Pesticide Precautionary Statement

PESTICIDE PRECAUTIONARY STATEMENT

This publication reports research involving pesticides. It does not con- tain recommendations for their use, nor does it imply that the uses dis- cussed here have been registered. All uses of pesticides must be registered " by appropriate State and/or Federal agencies before they can be recom- mended.

• CAUTION: Pesticides can be injurious to humans, domestic animals, de- sirable plants, and fish or other wildlife--if they are not handled or applied properly. Use all pesticides selectively and carefully. Follow recom- mended practices for the disposal of surplus pesticides and pesticide • containers.

I_Ol.l.OW THI[ LAUI[L.

U.S.ilrAatli[ll; OFAGEICULIUi|

i o

° •

" CONTENTS Page Methods ...... 1 Results and Discussion ...... 2 Plantation Total ...... 4 Acer spicatum ...... 6 Alnus crispa ...... 6 Amelanchier spp ...... 7 ' Betula papyrifera ...... 7 Cornus rugosa ...... 7 Corylus cornuta ...... 8 Lonicera canadensis ...... 8 Lonicera hirsuta ...... 8 Lonicera oblongifolia ...... 9 • Populus tremuloides ...... 9 • Prunuspensylvanica ...... 10 Prunus virginiana ...... 10 . Rosaacicularis ...... 11 Rubus strigosus ...... 11 Salix spp ...... 12 Conclusions ...... 12 Literature Cited ...... 12 Appendix ...... 15 | TALL SHRUB LAYER BIOMASS IN CONIFER ' PLANTATIONS OF NORTHEASTERN MINNESOrrA

Lewis F. Ohmann Principal Plant Ecologist

Management for wildlife habitat on public land hare (Lepus americanus) (Grigal and Moody 1980, occursprimarily through manipulation of forest stands Mooty 1976, Peek et al. 1976). ! for timber management. Common examples of stand manipulation are: regeneration of new stands via This paper describes the impacts of some common harve§ting followed by natural reproduction, seed- silvicultural practices on biomass production by tall ing, or_planting; release of regenerated stands from shrubs and small broadleaf trees in conifer planta- competing vegetati'on; and later thinning or other tions successfully established between 1945 and 1974 stand improvement treatments. Because good tim- near Isabella, Minnesota. Portions of this biomass-- ber management is not necessarily good wildlife hab- the leaf and current annual woody twig growth-- itat management (Thomas et al. 1976), land man- serve as potential browse for wildlife. agers should know the impacts that silvicultural practices might have on wildlife habitat. This kind METHODS of information can help meet legislative and policy requirements that. land managed by the USDA For- Study plantations on relatively uniform soils est Service support an abundance and diversity of (coarse loamy Dystric Eutrochrepts) _ were selected wildlife, from within a 33-mi 2 (85.5 km 2) area near Isabella, Minnesota (lat. 47° 35'N, long. 91°. 20'W). Parts of Large-scale planting programs in the Lake States the area had been logged for red pine (Pinus resinosa began with the Civilian Conservation Corps in 1933 Ait.) and white pine (Pinus strobus L.) during north- (Krefting 1975). By 1966 more than 1 million acres ern Minnesota's early logging period (ca. 1900-1920). (404,700 ha) of conifers had been planted in Michi- However, the sample was selected from plantations gan (Gysel 1966). The Superior National Forest of established following the clearcut of natural stands northeastern Minnesota has established about 300,000 of pine (jack pine, Pinus banksiana Lamb.) or mixed acres (121,410 ha) of conifer plantations. Conifer pine and spruce (usually black spruce, Picea mariana plantations make up about 10 percent of the Forest's (Mill.) B.S.P.) during the mid-1940's through the Isabella and Halfway (now part of the Kawishiwi) 1970's. Ranger Districts, where this study was conducted (Peek et al. 1976). Sample selection was designed to represent a spec- .... trum of plantation ages and the silvicultural tech- The acreage of conifer plantation in northeastern niques practiced in the study area over the 30-year • Minnesota will probably increase in response to more period. These conditions were represented on a cover- intensive forest management mandated by recent type map with overlays (Hagen and Meyer 1977) Federal legislation that restricts timber harvest in that depicted plantation establishment dates, tree the Boundary Waters Canoe Area Wilderness. species planted, and site preparation and plantation ' release methods. Based on the overlays, plantations • I tatiTheons prinforbrowcipalseverte(food)brateare moowildlisefe(Alcesusingalces),thesewphitelan-- tailed deer (Odocoileus virginianus), and snowshoe _Personal communication with D. F. Grigal.

I were arbitrarily classed by 5-year segments from tallied in only one-half of each mil-acre plot in plan- earliest available (1945-1949) to most recent (1970- tations where shrubs were especially abundant. 1974): The plantations were also categorized on the basis of two silvicultural practices: (1) site prepa- Diameters were used to estimate ovendry weight ration for planting; and (2) release from competing of the leaves, current annual woody twigs, and total vegetation; and on the species of trees planted, aboveground stem wood through use of allometric : estimation equations previously developed for trees Site preparation methods included: (1) no record and shrubs of this area (Grigal and Ohmann 1977, of a site preparation treatment applied; (2) mechan- Ohmann et al. 1976). ical disking of the ground surface, probably by an Athens-type disk plow (Rudolf 1950); (3) dozer-blade Biomass estimates for leaf (LEAF), current an- Shearing of woody stems and scraping of at least nual wood twig (TWIG), and total aboveground stem some of the forest floor material and depositing it in wood (WOOD) were summarized by plantation Ca- specific areas, generally windrows (rock-raking) (No- tablishment period (YEAR), method of plantation ble et al. 1977); and (4) chemical herbicide (usually release (RELEASE), method of site preparation either 2,4-D, or a mixture .of 2,4-D and 2,4,5-T) to (PREP), and tree species planted (SPECIES). the post-logging vegetation. Plantation release methods included: (1) chemical (in more recent years Total plantation shrub layer biomass and biomass usually 2,4-D herbicide applied by helicopter (Hagen of species that were represented in at least 15 of the and Meyer 1977), but earlier could have included a 28 treatment categories were, after logarithmic mixture of 2,4-D and 2,4,5-T applied less effectively transformation, analyzed for difference among treat- from fixed-wing aircraft); (2) manual (hand removal ments and treatment combinations through three- of competing vegetation one or more times); (3) man- way analyses of variance (ANOVAs). • ual plus chemical (a combination of both types of The effect of treatment categories was also as-_ release, usually a manual release followed by a sessed through Multiple Classification Analysis chemical release several years later); and (4) un- (MCA). MCA is a multivariate technique for exam- known (no record of release). Tree species planted ining the simultaneous interrelations between sev- were categorized as: (1) pine (either red pine, jack eral treatments and a dependent variable (Andrews pine, Or a combination of both); (2) pine and spruce et al. 1973). It is especially useful when the treat- (white spruce, Picea glauca (Moench) Voss); or (3) ments are correlated to some degree and where the spruce, attributes examined are not experimentally manip- The plantations were randomly chosen on the cover- ulated (Nie et al. 1975), as in this study. A unique type map and overlays by use of random numbers feature of MCA is the ability to show the effect of selected from right angle baselines drawn on the each treatment category on the dependent variable map. Samples were drawn until at least 3 sites were (biomass) both before and after taking into account picked representing all categories of species planted, the effects of the other treatment(s) being consid- site preparation, and plantation release practiced ered. The classification computes a grand mean for within each of the 5-year establishment periods. Be- the dependent variable (biomass) and for category cause silvicultural technology changed, not all of the means of each treatment, expressed as deviations listed practices were represented in each establish- from the grand mean. These deviations indicate the ment period. For example, disking for site prepara- effect of each treatment category. The analysis also tion was.not applied in plantations established after provides values for each treatment category, ex- 1964, chemical site preparation was not practiced pressed as deviations from the grand mean after ad- before 1965, no plantations were established without justment for other treatments. These adjusted values • site preparation later than 1965, and rock-raking indicate the '_true" effects of the treatment category was not used until 1960. on the dependent variable--the effects remaining • • after variation due to other treatments has been ac- Diameters of shrubs and deciduous tree species counted for. uncler 1 inch (2.5 cm) diameter at breast height (d.b.h.) weremeasured to the nearest one,tenth of an inch RESULTS AND DISCUSSION (0.25 cm) at a point 6 inches (15 cm) above ground level in circular mil-acre (4.05 m2) plots. Shrub di- The sample consisted of nine or more plantations ameters were tallied by species in 10 plots distrib- for each of the establishment periods except for 1970- uted systematically throughout a homogeneous por- 1974 when not enough plantations were available tion of each plantation. For efficiency, stems were for sampling (table 1). Some of the PREP, RELEASE,

