. Thesis. Michigan State Univ., East Iansing. 88pp. . 1988. Personal communication. Michigan Dep. Nat. Res., Rose lake, MI. Caveney, N. 1989. Personal communication. Michigan Dep. Nat. Res., Vanderbilt, m. Clary, W.P., and F.R. larson. 1971. Elk and deer use are related to food sources in Arizona ponderosa pine. U.S.D.A. For. Ser. Res. Note RM— 202. 4pp. _, and P. F. Ffolliott. 1966. Differences in herbage-timber relationships between thinned and unthinned ponderosa pine stands. U. S. D.A. For. Ser. Res. Note RM-74. 4pp. Collirflgwood, G. H., and W. D. Brush. 1974. Knowing your trees. American Forestry Washington D. c. 374pp. Collins, W.B., and P.J. Urness. 1981. Habitat preferences of mule deer as rated by pellet-group distributions. J. Wildl. Manage. 45:969- 972. Crouch, G.L. 1986. Effects of thinning pole-sized lodgepole pine on understory vegetation and large herbivore activity in central Colorado. U.S.D.A. For. Ser. Res. Pap. I‘M—268. lopp. DeGraaf, R.M., and D.D.Rudis. 1986. New England wildlife: Habitat, natural history, and distribution. U.S.D.A. For. Ser. Gen. Tech. Rep. NE-108. 491pp. Dickmann, D.I. 1989. Personal communication. Dept. of Forestry, Michigan State Univ., Fast Iansing. _, W. J. O'Neill, and N. Caveney.1987.Wide-spaced red pine. a _multiple use opportunity. Nor. J. Appl. For. 4. 44—45. Dietz, D.R. 1958. Seasonal variation in the nutritive content of five southwestern Colorado deer browse species. M.S. Thesis. Colorado State Univ., Ft. Collins. 110pp. Dills, G.G. 1970. Effects of prescribed burning on deer browse. J. Wildl. Manage. 34:540-545. Eyre, F.H., and P. Zehngraff. 1948. Red pine managetent in Minnesota. U.S.D.A. Agric. Circ. No. 778. 70pp 93 Fontenot, J .P. 1971. Nitrogen metabolism and nutrition in ruminants. Pages 575-599 i_n D.C. Church (ed). Digestive physiology and nutrition of rumirants. Vol. 2. Oregon State Univ. Press, Corvallis. 801pp. Fowells, H.A. 1965. Red pine. Pages 432-446 i_n H.A. Fowells (cotp). Silvies of the forest trees of the United States. U.S.D.A. Agric. Hardb. NO. 271. Freese, F. 1978. Eletentary forest sampling. U.S.D.A. Agric. Handb. No. 232. 91pp. Grisez, T'.J. 1968. Growth and development of older plantations in northwest Pennsylvania. U.S.D.A. For. Ser. Res. Pap. NE-104. Gysel, L.W. 1966. Ecology of a red pine plantation in Michigan. Ecology 47:465-472. , and L. J. Lyon. 1980. Habitat analysis and evaluation. Pages 305- 328 in S. D. Schemitz (ed). Wildlife managetent techniques manual. The Wildlife Society. 686pp. Halls, L.K. 1970. Growing deer food amidst southern timber. J. Range Manage. 23:213-215. . 1973. Managing deer habitat in loblolly-shortleaf pine forest. J. For. 71:752-757. , and C.E. Boyd. 1982. Influence of managed pine stands and mixed pine/hardwood stands on well being of deer. U.S.D.A. For. Ser. Res. Pap. SO-183. 18pp. Hobbs, N.T., and R.A. Spowart. 1984. Effects of prescribed fire on nutrition of mountain sheep and mtule deer during winter and spring. J. Wildl. Manage. 48:551-560. Hurst, G.A., and R.C. Warren. 