oak, white oak, red maple, black Bird Community Responseto cherry, and pitch pine. Dominant groundcover consisted of leaf litter, Timber Stand Improvementand Vacciniumsp., and variousspecies of ferns. Canopyclosure averaged be- tween 55 and 76%. Extent of defolia-

Snag Retention Downloaded from https://academic.oup.com/njaf/article-abstract/7/1/35/4802582 by DigiTop USDA's Digital Desktop Library user on 20 September 2018 tion by gypsymoths on all areaswas <5% duringthe study. TSI thinnings on two areas and H. Lee Stribling,Department of Zoologyand Wildlife Science, TSI + SNAG thinningson the other 311Funchess Hall, AlabamaAgricultural Experiment Station, two areas were conducted during 1983-84. All study areas were thin- AuburnUniversity, AL 36849-5414;Harvey R. Smith, nined to 400-550 /ha and basal NortheasternForest Experiment Station, USDA Service, 52 areas of 24.1-25.7 m2/ha (105.2-112.0 Mill PondRoad, Hamden, CN 06514;and Richard H. Yahner, ft2/ac).Because thinnings were part of a TSI program,the percentageof each Schoolof Forest Resources, The Pennsylvania State University, standthat was cut variedslightly ac- UniversityPark, PA 16802. cordingto numberof treeson a partic- ular studyarea that were dead,dying, malformed,or low-quality.Generally, ABSTRACT.Avian surveyswere con- (6 ft.) tall (Thomaset al. 1979).Dead thinningtook about 20 to 25% of the ductedfrom May-July, 1985 on four study or diseased trees with abundant basal area within each stand. Residual areas in the MoshannonState Forest, numbersof arthropodsare also im- treeheight averaged 19 m on all areas. ClearfieldCo., PA. Wecompared bird pop- portantfeeding sites for somebirds. Avian data and the majority of the ulations and communitiesin Timber Stand TS! cuts that eliminate most of these vegetationdata were collectedin 1985. Improvementareas to TimberStand Im- deadand diseasedtrees may signifi- A totalof 256and 277snags was re- provementareas with snags retained (TSI canfly affectbird communitieswithin servedon the 2 TSI + SNAG study vs. TSI + SNAG, respectively).Bird these standsby removingnesting, sites.These snags were distributedas numbersaveraged 1.7 timeshigher on the roosting,and feedingsites. evenly as possibleover 20-ha areas TSI + SNAG area than on the TSI area. Researchon relationshipsof snags (12.8 and 13.9 snags/ha,respectively) Averagespecies richness also was signifi- to avian populationshas been con- within each of the TSI + SNAG cantlyhigher on TSI + SNAG.Retention ductedin many areasof the United stands. As many snagsas possible of snagswas important to bark-gleaning States (Conner 1978, Evans and were removedduring thinning opera- and cavity-nestingbirds. During TSI Conner1979), but verylittle work has tions on the two TSI stands. A careful operations,forest managers can improve beendone in theNortheast, especially survey of the TSI stands revealed habitatfor cavity-nestingand bark- with regardto the effectsof snagre- somesnags remaining after thinning. gleaningbirds by leaving snags. The costs tentionand timber-management prac- However, TSI + SNAG stands had ofleaving snags is low,and the long-term tices on forest birds. 2.2 timesas many snags and 3.6 times benefitsprovided by insectivorousbirds Our objectiveis to comparebird asmuch basal area in snagsas did the outweighthe costs. populations and communities in TSI standsafter the thinningopera- North.J. Appl.For. 7(1):35•38, March 1990. standssubjected to TSI cuts versus tion. Snagsretained on the TS! + standssubjected to modifiedTSI cuts SNAG areasaveraged 39 cm dbh and with snag retention (TSI + SNAG). 