. and SPECIES categories also had limited represen- total estimated LEAF, TWIG, and WOOD biomass. tation andonly a minimum 3 samples were selected Although not as dominant, the smaller shrubs can from each of them because of their scarcity in the provide browse, especially because they tend to pro- plantation population (table 1). duce a higher proportion of their biomass as leaves and twigs. For example, for the tall shrub and tree Two of the independent variables, YEAR and PREP, species there was a generally consistent ratio of LEAF were significantly positively correlated. This prob- to WOOD biomass of about 0.14, and TWIG to WOOD ably reflects changes in site preparation practices, biomass ratio of about 0.04. Many of the smaller YEAR and SPECIES, as well as PREP and SPECIES shrubs, however, had higher LEAF and TWIG TO were also slightly positively related. YEAR and.RE- WOOD biomass ratios; for example, Rubus strigosus LEASE, PREP and RELEASE, and SPECIES and 0.66 and 0.17, Rubus parviflorus 0.79 and 1.0, Rosa RELEASE were weakly negatively related: acicularis 0.38 and 0.16, Rosa blanda 0.29 and 0.17, IndependentVariable Pearson'sR Probability and most of the Lonicera species (L. oblongifolia 0.33

iI YEAR-PREP 0.7134 0.000 and 0.21, L. hirsuta 0.22 and 0.08, and L. dioica 0.40 YEAR-RELEASE " -0.1100 .216 and 0.15). Not only do these species have browse YEAR-SPECIES .2022 .073 potential but some, Rosa and Lonicera for example, PREP-RELEASE -0.0691 .312 are highly preferred snowshoe hare and white-tailed RELEAPREP-SSPECIEE-SPESCIES -0..13157698 ..162131 deer foods during• some seasons (Grigal and Moody 19822). It is important to recognize the degree of correlation because positively related treatments, when consid- In addition to total plantation biomass, 15 species were represented by sufficient biomass in the various ered together, will explain less of the variation than categories to test for differences among treatments the sum of them considered separately. On the other through ANOVA. Three-way ANOVAs were con- hand, negatively related treatments that are posi- ducted to examine the combined effects of the sil- tively related t'o the dependent variable will, when vicultural practices on the estimated LEAF, TWIG, considered together, explain more of the variation in and WOOD biomass values in addition to individual . the dependent variable than the sum of them con- treatment effects. Some ANOVAs could not be considered separately. The independent variables YEAR ducted because the close relation between chemical and PREP are correlated sufficiently so that when site preparation and spruce species planted (PREP considered together in a three-way ANOVA or MCA and SPECIES) resulted in a singular matrix. Of the the amount of variation explained is less than ac- 96 3-way ANOVAs conducted, 38 LEAF, TWIG, or tually present. WOOD cases were significant at the p = I>0.10 level None of the 28 woody species was found in the for one or more sources of variation (see Appendix). sample plots within every plantation, and only 7 species were tallied in more than half (27) of the While ANOVA indicates, the presence of differ- plantations (table 2). However, these seven (the tall ences among treatment categories, it does not indi- shrubs Corylus, Amelanchier, Alnus, Prunus, and rate which treatment categories differ from the oth- Salix; and the trees under 1 inch d.b.h., Acer and .. ' Populus) accounted for more than 80 percent of the 2personal communication with Lynn Rogers.

Table l.--independentvariablecategoriesagainstwhich the biomass estimatesof the tall shrub layer (LEAF), current.annualwoody twig growth (TWIG),and total abovegroundwood (WOOD) were tested from 53 conifer plantationsnear Isabella,Minnesota

• I 'Establishmentperiod Site preparat]on Plantationrelease ...... Species planted i YEAE .Plantations PREP Plantations RELEASE Plantations "SPECI"ES• Plantations Number Number Number Number

I 1945-1949• 9 ' None 11 Chemical 17 Pine 38 1950-1954 9 Disk 21 Manual 21 Pine and Spruce 12 I 1955-1959 9 Rock-rake 18 Manual plus chemical 3 Spruce 3 !.960-1964 13 Chemical 3 No record 12 196.5-1969 9 1970-1974 4

3 Table 2.--Mean and standard error estimates for leaf, current annual woody twig, and total abovegroundwoody stem biomass (ovendry) of shrub layer species in conifer plantations near Isabella, Minnesota LEAF TWIG WOOD

Species P1antat i ons _ SE _ SE -X SE Number Pounds/ acre

Total (stand totals all shrub layer species) 53 285.6 26.5 75.9 7.4 2,059.4 212.7 Rubusstri gosus 51 20.2 4.0 5.1 .8 30.6 6.2 Corylus cornuta 50 110.4 15.0 26.0 3.3 841.7 II0.I _i_ 48 6.2 .9 2.7 .4 16.4 2.5 _meTanc'n_-_er spp. 43 20.6 3.1 5.7 .8 166.8 31.4 Alnus _ 34 109.7 20.1 33.2 6.1 935.1 172.7 .__ tremuloides 34 43.0 12.9 5.2 1.4 252.6 72.0 nu______ssPru _Densy_ 31 9.3 2.0 2.4 .5 83.6 17.3 Lonicera canadensis 26 2.3 .4 .8 .I 13.6 2.5 _JaT_xspp. 24 26.1 8.6 3.2 1.0 233.4 87.6 _-et-u_Fapapyri fera 19 2.0 .6 .8 .4 18.2 8.2 Lonicera oblongifolia 18 .8 .4 .5 .3 2.5 1.3 _catum 15 21.0 5.4 12.1 3.3 240.3 58.1 Prunus virginiana 15 12.4 2.5 2.7 .7 65.9 17.0 _s rugosa 13 21.4 9.7 4.5 2.0 85.6 42.9 _r_uta 12 1.7 .4 .7 .2 7.8 3.7 _rum 8 26.8 15.9 21.7 9.8 200.9 118.7 Rosa_ 6 2.4 .8 1.4 .5 8.2 3.4 ITTE_sspp. 4 1.3 1.2 .6 .5 11.7 11.3 • _s americana 3 2.5 1.5 .9 .6 19.4 13.6 Lonicerav_ 3 2.1 1.3 .4 .2 9.0 4.1 _'-6-nIcera _ 2 .8 0 .3 0 1.9 0 P_Pop_b_ samea 2 2.8 2.8 .6 .6 25.2 24.1 Alnusrugosa 1 31.4 29.7 311.8 _s_ni fera 1 23.2 - 3.5 - 124.9 - _s_ 1 15.1 - 3.3 - 86.9 - _h irsuta 1 .1 -- .1 - .4 - iTCE______sp_orus 1 55.3 __ 70.2 _ 70.2 - Viburnumrafinesquianum 1 8.3 _ .9 _ 32.3 _ ...... --

ers. Means that were detected by the Student- ELT-13 (UDLDC)Upland Deep Loamy Dry Coarse, Newman-Keuls test as significantly different (p = and ELT-14 (UDLDM)Upland Deep Medium Loamy I>0.05) included 10 categories of YEAR, four each of Dry. These ELTs are estimated to occupy 10, 8, and PREP and SPECIES, and two of RELEASE (see Ap- 11 percent of the land area within the Superior Na- pendix), tional Forest, respectively. The results presented here would be most applicable to conifer plantations lo- •The relation of biomass to each treatment cate- cated on these ELTs. gory is difficult to interpret because of the complex relation Of treatments and categories and their influence on the resulting biomass values. The ANOVA Plantation Total and MCA analyses help to clarify these relations by making it possible to examine the influence of a The treatment PREP explains a statistically sig- treatment and its categories after adjustment for the nificant amount of the variation in LEAF and WOOD ' effect of the other treatments. Below, I present sum- plantation biomass. Because total biomass is corn- mary interpretations of these analyses. However, be- prised of a combination of species for which greater cause the interpretations are based onadjusted val- or lesser biomass for any one species in any treat- ues, the summary does not in all cases conform to ment category may balance out, it is not surprising the influence levels depicted in Appendix figure 1. that the variation accounted for by a treatment may be statistically significant, but not very high. For Ecological landtypes for the Superior National LEAF and WOOD, the treatment PREP is most im- Forest have recently been described although they portant, explaining 15 and 18 percent of the biomass have not been extensively mapped. The most prob- variation, respectively. The other statistically non- able ELTs for the plantations in this study are: ELT- significant variation trends for LEAF and WOOD 11 (UDHD) Upland Deep Loamy over Sandy Dry, are SPECIES (12 and 14 percent), RELEASE (7 and 11 percent), and YEAR (2 and 5 percent). For TWIG There were no statistical differences in total shrub the ranking is PREP 14 percent, RELEASE 10 per- stratum biomass between 5-year establishment pe- cent, SPECIES 9 percent, and YEAR 2 percent, riods. Plantation shrub layer totals were lower than those found in either 5-year-old post-logging or ma- The most important treatment categories associ- ture aspen stands on good sites, 5-year-old aspen ated with abowe-average total plantation biomass stands after fire, and mature stands of mixed pine- values were older plantations that had been pre- aspen-birch (table 3). Plantation totals were higher pared by disking, planted to pine, and later released than shrub layer biomass in mature jack pine stands from competing vegetation by manual methods. Cat- on shallow soils over bedrock. They were about the egories associated with below-average biomass'val- same as those found in similar conifer plantations ues were sites that had been prepared by chemical and mature stands in the Isabella area and in nat- methods for planting to spruce and later released by urally regenerating 5-year-old conifer stands after chemical methods, fire (table 3).

l- Table 3.--Mean biomasso estimates (ovendry) of tall shrub layers from various studies in northeastern Minnesota

..Stand age (years) __ _i- - - , Source I-5 6-10 11-15 16-20 21-25 26-30 Mature Pounds'/acre

This study_/ 03_z 2,332 2,035 2,086 2,914 2,631 • 2(4)':v (9) (13) (9) (9) (9) Kernik_ Plantations 1,202 230 2,265 2,276 2,339 (I) (I) (2) (3) (7) White cedar 1,331 (I) Black spruce 2,747 (2) Mixed red and white pine 4,943 With subcanopy of aspen-birch (1) Ohmann and Grigal (1979)2-/ Jack pine-fir 2,928 (I) Jack pine-oak 5,705 (I) Jack pine-black spruce 3,074 (I) Aspen-birch 15,387 (1 ) SiIkworth (1980)5-J

l Aspen on good sites 32,834 3,729

Green ,r1979_,__ (2) (2) Jack pine on shallow soil 143 jack pine on deep till 2,680 (15)

1-/Plantation conifer stands after clearcut harvest of jack pine-black spruce. Includes deciduous tree species under 1-inch d.b.h, and tall shrubs. • -_4Number of stands in parentheses. _Personal communication. Plantation conifer stands after clearcut harvest of jack pine-black spruce, and mature uncut communities. Shrub layer includes deciduous tree seedlings under 1-inch d.b.h and tall shrubs. -14Natural conifer (seedi and aspen (sucker) stands 5 years after fire. Shrub layer i total-includes deciduous trees and tall shrubs _Natural sucker stands after whole-tree harvest of aspen; uncut mature aspen stands I on good sites. Shrub layer data includes regenerating aspen and tall shrubs 5 years after T loggi_ but includes tall shrubs only in the uncut stands. ZZMature, Uncut, even-age jack pine stands more than 50 years old on either shallow soils over bedrock or on deep tills. The tall shrub layer data includes tall shrubs and tree seedlings. However, tree seedlings comprised a small proportion of the layer (a maximum of 8.6 pounds//acres in any stand). Most of the seedlings were decidous tree species.