1982. Impacts of Silvicultural practices in loblolly pine plantations on white-tailed deer habitat. Pages 484-487 ' E. James Jr. (ed). Proceedings of the second biennial southern Silvicultural research confereuce. Nov. 4-5, 1982. Atlanta. Keay, J. A., and J. M. Poek.1980. Relationships between fire and winter habitat of deer in Idaho. J. Wildl. Manage. 44. 372-380. Kennedy, P.C., and LP. Wilson. 1971. Understory vegetation associated with Saratoga spittlebug damage in Michigan red pine plantations. Can. Entan. 103:1421-1426. Kirkland, G.L., Jr. 1977. Responses of small mammals to the clearcutting of northern Appalachian forests. J. Mammal. 58:600-609. 94 Krebs, C. J. 1966. Detographic changes in fluctuating populations of Mi—crotus californicus. Fool. Monogr. 36: 239-273. Krefting, L.W., and R.L. Phillips. 1970. Improving deer habitat in upper Michigan by cutting mixed-conifer swamps. J. For. 68:701-704. Kruse, W.B., 1972. Effects of wildfire on elk and deer use of a ponderosa pine forest. U.S.D.A. For. Ser. Res. Note lad-226. 4pp. lay, D.W. 1956. Effects of prescribed burning on forage and mast production in southern pine forests. J. For. 54:582-584. . 1957. Browse quality and the effects of prescribed burning in southern pire forests. J. For. 55:342-347. . 1967. Browse palatability and the effects of prescribed burning in sorU'uern pine forests. J. For. 65:826-828. leege, T.A. 1969. Burning seral brush ranges for big game in northern Idaho. Trans. North American Wildl. Conf. 429-438. , and W. O. Hickey. 1971. Sprouting of northern Idaho shrubs after prescribed burning. J. Wildl. Manage. 35: 508-515. leopold, B.D., P.R. Krausman, and J.J. Hervert. 1984. The pellet-group census technique as an indicator of relative habitat use. Wildl. Soc. Bull. 12:325-326. loft, E.R., and J.G. Kie. 1988. Comparison of pellet-group and radio triangulation methods for assessing deer habitat use. J. Wildl. Manage. 52:524-527. lothner, D.C., and D.P. Bradley. 1984. A new look at red pine financial returns in the lake States. U.S.D.A. For. Ser. Res. Pap. NC-246. 4pp. lowe, P. 0., P. F. Ffolliott, J. H. Dieterich, and D. R. Patton. 1978. Determining potential wildlife beefits from wildfire in Arizona ponderosa pine forests. U. S. D.A. For. Ser. Gen. Tech. Rept. RM-52. 12pp. qundgreu, A.L. 1983. New site productivity estimates for red pine in the lake States. J.For. 81:714-717. Lyon, L.J., T.L. lonner, J.P. Weigrand, C.L. Marcum, W.D. Edge, J.P. Jones, D.W. NbCleery, and L.L. Hicks. 1985. Coordinating elk and tiumber managetent: Final report of the Montana Cooperative Elk- Iogging Study. Mont. Dept. Fish, Wildl., and Parks. Bozeman, MT. 53pp. 95 MacArthur, R.H. 1972. Geographical ecology: patterns in the distribution of species. Harper and Row, New York. 269pp. Maynard, L.A., J.K. loosli, H.F. Hintz, arud R.G. Warner. 1979. Animal nutrition. FbGraw-Hill, New York. 602pp. McConnell, B. R., and J. G. Smith. 1970. Response of understory vegetation to understory thinning in easternWashington. J. Range Manage. 