13.6 m tall. An effort was made to se- Silviculturalpractices thatdisturb Results of our studywill be valuableto lect snagsfor retention that already forestunderstory and overstory vege- forestwildlife managers interested in had cavitiespresent. Retained snags tationaffect birds in differentways minimizing the effect of thinning averaged 1.1 cavities per . Re- dependingon a spedes'particular hab- operationson forestbirds. rainedsnags were predominatelyred oak, red maple, black cherry, and itat requirements(Conner et al. 1975, STUDY AREA Titteringtonet al. 1979, Crawfordet pitch pine. The majority of these al. 1981,Conner et al. 1983,Niemi and Thestudy was conducted at•Vlo- snagswas in the 4.0 stageof decay Hanowski 1984). Timber stand im- shannon State Forest, Clearfield Co., based on the series of tree deteriora- provement(TSI) is a practicewhereby Pennsylvania,in 1985. Four study tion (Thomaset al. 1979).In stage4 of forestsare thinnedby the removalof areaswere chosen based on similarity treedecomposition, most major limbs dead,dying, diseased, suppressed, or of elevation(500 to 600 m), slope are missing,and bark is slippingover low-qualitytrees (Smith 1962). Many (about 20%), aspect(southern), and 50% of the tree. treesremoved in TSI operationscould vegetationstructure and composition. METHODS be classifiedas snags.Snags are de- The TSI study areas were 70.4 and fined as any deador partly deadtree 59.5 ha and shared a common border Birdsurveys covered a central10-ha large enoughto be used by birdsfor alongone side700 m in length.The area (1000 x 100 m transect)within nesting or roosting,usually at least TSI + SNAG areas were 46.1 and 59.5 eachstudy site. Survey belts were lo- 10 2 cm (4 in.) dbh and at least1.8 m ha and were locatedaway from the cated as far as possiblefrom study TSI study sites9.7 km and 43.4 km, areaborders to minimizeedge effects. respectively. Surveys were conducted between 0600and 1000EDT, and at leasteight Vegetation 1This researchwas supportedby funds wererun on eacharea from late May- provided by the USDA Forest Service, Timbertypes on all 4 studyareas earlyJuly 1985.Surveys were distrib- NortheasternForest Experiment Station. and on areassurrounding study sites uted betweenstudy areas as evenly in AlabamaAgricultural Experiment Station were very similar. Dominant over- time (hour and day) as possible.All Journal SeriesNo. 15-892157P. story trees consistedof northern red birdssighted or heardwithin 50 m of

NJAF7(1990) 35 either side of the transect center line We found that snagswere impor- areas becausethe snags provided were recordedand mappedby an ob- tant to birdsoccupying bark-gleaning more foraging substrate for bark servermoving at a slow walk (Conner and cavity-nestingguilds. When birds gleanersand a greater number of and Dickson1980). "Fly-overs"were were stratified into foraging and hollow treesfor cavitynesters. excluded. Surveys were not con- nestingguilds (Table 3), bark-gleaning Others have shown also that avail- ducted on days when it rained or foragersand cavitynesters were sig- abilityof snagsin forestsis positively when wind exceeded 20 km/hr. nificantlymore numerous(F = 6.21, correlated with cavity-nesting bird Numbers of individual birds of each df = 30, P < 0.0185,and F = 9.08, df abundance, diversity, and species speciesand numbers of individual = 30, P < 0.0052,respectively) in the richness (e.g., Balda 1975, Yahner TSI + SNAG area. Members of these birdsof all speciescombined were de- 1987). Conversely,removal of snags Downloaded from https://academic.oup.com/njaf/article-abstract/7/1/35/4802582 by DigiTop USDA's Digital Desktop Library user on 20 September 2018 terminedfor eachsurvey. These mea- two guilds, brown creeper, white- has been shown to decreasecavity- suresof avian populationswere ana- breasted nuthatch, and downy and nestingbirds by asmuch as 52% (Scott lyzed using ANOVA