5 habits. Site preparation through chemical treatment " A cer spicatum and plantation release by manual plus chemical treatment would normally be expected to enhance Mountain maple was recorded in only 15 plan- sprouting rather than result in less biomass as found tafions, but was represented in all treatment cate- here, unless the herbicides were applied shortly after gories. Probably because of its low occurrence, none the stems of mountain maple were severed. of the treatments or treatment combinations ex-

•iationplained ina mountainstatistically maplesignificantbiomass.amoRELEASEunt of the var-ex- Alnus crispa plained from 25 to 35 percent of the biomass variation, YEAR explained 9 to 16 percent, PREP explained 4 Green alder was tallied in 34 plantations and was to 8 percent, and SPECIES explained 4 percent, represented in all treatment categories except plantations prepared by chemical methods and planted Above-average biomass values were associated with to spruce. The treatment YEAR was the dominant older plantations that had received only a disk treat- influence on green alder biomass. Treatment com- ment before planting; those that were planted to pine; binations YEAR-PREP-SPECIES, YEAR-PREP- and those that were released manually. Below-av- RELEASE, and YEAR-SPECIES-RELEASE all ex- erage biomass values were associated with planta- plained a significant amount of the variation in green tions that had been prepared for planting by chem- alder biomass. YEAR explained 46 to 53 percent of ical.methods or those that had been released from the variation in LEAF, TWIG, and STEM biomass. competing vegetation through a combination of The other treatments explained less than 2 percent manua| plus chemical treatments, each.

Mountain maple is a prominent understory tall Above-average biomass values were associated with shrub or small tree of northern forest types. It grows 5- to 15-year-old plantations, and to a minor degree, in either sunlight or shade but does best where nei- those sites that had been prepared for planting to t_her shade nor sunlight is extreme (Hosier 1974). pine by rock-raking and later manual release from ' The species repr.,duces most commonly by sprouting competing vegetation. The most adverse treatment from undergrouI, d stems, but it can also reproduce category was chemical site preparation: the species from seed, layered branches, and root suckers (Hos- was not recorded in those plantations in spite of its ier 1974). Ground surface disturbance can eradicate presence in other plantations of the same age cate- mountain maple if the plants are uprooted and the gory. stems severed. However, if the stems are simply tipped over or mashed down, or if they are just severed at Green alder is a circumpolar species (Johnson 1968) the ground surface, sprouting will be stimulated common in the Lake Superior region (Rosendahl (Hosier 1974). The species is resistant to foliar ap- 1955). It is a common dominant understory shrub of plication of 2,4-D and 2,4,5-T herbicides (Benzie 1977). conifer plantations in northeastern Minnesota (Peek Herbicides can stimulate sprouting (Hosier 1974), et al. 1976). After establishment by seed, alders can but the shrub can be controlled by severing the stem spread or persist by sprouting, layering, under- and applying 2,4,5-T to the stump, ground stems, and suckers (Healy and Gill 1974). Alder is susceptible to both 2,4-D and 2,4,5-T her- In a study of mature virgin forest communities of bicides (Arend and Roe 1961). Green alder's capacity the region, mountain maple was recorded in 18 per- for reproducing by vegetative means after disturb- cent ofjack pine-oak community-type stands, 29 per- ance and its susceptibility to herbicides may explain • cent of jack pine-black spruce stands, 71 percent of the biomass distribution found in this study. More jack pine-fir stands, 77 percent of aspen-birch stands, biomass was found in intermediate-age plantations and 100 percent of the maple-aspen-birch stands that were mostly prepared for planting by rock-rak- (Ohmann and Ream 1971). In this region, the species ing (14 of 17 plantations in that time period) where is apparently more representative of mesic conifer alder could have responded to the disturbance by and hardwood communities than of dryer jack pine producing more aerial stems vegetatively. Less bi- community-types. Mountain maple's minimal rep- omass was found in plantations subjected to chem- resentation makes it difficult to develop strong con- ical treatment. clusions about the species' response to silvicultural treatments. The shrub's biomass distribution is about what would be expected based on its known growth - I

A melanchier spp. of the variation in birch biomass. YEAR explained from 48 to 55 percent of the variation, followed by PREP (3 to 15 percent), RELEASE (3 to 6 percent), Juneberry was sampled in 45 plantations and was and SPECIES (1 to 5 percent). represented in all treatment categories. The treatment combination YEAR-SPECIES-RELEASE ex- The most important treatment categories associ- plained a statistically significant amount of the var- ated with above-average white birch biomass were iation in juneberry biomass. YEAR explained 18 to 5- to 10-year-old plantations that had been prepared 20 percent of the LEAF, TWIG, and STEM biomass for planting by rock-raking and then later chemi- variation; SPECIES explained 14 percent; .RE- cally released. The most important treatment cate- LEASE explained 13 to 15 percent; and PREP ex- gories associated with below-average birch biomass plained 12 to 14 percent, were older plantations that had received no site prep- l aration with no subsequent release recorded. Be- I The treatment categories associated with above- cause white birch is a tree species, the low biomass average biomass values were: (1) older plantations values associated with the older plantations proba- that had been prepared for planting to pine by disk- bly show the growth of birch out of the 1-inch di- ing, (2)older plantations that had received no site ameter class. preparation before planting, (3) plantations that had been manually released from competition, and (4) White (paper) birch, a widespread tree species in plantations where no release activity was recorded, northern North America (Viereck and Little 1972), The most important categories associated with less is common throughout the Lake States (Folwells 1965). biomass were the chemically treated plantations, es- It grows on a wide range of soils and soil moisture pecially those receiving both chemical site prepa- conditions, reproducing from seed (given a suitable ration for planting spruce and also later chemical seed bed) or from sprouts after cutting or fire (Fol- releas e. Shrubs of the genus Amelanchier grow most wells 1965), especially if the stems are less than 60 frequently in moist soils (Larson 1974), but some years old (Tabbs 1977). White birch is a fast-growing species of the genus are commonly found on upland but short-lived tree that is normally replaced after sandy soils and dry rocky ridges. Juneberry is a com- one generation. However, in the northern Lake States, mon understory shrub of conifer forests and plan- it may maintain itself through sprouting and gap rations in northeastern Minnesota (Lakela 1965, Peek replacement to become a component of mature fir- et al. 1976). Juneberry reproduces by seed and from birch communities. White birch is susceptible to both suckers. Although its growth and reproduction are 2,4-D and 2,4,5-T herbicides (Arend and Roe 1961). best in full sunlight, it also tolerates shade (Larson 1974). Juneberry is resistant to 2,4-D herbicide but More white birch biomass was found in planta- is susceptible to 2,4,5-T herbicide (Benzie 1977). tions where site preparation and herbicide release were practiced; less biomass was found in planta- Consistent with the species' growth habits and tions where no site preparation and/or release treat- response to management practices, more juneberry ments were applied. These findings are consistent bi0mass was found in older plantations where it would with birch's sprouting ability after disturbance and have had time to recover from site disturbance (disk its need for mineral soil seedbeds. Absence of white preparation for planting and manual plantation re- birch in the three plantations chemically prepared lease). Juneberry biomass was also greater in those for planting to spruce in the 1965-1969 period is plantations that received no site preparation for consistent with the species' susceptibility to herbi- planting, Less biomass was associated with planta- cides. tions that received chemical treatment (assuming a mixture of D-T herbicide). Corr_us rugosa ' Round-leaf dogwood was tallied in 13 plantations Betula papyrifera and was not represented in the youngest establish- ! ment period (1- to 5-year-old) or on sites prepared by

White (paper) birch, (<1 inch d.b.h.) was tallied chemical treatment for planting spruce. None of the in 19 plantations and was represented in all treat- treatment combinations explained a statistically sig- ment categories except, plantations treated with nificant amount of the variation in dogwood biomass. Chemicals before planting spruce. The treatment PREP explained 35 percent of the variation, followed combinations YEAR-PREP-SPECIES and YEAR- by RELEASE (28 to 29 percent), YEAR (17 to 18 SPECIES-RELEASE explained a significant amount percent), and SPECIES (2 percent).