23:208-212. Miller, H. G., J. M. Cooper, J. D. Miller, and O. Pauline. 1979. Nutrient cycles in pine and their adaptation to poor soils. Can. J. For. Res. 9: 19—26. Moran, R.J. 1973. The Rocky Mountain elk in Michigan. Res. and Dev. Rep. No. 267. Mich. Dept. Nat. Res., lensing, MI. 93pp. Murphy, D.A., and J.A. Coates. 1966. Effects of dietary protein on deer. Trans. N. Am. Wildl. Conf. 31:129-139. Nagy, J.G., and J.B. Haufler. 1980. Wildlife nutrition. Pages 129-142 i_n S.D. Schemitz, ed. Wildlife maragetent techniques manual. The Wildlife Society, Washington, D.C. 577pp. Neff, D. J. 1968. The pellet-group count technique for big game trend, census, and distribution: a review. J. Wildl. Marage. 32: 597-614. NCAA. 1987. National Oceanic and Atmospheric Administration. Climatological Data, Annual Summary, Michigan. Vol. 102. . 1988. National Oceanic and Atmospheric Aduministration. Cliumatological Data, Annual Summary, Michigan. Vol. 103. Oswald, B.P., and W.W. Covington. 1983. Changes in understory production following a wildfire in southwestern ponderosa pine. J. Range Marage. 36:507-509. Pearson, H.A. 1968. Thinning, clearcutting, and reseedu'ng effect deer and elk use of ponderosa pine forests in Arizona. U.S.D.A. For. Ser. Res. Note RM-119. 4pp. Philleo, B., J. B. Gavanagh, and D. P. Olson. 1978. Browse utilization by deer in relation to cutting and prescribed burning in southeastern New Hampshire. Trans. Northeast Sec. Wildl. Soc. 35:16-26. Pregitzer, K. Sm1988. Persoral communication. Michigan State Univ. , Fast lansing,MI . Radwan, M.A., and G.L. Crouch. 1974. Plant characteristics related to feeding preference by black-tailed deer. J. Wildl. Manage. 38:32- 41. 96 Reynolds, H.G. 1967. Chemical constituents and deer use of sore crown sprouts in Arizona daparral. J. For. 65:905-908. . 1969. Improvement of m habitat on southwestern forest lands. J. For. 67:803—805. Ricklefs, R.E. 1979. Ecology. Cniron Press, New York. 966pp Robbins, C.T. 1983. Wildlife feeding and nutrition. Academic Press, New York. 343pp. Roberts, T.A., and R.L. Tiller. 1985. Mtule deer and cattle responses to a prescribed burn. Wildl. Soc. Bull. 13:248-252. Rogers, L.L., J.J. Mooty, and D. Dawson. 1981. Foods of white-tailed deer in the upper Great lakes region- a review. U.S.D.A. For. Ser. Gen. Tech. Rep. NC-65. 24pp. Roppe, J.P., and D. Hein. 1978. Effects of fire on wildlife in a lodgepole pine forest. Southwestern Nat. 23:279-288. Ross, B. A., J. R. Bray, and W. H. Marshall. 1970. Effects of long—term deer exclusion on a Pinus resinosa forest in north-central Minnesota. Ecology 51: 1088- 1093. Short, H.L. 1981. Nutrition and metabolism. Pages 99-127 i_n O.C. Wallmo (ed). Mule and black-tailed deer of North America. Univ. of Nebraska Press, Lincoln. 605pp. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw Hill, NY. 312pp. Summers, L.M. 1977. Atlas of Michigan. Mich. St. Univ. Press, East lansing, Ml. 242pp. Spiegel, L.E., C. H. Huntly, and G. R. Gerber. 