procedures(Lu- hairy woodpeckers,may have been 1979) in some areas. ginbuhlet al. 1987). greaterin the TSI + SNAG than TSI Useof snagsis not alwaysrestricted Birdswere organizedinto foraging and nesting guilds (Root 1967). The numberof birdswithin eachguild on Table 2. Averagenumber of birdsseen or heardper surveyat MoshannonState TSI and TSI + SNAG study areas Forest,Clearfield Co., PA,in timberstand improvement with snagsretained (TSI + were comparedusing ANOVA proce- SNAG)versus timber stand improvement (TSI) studyareas from May-July 1985. dures. Study Area Tree diameter,density, and canopy TSI + SNAG TSI closurewere quantifiedin 0.05 ha cir- Area 1 Area 2 Area 1 Area 2 cular plots locatedat 50-m intervals along the 1000-mtransect used for Numberof surveys: 8 8 8 9 bird surveys. Low-growing woody American redstart 1.50 3.44 3.38 2.25 vegetation was quantified in the (Setophagaruticilla) centerof eachplot by countingwoody American robin 0.75 1.56 0.13 1.50 stemspresent within a 1-me circleat (Turdusmigratorius) heightsof 0.1, 0.5 1.0, 1.5, and 2 m Bluejay 0.25 0.44 0.13 -- abovethe ground.Percent canopy clo- ( Cyanocittacristata) Browncreeper 0.75 1.89 0.25 -- surewas estimatedby viewingverti- (Certhia americana) cally through a 5-cm tube. ANOVA Black-cappedchickadee 0.13 1.00 1.50 0.13 procedureswere usedto determineif (Parusatricapillus) differencesexisted in vegetationstruc- Brown-headed cowbird -- 0.11 -- 0.25 turethat couldexplain abundance and (Molothrus ater) distribution of birds on the study Black-throatedgreen warbler -- 0.11 0.88 0.38 areas. (Dendroica virens) Chippingsparrow 2.00 2.67 1.13 1.38 RESULTS AND DISCUSSION (Spizellaaborea) Chestnut-sided warbler 1.13 2.00 0.25 -- Study areaswith snagscontained (Dendroicapensylvanica) more birds and a higher number of Commonyellowthroat 10.38 3.22 0.88 1.63 species than study areas without (Geothlypistrichas) snags. Bird numbers averaged 1.7 Dark-eyedjunco 0.38 0.56 1.25 0.50 timeshigher (F = 37.9, df = 30, P < (Juncohyemalis) 0.0001) on the TSI + SNAG area than Downywoodpecker 0.75 0.78 0.50 0.50 on the TSI area (Table 1). In some (Picoidespubescens) casesspecies, e.g., brown creepers, Easternwood-pewee 1.88 2.56 2.00 2.00 which rely on snagsfor foragingand (Contopus virens) nesting,were more than 7 timesmore Great crestedflycatcher 0.50 0.56 -- -- (Myiarchuscrinitus) plentifulon TSI + SNAGareas (Table Hairywoodpecker 0.63 0.56 0.13 0.25 2). Speciesrichness was also signifi- (Picoides villosus) cantlyhigher (F = 18.9,df = 30, P < Indigobunting 0.63 0.89 1.88 2.25 0.0001)on TSI + SNAG comparedto (Passerinacyanea) TSI (Table 1). Leastflycatcher 0.13 5.56 1.25 5.25 (Empidonaxrainlinus) Ovenbird 0.38 2.00 1.88 2.25 Table 1. Mean number of birds and (Seiurusaurocapillus) meannumber of bird speciesper survey Pileatedwoodpecker 0.38 0.11 -- -- conductedat MoshannonState Forest, (Dryocopuspileatus) ClearfieldCo., Pennsylvania,on timber Red-eyedvireo 4.38 5.44 7.00 5.00 (Vireo olivaceus) standimprovement with snagsretained Rose-breastedgrosbeak 1.88 1.89 0.25 0.38 (TSI + SNAG) and timber stand im- (Pheucticusludovicianus) provement(TSI) studyareas May-July, Rufous-sided towhee 8.25 6.33 2.25 0.50 1985. Numbersin parenthesesindicate (Pipiloerythrophthalmus) numberof surveysconducted. Scarlettanager 2.50 2.78 1.38 0.38 Study Area (Pirangaolivacea) Veery 0.50 1.67 0.13 -- TSI + SNAG TSI ( Catharusfuscescens) Number of birds 54.9 (17)* 31.9 (16) White-breasted nuthatch 2.75 3.33 1.38 2.75 Speciesof birds 18.5 (17)* 13.2 (16) (Sitta carolinensis) thrush 0.25 1.33 -- -- * ANOVA indicatessignificant difference be- tween values in TSI + SNAG versus TSI stands at (Hylocichlamustelina) P < 0.0001. Scientificnames according to Eisenmannet al. (1982).