7 , Older plantations that had been prepared for Although it is a pioneer successional species tha • planting by a disk treatment and that later received grows best where overhead shade allows more th_

o a manual plus chemical or a chemical release con- 30 percent of full sunlight, it will persist in matur tained above-average dogwood biomass. Younger conifer stands on light soils (Stearns 1974), as il plantations, sites prepared for planting by rock-rak- these plantations. Hazel is susceptible to chemic_ ing or chemicals and plantations with no recorded herbicides (Stearns 1974). Disk site preparation a_ release or only a manual release contained the least _ plied in midsummer turns up the underground ol dogwood biomass. Thus, greater dogwood biomass is gans of shrubs, setting back their growth for sever_ associated with less intensive site preparation and years (Eyre and Zehngraff 1948). More hazel bk

.. severe release; less dogwood biomass is associated mass was present in older plantations where haz_ with younger plantations, intensive site prepara- would have had sufficient time to recover from th tion, and least release, disk site preparation treatment and where no che_ icals were later applied for plantation release. Lea Round-leaf dogwood is an infrequent understory hazel biomass was found in plantations that had r_ shrub of many northern forest types, mostly dry to ceived a Chemical treatment to prepare the site fc well-drained woodlands (Lesser and Wistendahl 1974) planting and/or to release the conifers after estat and rocky wooded slopes (Rosendahl 1955, Lakela lishment. 1965). In a study of nearby mature virgin forests, this species was recorded only in the more mesic Lonicera canadensis hardwood and nonpine coniferous types (Ohmann : and Ream 1971). In another study of 33-year-old post- Fly-honeysuckle was measured in 26 plantatior fire communities in the area, round-leaf dogwood and was represented in all but one treatment cat_ Was recorded in the aspen-birch and birch commu- " nities but not in the jack pine-birch and jack pine gory (manual plus chemical release). None of t_ • communities (Ohmann et al. 1973). The distribution treatment combinations explained a statistically si! of dogwood biomass found here suggests that the nificant amount of the variation in honeysuckle b shrub is not very tolerant to site disturbance (rock- omass. The most important treatment was YEAI but it explained only 9, 13, and 12 percent of t_ raking) or canopy shade (less biomass in the plan- variation in LEAF, TWIG, and WOOD, respectivel: tations not released). RELEASE explained 6 to 8 percent, SPECIES 1 l 2 percent, and PREP 1 percent.

"_'_orytus cornuta Treatment categories associated with above-a_ erage biomass values were 10- to 15-year-old plal Beaked hazel was tallied in 50 of the plantations tations that had been prepared for planting to piv and was represented in all treatment categories. The treatment PREP explained 15 percent of the biomass by rock-raking and that were manually released frol competing vegetation. Categories associated wit variation, followed by YEAR (14 percent), SPECIES below-average biomass values were: (1) plantatior (11 percent), and RELEASE (9 to 10 percent), for older than 15 years, (2)more recent plantations th_ LEAF, TWIG, and WOOD. had been chemically prepared for planting to spruc, The most important treatment categories associ- and (3) plantations chemically released from con ated with above-average biomass values were older peting vegetation. plantations that had been prepared for planting to Fly-honeysuckle is an upright straggly shru pine or a combination of pine and spruce by disking (Lakela 1965) found commonly throughout cool moi_ and later released only by manual methods or for woods of the coniferous region (Rosendahl 1955). which no release was recorded. The most important is shade tolerant and can compete and persist und_ . categories associated with below-average biomass canopy shade, but it normally will not dominate values were younger plantations that had been site (Jackson 1974). It can reproduce by seed and • chemically prepared for planting to spruce and/or shoots from a spreading root system (Jackson 1974 later chemically released from competition.

Beaked hazel is a common understory species of conifer plantations in northeastern Minnesota (Peek Lonicera hirsuta et al. 1976). After initial establishment in a stand by seed, hazel can produce many aerial stems veg- Hairy-climbing honeysuckle was tallied in 1 etatively from underground stems (Stearns 1974). plantations and was not represented in the 197_ 1974 establishment period or in plantations released It can reproduce from seed and by root suckers (Jack- by a manual plus chemical combination. None of the son 1974). treatment combinations explained a significant amount of the variation in honeysuckle biomass. Because none of these honeysuckles normally PREP explained 32 to 41 percent of the variation in dominate forest sites, little effort has been made to bi0mass, foliowedby YEAR (31 to 36 percent), SPE- control them through management practices (Jack- CIES (3 to 20 percent), and RELEASE (8 to 11 per- son 1974). This study suggests that the biomass distribution of these shrubs is influenced primarily by cent), their susceptibility to chemical herbicides, and sec- The categories associated with greater biomass ondarily (at least, for fly-honeysuckle and swamp fly- were older plantations that had received only a disk honeysuckle) by reduced growth and vigor as the treatment or no site preparation before planting pine plantation canopy closes and shade increases about and plantations that received no release or only a 15 to 20 years after establishment. manual release. The categories associated with less bi0mass were younger plantations that had been rock- Populus tremuloides raked (or chemical prepared before planting spruce), and later chemically released from competing veg- Quaking aspen (less than 1 inch d.b.h.) was rep- etation: resented in all treatment categories and recorded in ' 34 plantations. The treatment combinations YEAR- Hairy-climbing honeysuckle is a high twining vine PREP-RELEASE and YEAR-SPECIES-RELEASE • (Lakela 1965) that is common but not abundant in explained a significant amount of the variation in cool, moist coniferous forests on dry, well-drained aspen biomass. YEAR explained 45 to 50 percent of soils (Grimm 1957, Van Dersal 1938). This shrub, a the biomass variation. PREP explained 14 to 16 per- strong competitor that can persist under forest shade, cent, RELEASE explained 6 to 9 percent, and SPE- can reproduce by seed or by shoots from a spreading CIES explained less than 1 percent. root system (Jackson 1974). • Treatment categories associated with above-av-

' Lonicera oblongifolia toe_aagelesserbiomassextentvalues, thosweree sitesyoungerthat hadplantations,been preparedand by rock-raking or disking, planting to pine, and later Swamp fly-honeysuckle was recorded in 18 plan- release by manual methods or for which no release rations and was represented in all treatment cate- was recorded. These latter categories were associ- gories except the 1970-1974 establishment period and ated with younger plantations but not the youngest manual plus chemical release category. None of the (1- to 5-year-old) plantation group. The most impor- treatments explained a significant amount of the tant categories associated with less biomass were the variation in biomass. The lack of statistical signifi- older plantations, plantations that had been chem- cance is probably the result of its occurrence in few ically prepared for planting spruce, and plantations plantations. The most important treatment Was that had been chemically released from competing yEAR, which explained 32 to 33 percent of the var- vegetation. The negative influence of age on aspen iation followed by PREP (18 to 19 percent), SPECIES is probably a result of aspen growing out of the size (i6 to 17 percent), and RELEASE (2 percent), class represented in this study. A separate analysis indicated that about 20 percent of the basal area and Above-average biomass was associated with 10- density of trees (greater than 1 inch d.b.h.) in the • to 20-year-old plantations that had received no prep- older plantations was attributable to deciduous aration or only a disk treatment before planting of hardwoods, mainly aspen and maple. pine and that were later chemically released from Competing vegetation. Below-average biomass was Aspen is a very common species and forest type associated with older plantations with plantations throughout the Lake States (Brinkman and Roe 1975) where no release was recorded, and with plantations and northeastern Minnesota (Grigal and Ohmann " that received only a manual release treatment. 1975, Ohmann and Ream 1971). It was recorded in 64 percent of the plantations studied here. Aspen is Swamp fly-honeysuckle is a small, erect, some- unique in that almost all stands originate as suckers what straggly shrub that grows in moist habitats that arise from existing root systems (Brinkman and (Van Dersal 1938). It is common but not abundant Roe 1975). Aspen is an intolerant pioneer species in northern forests of this region (Rosendahl 1955). that cannot reproduce successfully under its own Here it was recorded in 34 percent of the plantations, shade. However, if aspen trees are present in the

9 . CanOpy,they can continue to produce suckers that common, fast-growing, short-lived, intolerant pi serve as a reserve of aspen to regenerate a.site after oneer species of boreal forest ecosystems that d_ .fire or logging. Aspen grows on a wide range of sites, clines about 25 to 35 years after disturbance (Mark " frominfertilsaendstowaterloggemdineralsoilsand 1974,Fulton1974).Pincherryusuallybecomese_ peats(Brinkman and Roe 1975).Aeriallyapplied tablisheald tethr eloggingofamatureforest.Itgro_ 2,4-Dherbicidduringe mid-Julytomid-Augustwill fromburiedseedremainingintheforestfloorfrol killexistingaspentreesand sproutsbutwillresult priorcolonizationorfrompriordisseminationbybir_ inexcellentresuckeringthefollowingyear(Perala or smallmammals (Marks 1974).Pin cherryis 1977).Controllingaspenwhen establishingaconifer common understoryshrubin conifeplr antations

.. plantationrequiresheavy mechanicaldisturbance northeasternMinnesota(Peeketal.1976).Itisvet ora herbicidsite epreparation,alongwithlatersuc- susceptibltoe 2,4-Dor 2,4,5-Therbicide(Fults o cessivheerbicidore manualplantationrelease(Ps er- 1974). ala1977). Prunus virginiana Aspen biomass was negatively associated with plantation age, with least biomass in the oldest plantations. Several factors may account for this trend: Chokecherry was tallied in only 15 plantation (1) the capability of aspen (a tree) to grow beyond but was represented in all treatment categories. Non the size:class category in this study (I>1 inch d.b.h.), of the treatment combinations explained a statistJ (2) aspen shaded out by the conifers in older plan- cally significant amount of the variation in any c tations,(3)fewerherbicideorreleaseapplications thebiomassclasses.YEAR explainedfrom28 to3 appliedtoyoungerplantations,increasinglikelihood percentofthevariationinbiomass,followebyd SPE : of resprouts, and (4) a positive association with rock- CIES (25 to 28 percent), PREP (18 to 22 percent: • rake site preparation which was not practiced until and RELEASE (9 to 10 percent). • 1960. • Above-averagebiomasswas associatedwith5-t Prunuspensylvanica 10-year-oldplantationsthathad been preparedb: chemicaltreatmentbeforeplantingspruceandolde Pin cherrywas measured in31 plantationsand plantationsthathad receivedonlymanual releas representeind alltreatmentcategorieNons. e ofthe orwhere no releasewas recorded. treatmentcombinationswas statisticallysignifi- cant.PREP explained22 to23 percentofthevari- Chokecherrygrowsina widevarietyofsoilsan, ationin biomass,followedby YEAR (18percent), habitatsthroughoutmuch ofNorthAmerica(Vilk SPECIES (7percent),andRELEASE (3to5 percent), itis1974).Itgrowsbestinrich,well-drainemoisd, soilwithadequatesunlight,butitiscompetitivaen, Above-averagepincherrybiomasswas associated tolerantofshadeand drought(Vilkitis1974).Chc with plantationsbetween 5 and 25 yearsoldthat kecherryisfoundcommonly throughoutnortheast had been preparedforplantingto pineby diskor ern Minnesota(Lakela1965).The shrubwas re rock-rakmee thodsand latermanuallyreleasefrod m cordedinonly28 percentoftheplantationsofthi competing vegetation. Treatment categories associ- study, and in only 25 percent of 33-year-old post-fir ated with below-average pin cherry biomass were: jack pine stands in an earlier study (Ohmann et ad (1) the very young plantations; (2) the oldest plan- 1973). Chokecherry reproduces by seed, but it als tations that had undergone no site preparation be- can sprout prolifically (Vilkitis 1974). The shrub ma: fore planting and no later release from competing be somewhat resistant to both D and T herbicide vegetation; and (3)plantations where chemicals had (USDA 1961), but it is more susceptible to 2,4,5-' been used to prepare the site for planting spruce, than to 2,4-D. For example, it was found in contrc plots but not in aerially treated (2,4,5-T herbicide Except for the lower biomass found in the 1- to 5- hardwood plots in northern Michigan (Gysel 19571 • year-oldplantations(wherepinCherrywas encoun- •teredin onlyone ofthefourplantationssampled) Thisintermediatsuse ceptibilitytoherbicidemas: and the relativelyhighbiomasswithchemicalre- explaintheambiguousresultofmore chokecherr. leasetreatment,thedistributionofpincherryiscon- biomassfoundinplantationspreparedforplantinj sistentwithwhat isknown ofitsgrowthhabitsand by herbicidtre eatmentand inolderplantationsre responsetomanagement practices.Pin cherryisa ceivingno herbicidreelease.