1963. A study of the effects on elk browsing on woody plant succession in northern Michigan. The Jack Pine Warbler 41: 68-72. Stoddart, L.A., A.D. Smith, and T.W. Box. 1975. Range maragetent. MoGraw-Hill, New York. 532pp. Tappeiner, J.C., and A.A. Alum. 1975. Undergrowth vegetation effects on the nutrient content of litterfall and soils in red pine and birch stands in northern Minnesota. Ecology 56:1193-1200. Telfer, E.S. 1978. Silviculture in the eastern deer yards. For. Chron. 54:203-208. Thill, R.E., P.F. Ffolliott, and D.R. Patton. 1983. Deer arud elk forage production in Arizona mixed conifer forests. U.S.D.A. For. Ser. Res. Pap. fill-248. 13pp. 97 Thotas, J. W. 1979. Introduction. Pages 10-211n J. W. Thomas (ed). Wildlife habitats in managed forests. the Blue Mountains of Oregon and Washington. U.S.D.A. For. Ser. Agric. Handb. No. 553. 512pp. , H. Black, R.J. Scherzinger, and R.J. Pedersen. 1979. Deer and elk. Pages 104-127 i_n J.W. Thotas (ed). Wildlife habitats in managed forests: the Blue metains of Oregon and Washington. U.S.D.A. For. Ser. Agric. Harrib. No. 553. 512pp. Tobiessen, P., and M.B. Werner. 1980. Hardwood seedling survival under plantations of scotch pine and red pine in central New York. Ecology 61:25-29. Van Wagner, C. E. 1963. Prescribed burning iments: red and white pine. Can. Dept. For. Pub. No. 1020. 27pp. . 1970. Fire and red pine. Proc. Tall Timbers Fire Ecology Conf. 10:211-219. Venue, L.J., and D.E. Ullrey. 1972. Feeding and nutrition of deer. Pages 275-291 i_n D.C. Church (ed). Digestive physiology and nutrition of rumiu'ants. Vol. 3. Oregon State Univ. Press, Corvallis. 350pp. von Althen, F. W., W. M. Stiell, and R. B. Forster. 1978. Effects of four on the growth, yields, and financial returns of a 62-year-old red pine plantation. For. Chron. 54: 253-260. Willard, E.E. 1985. Personal communication. Univ. of Montana, Missoula, MI‘. Wolters, G.L. 1973. Southern pine overstories influence herbage quality. J. Range Manage. 26:423-426. . 1982. Icngleaf and slash pine litter alters herbage composition. J. Range Manage. 35:761-763. , and R.L. Sclmmidtling.1975. Browse and herbage in intensively naraged pine plantations. J. Wildl. Marage. 39:557-562. _, A. Martin, and H. A. Pearson. 1982. Forage response to overstory —reduction on loblolly-shortleaf pine-hardwood forest range. J. Range Manage. 35:443-446. wyoming Game and Fish. 1982. Handbook of biological techniques. Wyoming Game and Fish Dept., Cheyenne. 442pp. Table A-1. Species of vegetation found on the study plots in 1987 and 1988. COMMON NAME SCIENTIFIC NAME HERBACEDUS Alum root Heuchera richardsonii Amica ma Spp Bedstraw m boreale Beggar's tick m frondosa Bluegrass ia pratensis Bracken fern Pteridium aggilintmm Brote grass ms spp Bunclnberry Ms caradensis Buttercup Ranunculus spp carada mayflower Carada violet Clematis Clintonia Clover Clubmoss Dock Enchanter's nightshade False Soloton's seal Fescue Fleabane Goldenrod Hawkweed Maia_nth_em_m me my; canadensis Clematis viginiana Clintonia borealis Trifolium m m Spp Rt_m__\ex a_oetgs_e_lle Circaea Ma Smilacina