36 NJAF7(1990) Table3. Averagenumber of birdsper guild on timber stand improvement with snag costsof leaving snags•s low, and retention(TSI + SNAG)versus timber stand improvement (TSI) study areas in Mo- long-termbenefits proreded by insec- shannonState Forest, Clearfield Co., PA,from May-July 1985. Numbersin paren- tivorous birds have been documented thesesindicate number of surveysconducted. in many studies (Baldwin 1958, TSI + SNAG TSI Knight1958, Otvos 1965, 1979, Shook Guild (17) (16) and Baldwin 1970, Koplin 1972, Insectivore Jackson1979, Torgersenand Camp- bell 1982,Torgersen et al. 1983,Tor- Groundgleaner 4.6 2.9 Barkgleaner 6.0a 2.6 gersenet al. 1984).Insectivorous birds

Lower-canopygleaner 15.8c 8.3 canplay a veryimportant role in con- Downloaded from https://academic.oup.com/njaf/article-abstract/7/1/35/4802582 by DigiTop USDA's Digital Desktop Library user on 20 September 2018 Upper-canopygleaner 2.5 2.8 trolling forest-defoliating , Air sallier 5.8 5.3 especiallywhen these insects occur in Omniyore sparsepopulations (Dowden et al Groundforager 10.1c 3.5 1953,Morris et al. 1958,Campbell et Lowercanopy forager 3.0 2.4 al. 1977,Campbell et al. 1983,Smith Uppercanopy forager -- -- Nest 1985). All cavity-nestingand bark- Ground-herb 2.4 2.1 gleaningbirds that occurredin greater Shrub-sapling 24.7c 10.5 numberson TSI + SNAG studyareas Tree twig 10.1 9.7 comparedto TSI studyareas were in- Tree branch 3.8 2.9 sectivorous. Cavity 6.5b 2.6 In this studywe reservedabout 13 ANOVA indicatessignificant difference between TSI + SNAGversus TSI at: a = P < 0.05, b = P < to 14 snags/ha(5 to 6 snags/ac)during 0 01, and c = P < 0.0001. TSI operationsand found up to 7 times as many individualsof some to cavity-nestingbirds. For example, 77, P > 0.488) than the basalarea on bark-gleaning,cavity-nesting species m Virginiawhere snagswere left in a TSI areas(25.7 m2/ha, 112.0ft2/ac). On on those areas. Various speciesof clearcutstand, a varietyof cavity-and the TSI + SNAG, however, the basal snag-usingbirds differ in the number noncavity-dependentbirds were ob- area was distributed in fewer trees/ha of snagsthey requireper unit areaof servedusing the retainedsnags: red- than on TSI (419.5 versus529.5 trees/ land (Thomaset al. 1979).While this tailed hawks (Conner and Adkisson ha, respectively;F = 8.63, df = 77, P studydid notattempt to testthe effect 1974a),eastern bluebirds (Conner and < 0.01).The resultwas a morepatchy different levels of snag retention Adkisson1974b), and woodpeckers and opencanopy on the TSI + SNAG mighthave on bird populationlevels, (Conner and Crawford 1974, Conner areas:47.1% open canopyon TSI + we do feel that forest managerscan et al. 1975). SNAG versus27.3% open canopyon significantlyimprove insectivorous Besidesbark-gleaning and cavity- TSI (P < 0.001, F = 7.56). The more bird habitat by retaining at least as nestingguilds, three other guilds also many snagsas we did in this study. opencanopy allowed additional sun- In addition, we recommend that showed higher numbers in TSI + light to reachthe forestfloor, permit- SNAG comparedto TSI (Table 3). ting the low plantgrowth to flourish. managersattempt to selectsnags for Lower-canopy insectivores and Birdsthat responded to the moreopen retention that already have cavities ground-feedingomnivotes were 1.9 canopyon the TSI + SNAG plotsin- present.This is a positiveindication and 2.9 times more numerous on TSI clude edge