10 Rosa acicularis not be likely unless the scraping did not remove all . of the organic forest floor. Lower biomass would be associated with plantations prepared for planting by Wild, priekly rose was represented in all treat- chemical treatment if, as is probable for site prep- ment categories and recorded in 48 plantations. The aration purposes, the treatment was a 2,4-D and 2,4,5- treatment combinations YEAR-PREP-RELEASE and T herbicide mixture applied in early summer. YEAR-SPECIES-RELEASE explained a significant amount of the variation in rose biomass. YEAR explained 17 to 19 percent of the variation in LEAF , R ubus strigosus TWIG, and STEM biomass; PREP explained 7 to 10 percent; SPECIES explained 7 to 9 percent; and RE- LEASE explained 5 percent. ' Raspberry was represented in all treatment categories and recorded in 51 sample plantations. Above-average biomass v_lues were associated with The treatment combinations YEAR-SPECIES-RE- 5- to 15-year-old plantations on sites that had been LEASE, and YEAR-PREP-RELEASE explained a prepared by rock-raking, planted to a combination significant amount of the variation in raspberry LEAF, of pine and spruce, and later released from compe- TWIG, and WOOD biomass. YEAR explained from tition by manual plus chemical methods. Lower than 36 to 44 percent of the biomass variation, followed average biomass values were associated with both by PREP (22 to 27 percent), RELEASE (16 to 17 • the youngest and oldest plantations and with plan- percent), and SPECIES (7 to 8 percent). tations that had been chemically prepared for plant- " ing to Spruce. Above-average biomass was associated with sites that had been chemically prepared for planting to Wild prickly rose is a circumboreal species that spruce; and sites that were prepared by rock-raking grows on a broad range of sites. It is common in and later chemically released. Below-average values conifer forests (Rosendahl 1955) and abundant as a were associated with older plantations with no site . successional species on disturbed areas (Densmore preparation and no later plantation release record. and Zasada 1977). After rose is established by seed, it can spread over wide areas vegetatively by rhi- These results are fully consistent with what is zomes (Densmore and Zasada 1977). Because the de- known about raspberry growth habits and the spe- gree of disturbance is important, rose was more cies' response to management practices. Raspberry abundant in nonscarified than in scarified conifer is common in the forest types of this region (Rosen- regeneration plots in interior Alaska (Zasada and dahl 1955). The species can reproduce from seeds, Grigal 1978). In this case, the forest floor organic layers, sprouts, and underground stems (Core 1974). layers were totally removed by scarification, while It grows better in partial shade than in full sunlight there was rose vegetative reproduction from rhi- and can be rejuvenated by partial removal of over- zomes on the relatively undisturbed organic layers head shade or deep cultivation (Core 1974). Scari- of.the nonscarified plots _. Rose is resistant to 2,4-D fication plots contained more raspberry than non- " herbicide (Benzie 1977) but is susceptible to 2,4,5-T scarified conifer regeneration plots in interior Alaska herbicide applied in May or June (Smithberg and (Zasada and Grigal 1978). Raspberry is resistant to •, Gill 1974). Rose was absent from northern Michigan 2,4-D herbicide (Benzie 1977). More raspberry bio- plantations where 2,4,5-T herbicide was aerially ap- mass was found in younger plantations where her- plied (Gysel 1957). bicide was used to prepare sites for planting to spruce, where sites were rock-raked, and where it was used The results of this study are generally consistent to release planted conifers. These practices reduce with known rose growth habits and response to man- overhead shade but do not kill the resistant rasp- agement practices. More biomass was associated with berry unless a combination of 2,4-D and 2,4,5-T is sites prepared by rock-raking and those plantations used. Management practices involving herbicides re- treated by a manual plus chemical plantation resulted in an increase in raspberry cover in Itasca lease. If the herbicide was 2,4-D or if it was a D-T State Park, Minnesota (Hansen et al. 1974). YEAR •mixture but not effectively applied (applied in Au- was the most important treatment. There was less gust), more biomass would occur. Manual release raspberry biomass in older plantations where the might involve severing some rose aerial stems that ground was not disturbed before planting and where Would produce more sprouts. More biomass following no release treatment was later applied to remove rock-raking as a site preparation treatment would overhead shade.

o 11 • Sal app. CONCLUSIONS

Willow was recorded in 24 plantations and was Most of the responses to silvicultural practicq represented in all treatment categories except the found through this study are consistent with the e youngest (1- to 5-year-old) plantations. None of the ological requirements of the species, at least to t| treatments explained a significant amount of the extent that their requirements have been doc variation in biomass. In the case of LEAF and WOOD, mented. The silvicultural goal of successful esta RELEASE explained 23 and 24 percent of the yar- lishment and regeneration of conifer stands was m iation in biomass, PREP 18 percent, YEAR 14 and for these plantations, yet the stands contained t_ 15 percent, and SPECIES 1 percent. In the case of shrub layers similar to those in conifer stands n_ TWIG, YEAR explained 31 percent, RELEASE 18 urally regenerated after disturbance. The shrub la percent, PREP 18 percent, and SPECIES 2 percent, era were also similar to those found in several m ture conifer forest communities in the area. N surprisingly, the study plantations contained lc Above-average biomass was associated with the shrub layer biomass than did either regenerating oldest plantations that had received a disk treatment mature aspen stands on better sites. or no preparation before planting pine and that were later released by manual plus chemical or chemical The effect of a persistent shrub layer on the lol •methods. Below-average willow biomass was asso- term growth and yield of the planted conifers is ciated With: (1) younger plantations, (2) plantations relevant to the goal of successful re-establishm_ that had received rock-rake or chemical site prepa- of the stand (thinning was not included in the stud ration (those planted to spruce), and (3) older plan- The presence of a persistent shrub layer is, howe_ tations that had been given a manual release or where no release was recorded. Togeneralize, willow LEAF, very important to the suitability of these plantati( as wildlife habitat. Because a persistent shrub la: TWIG, and STEM biomass was greatest in planta- adds to both species and structural diversity, it tions receiving the least site preparation and the important not only to animals commonly though most release; wLlow biomass was lowest in planta- as browsers (deer, moose, and hare), but to invel tions receiving the most site preparation and the brate browsers and to many birds as well (Pro least release. 1979). Willows comprise a widely distributed genus with about 80 species throughout North America (Rosen- dahl 1955). At least 3 willow species grow in the LITERATURE CITED study plantations: bebb willow (S. bebbiana), pussy willow (S. discolor), and upland willow (S. humilis) Andrews, F. M.; Morgan, J. N.; Sonquist, J. A.; K] (Peek et al. 1976). Willows reproduce by seed, suck- L. Multiple classification analysis. 2nd edition era, sprouts, and root shoots; are short-lived; and do stitute for Social Research. University of Mi best in good soils and full sunlight (Rawson 1974). gan, Ann Arbor, MI; 1973. 104 p. Bebb willow grows both in moist and dry places; Arend, J. L.; Roe, E. I. Releasing conifers in the I pussy willow is comrLonly found in swamps and other States with chemicals. Agriculture Handb wet places, but it also grows well on dry soils; and Washington, D.C." U.S. Department of Agr: upland willow is common on dry soils (Grimm 1957). ture; 1961. 185 p. . Willows are susceptible to foliage application of 2,4- Benzie, John W. Manager's handbook for jack D and 2,4,5-T herbicides (Rawson 1974); however, in the north-central states. Gen. Tech. Rep. they may resprout from the roots and stem base the 32. St. Paul, MN: U.S. Department of Agricul_ season after treatment (USDA 1961). For example, Forest Service, North Central Forest Experb Gysel (1957) found willow in both check and treat- Station; 1977. 18 p. ment plots after 2,4,5-T herbicide was aerially ap- Brinkman, Kenneth A.; Roe, Eugene I. Quakin plied in northern Michigan. pen: silvics and management in the Lake St Agriculture Handbook 486. Washington, D.C.: More willow biomass associated with plantations Department of Agriculture; 1975. 52 p. receiving chemical release is consistent with the ape- Core, Earl L. Brambles. In: Gill, J. D.; Healy, V cies' ability to reproduce vegetatively following such eds. Shrubs and vines for northeastern wil treatment, p. 16-19. Gen. Tech. Rep. NE-9. Broomall, PA Department of Agriculture, Forest Service, North- Healy, William M.; Gill, John D. Alders. In: Gill, J. eastern Forest Experiment Station; 1974. 180 p. D.; Healy, W. M., eds. Shrubs and vines for north- "Densmore, R.; Zasada, J. C. Germination require- eastern wildlife, p. 6-9. Gen. Tech. Rep. NE-9. ments of Alaskan Rosa acicularis. Can. Field-Nat. Broomall, PA: U.S. Department of Agriculture, 91(1): 58-62; 1977. Forest Service, Northeastern Forest Experiment Eyre; F. H..; Zehngraff, Paul. Red pine management Station; 1974. 180 p. t in Minnesota. U.S. Department of Agriculture . Hosier, Paul E. Mountain maple. In: Gill, J. D.; Healy, Circular 778. Washington, D.C.; 1948. 70 p. W.M., eds. Shrubs and vines for northeastern Fowells, H. A. Silvics of forest trees of the United wildlife, p. 93-95. Gen. Tech. Rep. NE-9. Broomall, i States. Agriculture Handbook 271. Washington, PA: U.S. Department of Agriculture, Forest Serv- D.C.: U.S. Department of Agriculture; 1965. 762 ice, Northeastern Forest Experiment Station; 1974. p. 180 p. Fulton, John R. Pin cherry. In: Gill, J. D.; Healy, W. Jackson, Lawrence W. Honeysuckles. In: Gill, J. D.; M., eds. Shrubs and vines for northeastern wild- Healy, W. M., eds. Shrubs and vines for north- life. p. 26-28. Gen. Tech, Rep. NE-9. Broomall, PA: eastern wildlife, p. 71-82. Gen. Tech. Rep. NE-9. U.S. Department of Agriculture, Forest Service, Broomall, PA: U.S. Department of Agriculture, -Northeastern Forest Experiment Station; 1974. 180 Forest Service, Northeastern Forest Experiment p. Station; 1974. 180 p. Green, David C. Productivity of Pinus banksiana _ Johnson, Frederic D. Taxonomy and distribution of Lamb. on shallow soils in northeastern Minne- northwestern alders. In: Trappe, J. M.; Franklin, Sota. M. S. Thesis. St. Paul, MN: Soil Science De- J.F.; Tarrant, R. F.; Hansen, G. M., eds. Biology partment, University of Minnesota; 1978. 161 p. of Alder. p. 9-22. Portland, OR: U.S. Department Grigal, D. F.; Moody, N. R. Estimation of browse by of Agriculture, Forest Service, Pacific Northwest .size classes for snowshoe hare. J. Wildl. Manage. Forest and Range Experiment Station; 1968. 292 44(1): 34-40; 1980. p. Grigal, D. F.; Ohmann, L. F. Classification, descrip- Krefting, L. W. The effect of white-tailed deer and tion and dynamics of upland plant communities snowshoe hare browsing on trees and shrubs in