racemosa Festuca spp Brig n strig Solidago spp Hieracium aurantiacum Table A-1. (cont'd) 99 (XENON NAME SCIENTIFIC NAME Honeysuckle lonicera caradensis Horsetail Eggisetum arvense lettuce lactuca spp Lichen Various Mallow Mal__va Electa Milkweed Asclepias tuberosa Moss Various Mtullein Verbascum 3.1m Mushroom Various Orchardgrass Dactfllis glomerata Panic grass Panicum spp Pennywort Lysimachia ntmttmularia Pfiuegrass ems Spp Poison ivy Tbxicodendron radicans Polygoanm Polygonumm spp Pussytoes Antennaria Electa Pyrola EEO—1a __secunda Sedge Cat—rex SPP Selfheal Prunella vulggris Smooth aster m 1_a1ev_is Snakeroot Sanicula marilandica starflower Trientalis borealis Strawberry Fraflia viginiana Thimbleweed Anemone Virginiana Table A-1 . (cont 'd) 100 WNAME SCIENTIFICNAME Thistle Timothy Trillium Trout lily Violet Virginia creeper Western wheatgrass White avens Wild sasparilla Yarrow moor Basswood Beaked hazelnut Beech Big-tooth aspen Black cherry Brambles Chokecherry Currant Dogwood Elm Fir Hawthorn Cirsium arvense Phleutm pratense Trillium spp Emu—um ___americanmn Viola spp Agmpyrem __smithii Geum canadense Achillea millefolium LCO 115m Eagu_s ifolia m grand—identata ms serotina MS SPP m viginiara Ri_b_9s americara 92139.5 SPP Ms aumericara flog balsamea Crataw pruinosa 101 Table A-1. (cont'd) WNAME SCIENTIFICNAME Ironwood m vgg' iniana Oak Lerous SPP Red maple Air m Red pine 2iu_tu_s resinosa Ruussian crab m baccata Serviceberry Amelanchier mils Striped maple Acer mylvanicum Sugar maple 153.1“ __saccharum Trembling aspen 29w tremmuloides White ash Fraxinus americara White birch £91313 wifera White Spruce gees slaia Witchhazel Hamamelis vi ginlana 102 Am.oc m.c Am.mc m.m Am.cc N.H Am.mv H.HH unosa cache A~.~c m.v Am.aec m.mm Ac.Hc w.m AH.aHV a.am speeches Am.Hc o.mH AH.HV c.mo Ae.oc m.cH “a.mc c.~o mmos Ao.Hc a.n Am.ch w.sm Am.cc e.m Ae.mc H.H~ crosses: Ae.Hc c.o Ac.sv c.mm Am.Hc o.m Ao.mac s.Hm sundae Ae.oc c.o as.oc s.ce Ac.Hc s.m AH.oc c.cm oooeesmz As.cc m.H Ac.oc H.HH so.cc m.H as.oc m.mH ooucooaoo mam.cv o.c nAc.ec m.s eflc.cc s.~ vo.mc m.nm ooomom An.Hc e.m Ao.cac o.om Am.cc w.m Am.mc w.mm mmoenoao Ao.Hc m.~ AH.oc o.me Am.Hc m.a Ao.mac c.ce nmsoawsne mousse Ac.Hc N.H Am.oc H.HH Am.cc «.0 Aa.ac s.H mmnnmmeonm AH.NV N.» so.oc m.mm Ae.ac c.o Ae.HHV c.me snow cmxomnm AN.HV H.m Ae.ac m.sm Am.cc H.m Ae.mc «.mm mmmumooem AN.HV m.» Am.mac o.cm Ac.cc H.c Am.oc m.ms announce some mm w some as m Ammo mm w some as m mmHommm somezoo nc\~e m.mm .scoa ca mcofid Houucoo one sonoo\mou case mc\~e m.mm ocoeunobuoud on» on peace cowumpwooew "#60033: cos—sou you mesocwsoobu Ammo 9’3ng can as: 33082 .mtd manna. 103 flgwwflo who as; ucwuwmuflo 5H3 “economy? 98 mums; ucwuwuuflo fig goofing? E2893 .Aca.c v do 5 meow: $53G o>fluoaum mi .AoH.c v cc 5 memos gunman 95392 96 AH.~c ~.e so.sac o.mn so.cc c.H Am.wc m.aa scuba» Ao.Hc c.m so.efic m.aa Am.cc N.H Am.mc o.cH manhood case Aw.Hc m.m Ao.ov a.am Ao.cc m.~ AH.mc m.m~ seed noose AH.HV m.~ Am.csc c.cm so.cc m.m AH.oc o.mm subonanwum Ac.Hc H.