or open-canopyspecies, that primarycavity nesters (birds that + SNAGversus TSI study areas. Like- suchas great crestedflycatchers and constructcavities) have found a partic- wise, birds in the shrub-sapling chippingsparrows (e.g., Crawfordet ular tree's size and conditionaccept- nesting guild were 2.3 times more al. 1981,Thompson and Capen1988). able to spend time constructinga abundant on TSI + SNAG areas com- Further,fewer but largertrees in the cavity. A snag with cavitiespresent paredto TSI areas.Greater numbers TSI + SNAG plots provided more will probably provide more "bird- of ground-shrub-saplinginhabiting suitable habitiats for various bark for- years" of use than one in the same b•rdson the TSI + SNAGstudy areas agers,such as downy,hairy, and pi- stageof deteriorationwith no cavities may have resultedfrom the greater leatedwoodpeckers. present. availabilityof low-growingvegetation When planning TSI operations available. There was significantly MANAGEMENT IMPLICATIONS forestmanagers need to weigh these greaterwoody plant materialpresent advantagesof snagretention against from groundthrough 1.5-m on TSI + This studydemonstrates that forest the potentialdisadvantages of snag SNAG than on TSI sites(Table 4). This managers can support a greater retention,such as snagssusceptibility probablyprovided protected and pro- number of cavity-nestingand bark to fireand potential to fall (McClelland gleaningbirds by leavingsnags when and Frissell1975) or to harbor insects ductivelow foragingareas and shrub- and diseases. [] nesting sites for omnivorous,shrub- using standardTSI operations.The nesting species,e.g., rufous-sided towhees, dark-eyed juncos, and chippingsparrows. Other species, like Table4. Mean numberof woodystems at variousheights per 1 m2 sampleplot in common yellowthroats,that feed in MoshannonState Forest, Clearfield Co., PA.Counts conducted on timberstand im- low-canopyand nest in the shrub provementwith snagsretained (TSI + SNAG)and timber stand improvement (TSI) strata, also probablybenefited from studyareas from June-August,1985. the abundance of shrub cover on the TSI + SNAG area. Mean stems/m 2 Thedense growth of lowvegetation Stemheight on the TSI + SNAGareas is probably interval (m) TSI + SNAG TSI df P > F F a responseto the distributionof the 0.0-0.1 25.5 4.1 77 0.0002 15.18 remainingoverstory trees after the TSI 0.1-0.5 3.4 0.4 77 0.0001 19.34 operation.Overstory basal area on the 0.5-1.0 1.3 0.2 77 0.0035 9.05 TSI + SNAG areas (24.1 m2/ha, 105.2 1.0-1.5 0.7 0.2 77 0.0247 5.25 1.5-2.0 0.3 0.2 77 0.1783 1.84 ft2/ac)was no different(F = 0.49,df =

NIAF 7099O) 37 LITERATURE CITED TœRINGTON1981 Songbird population re- OTVOS, I S 1979 The effects of insectivorous sponseto silviculturalpractices in centralAp- bird activitiesin forestecosystems: An evalua- BALDA,R. P. 1975. Vegetation structure and palachia.J. Wildl. Manage.45:680-692. tion. P. 341-374 in The role of insectivorous breedingbird diversity.P. 59-60 in Proc.:A DOWDENP. B., H. A. JAYNœS,AND V. M. C.XR- birds in forestecosystems. Academic Press, symposiumon managementof forest and O•. 1953.The role of birdsin a sprucebud- rangehabitats for nongamebirds. USDA For. worm outbreak in Maine. J. Econ. Entomol. Root, R. B. 1967.The nicheexploitation pattern Serv.Gen. Tech.Rep. WO-1. 