p. ' within a Minnesota wilderness area. Ecol. Mon- northern Minnesota. St. Paul, MN: University of ogr. 45: 389-407; 1975. Minnesota Agricultural Experiment Station Tech. Grigal, D. F.; Ohmann, L. F. Biomass estimation for Bull. 302; 1975.43 p. some shrubs from northeastern Minnesota. Res. Lakela, O. A flora of northeastern Minnesota. Min- Note NC-226. St. Paul, MN: U.S. Department of neapolis, MN: University of Minnesota Press; 1965. Agriculture, Forest Service, North Central Forest 541 p. Experiment Station; 1977.3 p. Larson, Joseph S. Serviceberries. In: Gill, J. D.; Healy, Grimm, W. C. The book of shrubs. Harrisburg, PA: W.M., eds. Shrubs and vines for northeastern The Stackpole Company; 1957. 522 p. wildlife, p. 126-128. Gen. Tech. Rep. NE-9. Broom- Gysel, L. W. Effects of silvicultural practices on wild- all, PA: U.S. Department of Agriculture, Forest •life food and cover in oak and aspen types in north- Service, Northeastern Forest Experiment Station; ern Michigan. J. For. 55: 803-809; 1957. 1974. 180 p. , Gysel, L. W. Ecology of a red pine (Pinus resinosa) Lesser, W. A.; Wistendahl, Jean D. Dogwoods. In: plantation in Michigan. Ecol. 47(3): 465-472; 1966. Gill, J. D.; Healy, W. M., eds. Shrubs and vines . Hagen,. R.; Meyer, M. Remote sensing of vegetation for northeastern wildlife, p. 32-41. Gen. Tech. Rep. as related to black bear habitat, spruce budworm NE-9. Broomall, PA: U.S. Department of Agri- damage, and plantation release appraisals on the culture, Forest Service, Northeastern Forest Ex- Superior National Forest. Remote Sensing Labo- periment Station; 1974. 180 p. ' ratory. St. Paul, MN: College of Forestry and Ag- Marks, P. L. The role of pin cherry (Prunus pensyl- ricultural Experiment Station, University of Min- vanica L.) in the maintenance of stability in north- . nesota; 1977. 19 p. ern hardwood ecosystems. Ecol. Monogr. 44: 73- Hansen, H. L.; Kurmis, V.; Ness, D. D. The ecology 88; 1974. of upland forest communities and implications for Mooty, J. J. Year-round food habits of white-tailed -_ management in Itasca State Park, Minnesota. St. deer in northern Minnesota. Minn. Wildl. Res. Paul, MN: Agricultural Experiment Station Tech. Qtrly. 36(1): 11-18; 1976. Bull. 298, University of Minnesota; 1974. 44 p. : Nie, H. H.; Hull, C. H.; Jenkins, J. G.; Steinbrenner, K.; Bent, D. H. Statistical package for the social ...... sciences. McGraw-Hill, Inc.; 1975. 675 p.

• 13 Noble, M. G.; DeBoer, L. K.; Johnson, K. L.; Coffin, ment of Agriculture; 1950. 171 p. B. A.; Fellows, L. G.; Christensen, N. A. Quanti- Silkworth, Darin R. Leaching losses of nutrients tative relationships among some Pinus bank- lowing whole-tree harvesting of aspen. M. S. siana-Picea mariana forests subjected to wildfire sis. St. Paul, MN: Soil Sciences Department, ! and postlogging treatments. Can. J. For. Res. 7(2)" versity of Minnesota; 1980. 163 p. 368-377; 1977. Smithberg, Margaret; Gill, John D. Roses. In: Ohmann, Lewis F.; Grigal, David F. Early revege- J.D.; Healy, W. M., eds. Shrubs and vines tation and nutrient dynamics following the 1971 northeastern wildlife, p. 116-121. Gen. Tech. ] Little Sioux Forest Fire in northeastern Minne- NE-9. Broomall, PA: U.S. Department of ,4 .. sota. For. Sci. Monogr. 21; 1979. 80 p. culture, Forest Service, Northeastern Forest Ohmann, L. F.; Ream, R. R. Wilderness ecology: vir- periment Station; 1974. 180 p. gin plant communities of the Boundary Waters Stearns, Forest W. Hazels. In: Gill, J. D.; Healy Canoe Area. Res. Pap. NC-68. St. Paul, MN: U.S. M., eds. Shrubs and vines for northeastern Department of Agriculture, Forest Service, North life. p. 65-70. Gen. Tech. Rep. NE-9. Broomall, Central Forest.Experiment Station; 1971.55 p. U.S. Department of Agriculture, Forest Ser_ Ohmann, L. F.; Grigal, D. F.; Brander, R. B. Biomass Northeastern Forest Experiment Station; 1974. estimation for five shrubs from northeastern Min- p. nes0ta. Res. Pap. NC-133. St. Paul, MN: U.S. De- Thomas, J. W.; Miller, R. J.; Black, H.; Rodiek, J partment of Agriculture, Forest Service, North Maser, C. Guidelines for maintaining and Central Forest Experiment Station; 1976. 11 p. hancing wildlife habitat in forest managemer Ohmann, L. F.; Cushwa, C. T.; Lake, R. E.; Beer, J. the Blue Mountains of Oregon and Washing R.; Brander, R. B. Wilderness ecology: the upland 41st N. A. Wildl. and Nat. Resour. Conf. Tre • plant communities, woody browse production, and Wildl. Manage. Inst., Washington, D.C.; 197_ • small mammals of two adjacent 33-year-old wild- 452-476. _fire areas of northeastern Minnesota. Gen. Tech. Tubbs, C. H. Manager's handbook for northern h Rep. NC-7. St. Paul, MN: U.S. Department of Ag- woods in the north-central states. Gen. Tech. ] riculture, Forest Service, North Central Forest NC-39. St. Paul, MN: U.S. Department of ,4 Experiment S ation; 1973.30 p. culture, Forest Service, North Central Forest Peek, James M.; Urich, David L.; Mackie, Richard periment Station; 1977.29 p. J. Moose habitat selection and relationships to U.S. Department of Agriculture. Chemical cot forest management in northeastern Minnesota. of brush and trees. Farmer's Bulletin 2158. V_ Wildl. Monogr. 48; 1976. 65 p. ington, D.C.; 1961.23 p. Perala, D. A. Manager's handbook for aspen in the Van Dersal, W. R. Native woody plants of the Un north-central states. Gen. Tech. Rep. NC-36. St. States: their erosion control and wildlife val Paul, MN: U.S. Department of Agriculture, Forest Miscellaneous Publication 303. Washington, ]: Service, North Central Forest Experiment Sta- U.S. Department of Agriculture; 1938. 362 p tion; 1977.30 p. Vilkitis, James R. Common chokecherry. In: Gil Probst, John R. Oak forest bird communities. In: D.; Healy, W. M., eds. Shrubs and vines for no DeGraff, R. M.; Evans, K. E., eds. Management eastern wildlife, p. 23-25. Gen. Tech. Rep. N of north central and northeastern forests for non- Broomall, PA: U.S. Department of Agricult game birds, p. 80-89. Gen. Tech. Rep. NC-51. St. Forest Service, Northeastern Forest Experin Paul, MN: U.S. Department of Agriculture, Forest Station; 1974. 180 p. Service, North Central Forest Experiment Sta- Viereck, Leslie A.; Little, Jr., Elbert L. Alaska t: tion; 1979. 268 p. and shrubs. Agriculture Handbook 410. Wash • Rawson, J. W. Willows. In: Gill, J. D.; Healy, W.M., ton, D.C.: U.S. Department of Agriculture; 11 ' eds. Shrubs and vines for northeastern wildlife. 265 p. p. 147-149. Gen. Tech. Rep. NE-9. Broomall, PA: Zasada, John C.; Grigal, David F. The effects of • U.S. Department of Agriculture, Forest Service, vicultural system and seed bed preparatior Northeastern Forest Experiment Station; 1974. 180 natural regeneration of white spruce and as., p. ' ated species in interior Alaska. In: Hollis, C. Rosendahl, C. O. Trees and shrubs of the Upper Mid- Squillace, A. E., eds. Fifth North American Fo west. Minneapolis, MN: University of Minnesota Biology Workshop Proceedings. Portland, OR: Press; 1955. 411 p. Department of Agriculture, Forest Service, Pa Rudolf, Paul O. Forest plantations in the Lake States. Northwest Forest and Range Experiment Stat Tech. Bull. 1010. Washington, D.C." U.S. Depart- 1978: 213-220. APPENDIX LEAF, 1 TWIG, and 2 WOOD instances (Appendix table 8). The treatments as individual sources of sig- _, nificant variation (as variables 1, 2, or 3 listed in the combination) showed YEAR to be most important with significant F-ratios for 9 LEAF, 11 TWIG, and , This Appendix presents the results of the ANO- 10 WOOD instances. PREP was significant once for VAs and Multiple Classification Analysis (fig. 1). It LEAF and WOOD; SPECIES was significant once