~ AH.oc w.eH Ae.cc m.m Am.oc m.mm bosoHubmum Ac.ac N.m Am.mac c.c~ Am.cc n.H Am.oc a.o Hmosuaow Ae.cc m.a Ae.oc s.om AH.HV e.o Ao.wc c.mm mooom Ammo mm w some as w some mm w some as w mmHoamm uczmzco ms\me m.mm so.uc9oc .~n< manna 104 Am.3cm.o Aa.mcc.cm Am.3co.m Am.mco.~m “o.cco.o Am.eco.ne omoom 6.83. 3.3898 Rocco 3.883 8.893 Steam unosa caste maoccc naoccc maécc nsocco «3.833 baited 533:: 1933.3 someone 1333633336» «3.39m menznm who: Goon 6.33.8 8.893 8.3mm 86:3 Snood Becca: £336 3.33.3 8636 3.892 Sine 8533 89.2%: 8.333 3.33.3 3.333 8.84.3 Sacco 3.3:. 36838 Amino 34.3.8 $33.» 8.83.2. Sine soccer. $9530 .8303sz Am.3cm.3 AH.HHVH.33 Ae.cco.o Ae.mcw.o Ae.ccs.c Ac.~cs.m moncno maimé nausea cache animus 1333.3 $5.898 mung 3.393 6.33.3 8.892 3.3:.8 Sims 8.299 Emu soother Anion 333.3 Econ 8.33.8 Scout. Amines sofimoom Amino 3.33.3 €533 Shear. 8535 8635 .332 some mm w some .2 w some mm w some has some em a. some as w gm 383.28 mime mom gme 3.3 .83 cannot mood modem snobs ofi so Boom coflouooos. 36003.3: coeeoo no a”: asp—GEM 9,338 can as; 83% . mt< wanna 105 Aoa.o v 5 £6.3me mum mumflukumgm ugwmg 5H3 39H memm 93. 5 mug gag“ wfiumamm mi A03.o v my gag 98 gag #:meuuflU 5H3 30H 28m 95 5 989: 383% 330mm? 96 Am.ovm.o Am.o3vm.m3 Am.ovm.o AN.NVN.~ Am.ovm.o A>.ovn.o souumy Aw.ovw.3 A~.mvm.m Aw.ova.~ Am.vv¢.m3 Ah.ov¢.m Am.~vn.w3 mosuuma @333 $933823 mflo.ovo.o nAo.ovo.o ¢Ao.ovm.3 mAm.mvm.o3 m<fim.ovm.o gnaw.3v¢.¢ :Mmumoz 33.mvm.> An.mvw.nm Ao.mvw.oa Am.m3vm.mo A3.3Vm.n Am.mvo.hm >333 usbya Am.ova.m Am.nvm.m3 A3.3V>.m Am.mvm.vm A>.3Vo.¢ A~.33V>.mm auuwnsmuum Am.3v3.m Am.oavo.om Am.3vm.v Aa.mvn.om Am.ovm.m Am.mv>.mm umsoawumum Am.~vm.m Am.oavo.o~ Am.ov>.3 An.nvw.~3 Am.ovm.m Am.¢v3.VH Hummuawm Ammv mm » Ammv m4 w Ammv mm M Ammv mm w Ammv mm w Ammv m< w mmHummm gogazoo mn\ms m.mm mn\ms 3.o3 Ao.u:oov .mum mHQma 106 83.8 8.8 8.8.38 8.8 28.8 8.83 88.88 3.38 E3388 3.8 8.3 93.8 8.83 8.38 8.8 88.8 8.88 388633.338 8.38 8.8 3.8 3.33 8.8 8.8 3.8 8.88 38333388 3.38 3.8 3.8 8.8. 8.8 3.8 8.8 8.88 88888 83.8 8.8 3.8 3.3 23.8 8.8 3.8 8.8 5833:: 3.8 3.8 8.8 8.3 8.8 8.3 8.8 8.88 8802 3.8 8.8 8.8 3.33 3.8 8.8 3.8 3.3 8895333: 3.38 8.8 0.3.8 3.3 8.38 8.83 88.88 8.38 88963388 3.38 3.8 3.8 3.33 8.38 8.8 8.8 8.88 8895530 88.8 8.8 38.8 8.8 28.8 8.8 83.8 3.33 38333885 888:8 8.8 8.33 3.8 3.38 3.8 3.83 3.8 8.88 wwfimQEm 3.8 8.8 3.8 8.8 8.8 3.8 3.3.38 8.88 E88 :83on 8.8 3.8 3.8 3.33 8.8 8.8 3.8 3.3 8.8838938 88.8 8.3. 28.8 8.88 28.8 3.8 83.8 8.83 38.3.8808 38 mm 8 8.8 .2 8 38 .2 8 38 .2 8 mfiommm 8883 3653333. 8883 82:353. .8883 883.38 8:35 8.68388 88 938.8 3.83 888 833833538238 wfi so 3:88 3.303.398.9083» fig: 8305860 no 338 830885353“ 9333383883 use 37: 35303.8 .733 0388.3. 107 . 83.0 v 8 £38.83an 0.38 393.308.33.893 ugumg 533 30.3 28m $5 5 wage gunman 983388393 mi . 83.0 v 8 gum? 0.38 33.383696 ustwuuflv 533 30.3 8.58m 93 :3 8.3898 gag“ mgHOmnd 3.8 8.8 o.3 3.8 8.83 8.8 8.8 8.8 8.83 88.8.83 8333 88.8 8.8 83.8 8.88 88.8 3.83 83.8 8.88 8333 8353.3. 38 88 8 38 8.8 8 38 88 8 38 88 8 8838888 8.me BEE wmmH EH35. 8.8.888 . 3.8 8388.3 ‘Wfl‘uzunflfiflmmyaflwW