46(2):307-312. of the blue-graygnatcatcher. Ecol. Monogr BALDWIN,P. H. 1958. Overwinteringof wood- E•sœ•, E., ETAL. 1982. Thirty-fourth supple- 37:317-350. peckers in -infested spruce-fir ment to the AmericanOrnithologists' Union Scour,V. E. 1979.Bird response to snagremoval forestsof Colorado.Internat. Ornith. Congr. check-list of North American birds. Auk. in ponderosapine. J. For. 77:26-28. 12:71-84. 99(3):1-16. SHOOK,R. S., • P. H. B.•DW•N. 1970. Wood- CAMPBELL,R. W., R. J. SrOAN,• C. E. BU•,AiC. EVANS,K. E., AND R. N. COMER. 1979. Snag peckerpredation on barkbeetles in Engelmann Downloaded from https://academic.oup.com/njaf/article-abstract/7/1/35/4802582 by DigiTop USDA's Digital Desktop Library user on 20 September 2018 1977. Sourcesof mortalityamong late instar management.P. 214-225 in Managementof sprucelogs as relatedto standdensity. Can gypsymoth larvaein sparsepopulations. En- north central and northeastern for non- Entomol. 102:1345-1354. viron. Entomol.6(6):865-871. gamebirds. Workshop Proc., USDA For. Serv. SMrm, D. M. 1962.The practiceof CAMPBELL,R. W., T. R. TORGERSENAND N. SRI- Gen. Tech.Rep. NC-51. Wiley, New York. 578p. VASTAVA.1983. A suggestedrole of predacious JACKSON,J. A. 1979. Insectivorous birds and SMrr•, H. R. 1985.Wildlife and the gypsymoth birdsand antsin the populationdynamics of North Americanforest ecosystems. P. 1-7 in Wildl. Soc. Bull. 13:166-174. the westernspruce budworm. For. Sci.29:770- The role of insectivorous birds in forest eco- T•OMAS,J. W., œT•a,. 1979.Snags. P. 60-77 m 790. systems.Academic Press, Inc. Wildlife habitatsin managedforests: the Blue CONN•R,R. N., • C. S. ADmSSON.1974a. Red- K•GHT, F. B. 1958.The effectsof woodpeckers Mountainsof Oregonand Washington.USDA tailed hawks u 'tflizationof dearcuts. The Raven. on populations of the Englemann spruce For. Serv.Agric. Handb. No. 553.512 p. 45(4):83-84. beetle.J. Econ.Entomol. 51(5):603-607. TITTERINGTON,g. W., H. S. CRAWFORD,AND CAMPBELL,R.W. 1974b. Eastern bluebirds KoPek, J. R. 1972.Measuring predator impact B. N. BUR•SON.1979. Songbird responses to nestingin dearcuts.J. Wildl. Manage.38:934- of woodpeckerson sprucebeetles. J. Wildl. commercialclear-cutting in Maine spruce-fir 935. Manage.36(2):308-320. forests.J. Wildl. Manage.43:602-609. CAMPBELL,R. W., ,•a•v H. S. CP•W•ORD. 1974. LUGINBUHL, R. C., S. D. SCHLOTZHAUER,AND THOMPSON,F. R. III, AND D. E. CAPEN.1988 Woodpeckersforaging in Appalachiandearcuts. J. C. PARKER.1987. SAS/STAT guide for per- Avianassemblages in seral stages of a Vermont J. For. 72:564-566. sonalcomputers. Sas Institute Inc., Cary, NC. forest.J. Wildl. Manage.52:771-777. CAMPBELL,R. W., •T AL. 1975. Woodpecker 1028p. TORGERSEN,T. R., ,•a•D R. W. CAMPBELL.1982 nestinghabitat in cut and uncutwoodlands in MCCLEL!._•ID, B. R., AND S.S. FPaSS•LL.1975. Someeffects of avianpredators on the western Virginia.J. Wildl. Manage.39:144-150. Identifyingforest snags useful for hole-nesting sprucebudworm in northcentral Washington Environ. Entomol. 11:429-431. CAMPBELL,R. W. 1978. Snag managementfor birds.J. For. 73(7):414-417. TORGERSEN,T. g., g. g. MASON, AND H. G cavity nestingbirds. P. 120-128 in Manage- MORRIS, R. F., œTAL. 1958. The numerical re- PAUl.1983. Predation on pupaeof Douglas-fir ment of southernforests for nongamebirds. sponseof avian and mammalianpredators tussockmoth, Orgyiapseudotsugata (McDun- WorkshopProc., USDA For. Serv.Gen. Tech. during a gradationin the sprucebudworm. Ecol. 39:487-494. nough)(Lepidoptera: Lymantriidae). Environ Rep.SE-14. Entomol. 