1 Speciesalso includesbiomassmobyre treatmentdetailed tabular(Appendixinformationtables 4-7).on RELEASEfor TWIG andwastwicesignificanteach forforLEAF2 LEAF,and 1WOOD;TWIG, and 2 WOOD instances. Overall interaction effect was Of the 96 three-way ANOVAs conducted, 38 LEAF, significant in YEAR-RELEASE-PREP twice each for TWIG, or WOOD cases were significant at the p = LEAF, TWIG, and WOOD; YEAR-SPECIES-RE- I> 0.10 Ievel for one or more sources of variation: LEASE was significant in 2 LEAF, 1 TWIG, and 3 model, main effect, independent variable 1, inde- WOOD instances; and YEAR-PREP-SPECIES was pe.ndent variable 2, independent variable 3, overall significant once for LEAF and WOOD. Interaction interaction, interaction between variable 1 and 2, between YEAR and PREP was significant twice each interaction between variables 1 and 3, and inter- for LEAF and WOOD and once for TWIG; interaction action between variables 2 and 3 (Appendix table 8). between YEAR and RELEASE was significant 7 times ANOVA main effects were significant at p = I>0.10 for LEAF, 5 times for TWIG, and 6 times for WOOD. . most often for YEAR-SPECIES-RELEASE with 4 PREP-RELEASE and YEAR-SPECIES interactions times each for LEAF, TWIG, and WOOD; for YEAR- were significant once each for LEAF and WOOD, and RELEASE-PREP with 3 times each for LEAF, TWIG, SPECIES-RELEASE was significant once each for • and WOOD; and for YEAR-PREP-SPECIES with 2 LEAF, TWIG, and WOOD (Appendix table 8).

! j

' 15 + YEAR PREP RELEASE SPECIES _ .++.++__.._._.__o%#_o_ o__+_o* _o__._,.._o_

. Acerspicatum LEAF O O 0 O 0 O O! • 0 •w,_° eCo • 0o.o OoOe o wooDeo C • • 00io OOeO 0 Alnuscrispa LEAF OOC • O O O!O O 0 O • 0 •w,_O0.Coo oolo ooO. o wooOOCooo o olo ooO• o

Amelanchierspp. LEAF 0 • • 0 0 • o iO O 0 O 0 0 -,-wO,Ge coo o oio • o • o o wooe__O_____e_O• •1o ooOo o Betulapapyrifera LEAF AQ___O 0 O 010 0 O O 0 O .w._tO___-__ C • • o OO0 • woo_o__O_eo-_ • o,o o • o o . Cornusrugosa LEAF __tQQ OC' O OOO 0 0 • +w,__tO0 OC'. O00o o wooo_OO_O •Co O00o o Coryluscornuta LEAF__O.___o0O__0o • 01o: • 0 • • o ' +w,_Oo__.eoo • o!o • o • • o woooO°_eo° •o!o eo•e o Loniceracanadensis LEAF £t££ 0 • •10 O 0 0 0

TWIG ....0 O 0 0 0 O • 0 O 0 0 0 woo211oooo • o• Oo o o Lonicerahirsuta LEAF __O0 £ C 0 • • • 0 • TWIG .... ooo Ooo o• o • • wooo__eOo CC'O e e 0 • Lonicera oblongifolia LEAF __ O 0 Q . 0 0 O 0 " O ___0____ +w.o_ooO. •CO O. • 0 ' woo__e_o o___• oC,O O • • 0 Populustremuloides LEAF Q__O__£ __ CC'O 0 " OO 0 +w,o__C_._o CC' • o. 0 • o woooQCO_e CC'e o.Oe o Prunuspensylvanica LEAF 0£ 0 Q OC' 0 0 0 0 O 0 +w._O2_o___o Oolo •oeO o wooeo.___o OCio o o o• o Prunusvirginiana LEAF _00£ • 0 O!O • 0 0 • • +w._O_CoQe oe!• •OoO • wooeo____C_• Q • o ol • • 0 o • • Rosaacicularis LEAF Q 0 0 • 0 0 O IO • 0 • 0 0 +w,__eooe__o o o,o • o o o o . wooOooo e_o o eio • o o o o

Rubusstrigosus LEAF __0____0__2 Q___ " Oi_) 0 oOO 0 +w,o___Q£o • ol. o o.. o woo__0oo __ . o,,, o oO0 o Salixspp. LEAF__Oe o 0 oIo 0 • 0 O, o +w,o_tO___ o O!o 0 • O0 o . woo_O_° e___ 0 o!e 0 eO0 o PLANTATITOONTAL LEAF0 0 0 • • • O IO • 0 • • o Tw.oG-5-5_e • oJe • o • • o WOOD • 0 o O0 O01o • O0 • o

, O :" SLIGHT POSITIVE O: : PO'¢ITIVI ----STRONGLY • --POSITIVE' , OR

----OR NEGATIVE , • ORj_IEG/ NI_GATI_fE

Figure 1.--Diagrammatic representation of classification analysis results depicting the influence of silvicultural treatments on the major tall shrub and deciduous tree (under 1-inch d.b.h.) species in conifer plantations near Isabella, Minnesota. Shaded circles represent a negative influence and open circles represent a positive influence.

16 o J ,::_1"o r'.-. 00 _o _D ("_ Od <"_ I".-. ,,_ O0 '_ O_ CO _l'_ l r--IO I".. O0 '_1" '_" I'_ _l,f} I._ 0 CO _[} 0"}00 {_1 I- I... l 0_0 _ Od _ _l_ ,l_ _l.l_ 1"4 _ 0 0

0 I •lm " 4.J E I _ r'_,.l_C_l 00 ,::_I- _ ,,.0o'_ _ I'.__,_ _ .('_ ¢_ _ ...... I • I • • I I I I • I -0 " 0 _._ m 0 _IV_ t l ,--I _ _ ,--4,--i I I ,:::JI- I I l I v--I_ _ _._

! _" I"... O0 O0 _') 0 01'_. _00 _ _. r..l_ ur_ _l._ _ Od _-.__ _ •

o._ I_I _00_4 __l ",ll_, _,______g'_ QJ _ l E 0 , I _,-- Iz o_,_-r--.,--_cocn _ cO_n ,--_o,_ ,-__ ,,_1<-_ cO cO o'_,--_,--_ o'_ C::_--- I 0 ll_ i'l-- I eJ '_:_I"I'_ I'_ ':_" O_ l._eJ I--I O_OOl " ell lll_ l._ l'-

, o (_J L _ _0 0"_ •_ L- I i_I I

¢_ IC- c_J o,J _0 ,:_l- 00 co o,j _.. r,. ,.1., ,-_ c) _._ o,_ o'_ _ o'_ ,--, o"_ o'_

0 , 4 C',jc_"}@,J _ @,1 _" (._ ,..-4

_ I 0_" _ 00 C',dr_. r_ _'_ c%1 O'_u_ _od ,---_kO ,_ c,_ i--400,--4 C_ ,:::t- O_ L (_ 0 • L} rO t._ _ kO ,_:t-0 or_ 00(%1 CxJ,_:t-,_:_- 04 CO0 0 L _:_ IU.I _0 • • • • • _J 0 _1L _ _ _1 _-_ r-_ _

_'-_- l¢_ o_ 0o eJ _ ,_- _r__ eJ _ ,l, r.. _r_ , _ _, I !