12:1678-1682. C,•PnELr,R. W., • J. G. DicIcsoN.1980. Strip NIœM•,G. J., • J. M. HANOWSm.1984. Rela- TORGERSEN,T. R. ETAL. 1984. Avian predators of transectsampling and analysisfor avianhab- tionshipsof breedingbirds to habitatcharacter- itat studies. Wildl. Soc. Bull. 8:4-10. Douglas-firtussock moth, Orgy/a pseudotsugata isticsin loggedareas. J. Wildl. Manage.48: (McDunnough)(Lepidoptera: Lymantriidae) C•,,iPnœ•,•,R. W., J. G. DICKSON,B. A. LOCK•, 438-4'k3. in southwesternOregon. Environ. Entomol • C. A. SEGELQraST.1983. Vegetation char- Orvos, I. S. 1965.Studies on avianpredators of 13:1018-1022. acteristicsimportant to commonsongbirds in Dendroctonusbrevicomis, Le Conte(Coleoptera: YAHNER,R. H. 1987.Use of even-agedstands by east Texas. Wilson Bull. 95:349-361. Scolytidae)with specialreference to Picidae. winter and spring bird communities.Wilson CRAWFORD,H. S., R. G. HOOPER,AND R. W. Trr- Can. Entomol.97(11):1184-1199. Bull. 99:218-232.

planted in every stateof the United Racial Diversity of Black Locustin Statesand abroad,particularly in Ger- many, and is consideredone of the mostwidely planted trees in the world Growth Rate and in Susceptibility (Cuno 1930, Feher 1935, Blumke 1950). Maryland alone plants about to the LocustTwig Borer 100,000black locust trees annually. Blacklocust is a highlyvalued tree from severalaspects. As a memberof JohnB. Genysand Dan M. Harman,The University of the familyLeguminosae, it is capableof Maryland,Frostburg, MD 21521 addingnitrogen to the soilby the pro- cessof associatednitrogen-fixing bac- teria. This qualifiesit for reclamation ABSTRACT:Twenty-five different geo- the mostproductive source had the kiln- of nitrogen-poor strip mine spoil graphicsources of blacklocust trees from dry weightsequivalent to 70,000kg/ha. banks.Wood of blacklocust is highly Ohio,Maryland, West Virginia, and Vir- This suggeststhat selectedblack locust notedfor its hardness,strength, and giniawere studied for 3 yearsin a repli- sourcesmay qualifyfor short-rotation durability(Raber 1936, Cuno 1930).It catedexperiment in Maryland. Significant massyield of energy. The most out- is highly resistantto rotting and has differenceswere found in heights,diam- standingin kiln-dry weightper square long been used for outdoor framing eters,kiln-dry weights, and susceptibility growth spacewere somesources from and fenceposts (Cope 1942, Ginther levelsto locusttwig borer.None of these Tucker,Randolph, and Mineral counties 1933).Maryland foresterscan readily variableswere related to either longitude or of WestVirginia. sell its pole-sizedcuts at three times latitiude of the seedsource. However, North.J. Appl.For. 7(1):38-43, March 1990. the value of other speciesof similar high-elevationsources included more rap- age. By policy, one nearby large wood-using corporation considers idly growingpopulations than those from black locust too valuable for use as lowelevations. The third-year sprouts of Blacklocust isa deciduoustreena- pulpwood.Instead, it is marketedfor tive to the eastern United States. Its rustic fence rails and posts (pers naturalrange extends along the Appa~ comm., Bruce Brenneman, Westvaco, i Scientific Contribution No. 2029-AEL, lachianMountains from Pennsylvania Inc.). Many enterprisesin the Appala- CEES,University of Maryland,Frostburg, to Georgia,and west to Missouriand chian region depend solely on MD 21532. Oklahoma. In addition, it has been productsof blacklocust. Much infor-

38 NJAF7(1990)