ILl ,,,-,-4 ,,-,,-4 ,--,4 _ ,,,,,,,4 _'_ _ ._I I

O._'q II.J.I I • • • " • • " • • • I I • • • • • I • • I • •,-- _ CO_ICI'} '::_'',_C_ _ _ ,--4,"-0 ,-4 I I O0<_,-0 O,J,-0 ,....t q,_ I C_ ._ L _4 L I ,:::J"O_ r,_ _ _ ,--4,---, _I"

0 _l _l__ _l_ _ -0 0 I_'- I r,,. _) r.... i._ i'... _) ,_-oJ _,.-i _ _o !'..- 00 I'-.. o c_ o'} ,_- l,_ ,_- o i_-- '_ •or-. ! ',_I" _'_ ,1-._ _ _ 00 co _'_ c%1_1 _l_ I •, -I-J

_,-- I

II.l...J I I .... I I • • I • • I • • • ^o^e I _ .,.e I 0 • LI- _- I _ U.I ,r- _ _ I ._ ' ,. 0 i_- I O,_,--_UD ur_-.._DO,_u'_ 0,_ Qor-- o'_ _ C_Or-. unO_OJ 0,-_ ,-_ c" (...} I_ ' o•co_• • o_ooco_ _ _._o0 eJ_ ,_ o_e co• E o_l , • _>,_> L.. _ I_ I C_'},--i ,.--4,-.4 0">,--4 <'_,--IC%I C%J ':::I" 0 0 ,"-" 0". I

L_J_ tO.r-_-_ _ ¢L_II_LJ.J ! _-_° l I ' I _ • I ,_• I _ • 0 • • _ L} ,-- I _ I ,--4 _ I._ ,.-4

• ' _' _l I_ 000 t__r0-__ O_ _:_- O0 ,,0ur_ 0 _ O_l

• L_- L J_" I O..l Od _ l._ "ll l._ ._ ,::_ .._ I,_ I • • • • • • • • • • • _ E "_L_

, _ _ __- _I _- _ _ .,--,_,_ _

:'_l::_.E _::::3_I:l_l:_"-_"_ L u_ 0 glo E"; '_ "_- . _--O0 , _ _ _ E ,_Ioll,,oie; _.ilo_o Ul_ IO1"_ r,....= • _l _ _ U L ..i l=,io¢/,_I_ ii._ LL__o==o(_ _,.-_" F=F=• _ _- (" E (If _I:l E_O 0 0 o L._ I-- _ .a-_ _-_" lolP"I_ :a_

O_ -=. OhO r-40 I_ ¢MI_ _ ¢_J C) ¢_ M:) _ O1._ I_ L..r- I • • • • • • • • • • • • • • • •

=:X:: J

Z C Cx.ln_ _-.40,.I,-4 OO0 ',::I"O" O0 ¢%J _O eO I"_ CO O0 r I 00_) 0 uri CM 0_0 QO i_ 0 ,-4 ,-I O0 I,O _1" IJ- 4_ ,-=4_:I* O'_ OO ,::_ ,,_U_ _ _ ¢'_),--Ir'_. _ _-4o0 _ i e_ ' C_I_ _ "0_ ...... _ o-4 t_ _ v-4 t f') I

IZ I ,--I_) Oh ¢_ ¢_) Ohl._ _ _D _'=-e _$" _ID ,-4 _.=4_-.I _--00_ _=-I Ohl._ Oh I 0

O u_ _O O _ ¢%,IJ'*_ v-4 t.f'u)_ £%1 £%11-4 OO nO L=.I I O_ _ ¢%.I OO ¢_) OO _--4

" I :_=_I ' ,.==I (%1 _ _ ...... • i _- O "::I" ¢%1 ll) O_ ,.-0 _ ',::¢%1I" O0 ,---4,-4 ,*.-t ¢v.) I ¢O I"_ £'%1 O ":::I" I

IZ CxJ',_ r_ ,==0O0 nO C%1 ,--4O0 i.,") _ ¢'_ ,=-0¢'_),,::cI-o OO O0 ¢'O ,-=4_-4 O_ l I o- - ° °

o ::_1 o _ ,-4 ,_- O0

I _t__-_t_-4 _-_ 00¢_ ,--4 _d_) CO

•-_ _-4 C_ Z • _. .

.. UP) , 4..) _O ,--0C) OO r-..u_ O_ ,=..4 (_) M:) no M:) Oj O _£'%1 _ ¢:: ." o _ o-' _._oo _ _ ,-: _l_: " _o_o l _ l _ o0 Lo_

_1 I _ _ ._ "_ U _ ._Q _ • ,=.4)--I rio _ _ _I_ .p

• _ t ur) O r'_ oJ I"_ r'_. r"_.O", o_I cO ,.=4 co (%1 n*) _D _,=4 ',::I,--=a r'_ 0 I_ ...... _" ,_" CO CO _ _ _ _--a ,--_ I-4 _100 o*_ _- O I C_I ,--_,_I" ,--4 I_ _--_,.-__-_ _ 00 ,-4 (-- I . ! U Z C%ICJ I_. O'_ _ID l.lr) ('_)I.'_ O _) "::1" _="0,"'0 0"_ C) _0 OJ 0"_ ,--00d I._ ,...-0 O0 .r-

• • • • • • • • • • • • • • .r-

I OQ 0_I -C%,I _ _ e,-i _ I,,01

_0> ! I cO £:) _ • • • • • • • (3). , _oo,-. o _ _ _o_-_ __-_oo_ _

.r- Z I _"I {'_'} _;_I" 0'_ CY) ,--i _0 I.{') ,-'I '_t ,_- e-..l O_ t_O {Y) N 0'_ e-'q 0'_ I'",.- _-'I

o_ I_' i_.....oo_ i _ _ i " " l _4 .... ' " " ' " ® ,-'4 LI i I_. _=-0 _1" t_l _" _D U_ M:) £_ _-4 n*) n ml c_ _ r,_ U e-_ >-1 jQ "_Q cO _ 0 'c- 1 ooO_O _O00,-_ 0000Oo0 00_-_ eoun_ o0 cO,_l- o0r_.ej un (-

= o ,-'a 0'_ I.,t_ l,,l"_P... ,,_1" _._'_I"',,.. _',x,! ,,-,4 _ 0'_ _%) n,.C) C%1r'-.. _c)

• 0_1 £%1 O O , IZ I 00,..-4 _-=I,=-.4 O') _*-a C_ ,=-40j £%1 ',:::I" _ .4=)

i'L',v, _ _el It oC_ _ _ I C%• 1 O_ o" oo" vo_ i I _J I _ _ CO c >-I I U ¢1 ._ cO '_" I._ n'_ Ohio _:1" I_ Oh e_l ,_1- _ ' C_ _1"

I _1" _ . t_,--I '=_ _ (_0 _ Z

-_ L. _ _ •

• IL. _ . • _.__ .,- _,, _ _,_! _' ,- _- . =_ _o _ _3 E_ __ . o uo _ E _ _" O °_- _" Ll,--t_nl • : "; I ® _l®l==l==lml+_l_l_, == =,_,_,_,= = _ =_-__.,--<., _._"°°° °°° , °1o_ °,._-°'°'I,_1....11,_°,o_'_°_''°r_,XEI g _ _

. 19 2O r

I _, I

O_ Ch I'_. ,--COi ,-4 C)_ Cx_i_O ,--,C) I._ _ C) C%I 0 O_ L-a =- l • • • • • • • • • • • • • • • •

X " 0 'J cr_ ,:_- P'-. ,....4C',,J _) ,,:_'- _D (:X_ CO (',') r... ,:2" ('_ ,....4 L_ C: •,-- kOUr_ gOd_ r-,,0"_ u')_ Oh r,,.oO O0 ,_-Od O0 cO • • • • • • • • • " • • 0

t "t3 Q; :3 I C

Z ,---I,--I or) ,---ICM _'_1 ,--_ ,--* o0 ,--* ,--I (M('O t'_")(M ('_ C:: I 0 .. _ I _ _OC%I _ _-,*L_ r-4_ _ U _lP_l..I • • " • • • • LIC)

e- i ¢1_

ll[l l _'_ _ ("') _--..I _ O0 ,--I O0 ,.-,I l_ c,_ i • l l l

Z I _00 (M r-'_.,:::l- I"_ 0"_ _-., LI_ t..'_C%J_--.t,l:::J(',-v') I.O _ t"_') 0000_,'-4

v ILLJL. ,, • • • ,, • • I • • • I • • • • • •

U 0 "0 r_

rtt;0 • • • • e • • • • • • • • • • • • • • • o

I I

' I-- I Z O0 I._I_.. (3_r'.-l._ ,--_,--__ i._._ C) ,l* i'_ _'-'_CM 0": C_I P-* 0 (_ _ I _ _ll (_l ell _l-i _l-ll Ill (_l . ,_,,, __ . _ "._ ,_ • : .. _ "._ : • . . : _ • : • l_ ('_ r.14 _) I .1-- t-_ I .0 c- i C"JCMO _OCMLI l) (-_-)r,.._.(.,,,0rlll'lC:_' (_-_(_ _l'4_--4('%l('_C_'C::J'_,'14 0'_ I

_" _ eJ oo r% (DO i , I I I

Z I ,_"_')r'--,-_0"_I_,'__- 0'_1" _-"* _0 _-'* _-'_r_ _0 C_I _---, 0,-4 0 I..") k'M I

_J l(,,,,/) O0. O0..:IJ1) CM.....rl_ _--i p_ :- : :-_ •_ • : ., I ....C_l _lO tl- I 0 Z l r_ I

•_, u_ P- (/) (I[3 •

,- =,_,,:, ,_,,,o _ , .,.?, _ o •_ El _.,- _ _- T_ °_I ul c"

a_ = a. u L L _"l_ U .ll-)_lI _I_. _l I = • r- r'l"U _ =1_!_10/_1_-I0_!0 _ I0/_ tl_;°_l" e_')lL4"_ uUIoFI_I_I=I_"I_IE_ ILIL _ lOl0 0 .J I I IIINIoIL.C:_:="_ a_i_i_

23 24 =r

25 ..____._

o 29 U. S. GOVERNMENT PRINTING OFICE: 1982--566-928,"93