Nineteenth Century LumberSurveys for Bangor, Maine:Implications for Pre-EuropeanSettlement Forest Characteristics inNorthern and Eastern Maine

Jeremy S. Wilson

In the mid-19th century, Bangor, Maine, was a major center for the manufacturing and shipping oflumber, leading many to proclaim Bangor as the “lumber capital of the world.” The raw material forthis lumber was transported down the Penobscot River, which drains the largest watershed in Maine.Starting in 1832, lumber put up for sale and shipped from Bangor was surveyed. Between 1832 and1872, over 3.36 billion board feet (bbf) of white pine lumber was tallied. In the 19th century, therewere numerous inefficiencies and losses involved in the harvesting, transporting, and processing thatconverted a standing pine tree to lumber ready to be loaded on a ship; therefore, the 3.36 bbf surveyedprovides only a minimum estimate for the volume of white pine in the pre-European settlement forestof northern and eastern Maine. Even this minimum volume estimate for pine suggests a magnitude ofthis resource that is inconsistent with descriptions developed from early land survey informationand suggests that current interpretations of the pre-European disturbance regime may needre-evaluation.

Keywords: white pine, lumber records, pre-European settlement, natural disturbance regime

I n the mid-19th century, Bangor,Maine, rose to prominence as a majorcenter for lumber manufacture and ex-

port in the northeast United States (Figure1). Situated on a tidewater of the PenobscotRiver, Bangor was the shipping port for thelargest watershed in Maine starting in thelate 18th century. River drives on the Penob-scot and its tributaries provided log trans-portation from a forest area of nearly twomillion hectares (Figure 2). Starting in 1832,lumber put up for sale and shipped fromBangor was surveyed by the Surveyor Gener-

al’s Office for Penobscot County (Defenbaugh1906, Smith 1972). The amount of pine (pri-marily white pine [Pinus strobus L.]) lumbersurveyed in Bangor was considerable, andpeaked in the middle of the 19th century (De-fenbaugh 1906). The large volume of whitepine lumber manufactured from logs cut in thePenobscot watershed has implications for un-derstanding the character and composition ofthe pre-European settlement forests of north-ern and eastern Maine. The presence of whitepine in forest stands is often attributed to re-generation events after major canopy distur-

bances such as fires, windthrow, and aban-doned agricultural fields (Abrams 2001).White pine is a moderately long-lived spe-cies (Wendell and Smith 1990, Abrams2001); therefore, the persistence of pine instands preserves a record of relatively intensedisturbance events in the regions past.

Based on early 19th century land surveyrecords for a 1.65 million hectare area innorthern and eastern Maine that had consid-erable overlap with the Penobscot water-shed, Lorimer (1977) found that only asmall proportion of witness trees were whitepine. The dominance of witness trees by rel-atively shade-tolerant species such as spruce(Picea spp.), yellow birch (Betula alleghensisBritton), beech (Fagus grandifolia Ehrh.),and balsam fir (Abies balsamea [L.] Mill.) ledLorimer (1977, 1980) to conclude that thepre-European settlement forests were largelyin a “climax state,” and that the interval be-tween major wind and fire disturbanceevents was quite long in this region. Lumbershipping records that summarize productionfor a large region, such as the PenobscotRiver watershed, provide one method ofcritically evaluating estimates of white pineabundance and natural disturbance regimedeterminations developed from the earlyland survey records.

218 Journal of Forestry • July/August 2005

AB

ST

RA

CT

Lumber Surveyed in Bangor,Maine, during the 19th Century

Europeans first settled the Bangor re-gion in the 1770s. The Penobscot River andthe forest it drained were a major draw forsettlers; however, by 1816, only one millionboard feet (bd ft) of lumber were shippedannually from Bangor (Roberts 1875, De-fenbaugh 1906). A surge of new settlerscame into the Penobscot Valley in the1820s, and by 1831, more than 30 millionbd ft of lumber were shipped annually fromBangor (Roberts 1875, Defenbaugh 1906,Stanley 1963). Estimates are that at least 200million bd ft of lumber had been shippedfrom Bangor before 1832 (Whitman 1873,Roberts 1875, Defenbaugh 1906). Figure 3shows the volume of lumber manufacturedfrom different tree species that was surveyedin Bangor between 1832 and 1872. Thepeak for white pine lumber surveyed oc-curred in the mid-19th century. During theperiod from 1832 to 1872 over 3.36 billionboard feet (bbf) of white pine lumber weresurveyed in Bangor. By the 1860s, much ofthe readily available large white pine in thePenobscot River watershed had been logged,and spruce had become the largest compo-nent of lumber manufactured and shippedfrom Bangor. By 1880, the era of large pinelogging on the Penobscot was essentiallyover, and the bulk of pine surveyed was “sec-ond growth” and used for making pineboxes (Defenbaugh 1906, Smith 1972).

Inefficiencies and Losses. In the mid-19th century, there were numerous ineffi-

ciencies and losses involved in the harvest-ing, transporting, and processing thatconverted a standing pine tree to lumberready to be loaded on a ship (Coolidge 1963,Smith 1972). These inefficiencies and lossesmean that volumes of lumber surveyed inBangor provide only a minimum estimate;actual standing volumes of white pine in thepre-European settlement forest must havebeen considerably higher.

Volume Left in the Woods. Merchant-able log diameters and lengths as well asquality requirements of the mid-19th cen-tury meant that many white pines were con-sidered too small or defective and thus leftstanding or lying the woods. Coolidge(1963) reports that nearly 50% of large pineshad substantial rot or other defects. In addi-tion, large sections of cut trees that would beconsidered merchantable now were left inthe woods because of small sizes or defects.Conversely, some of the largest pine logswere too big to move or to drive in smallerstreams, so they were left to rot in the woods(Springer 1851, Houpt 1964). Large whitepines were also left standing in areas that weretoo far from drivable water or were too steep,making harvesting and transporting them dif-ficult (Defenbaugh 1906, Houpt 1964).

Volume Lost in Transit and Milling.Merchantable logs were harvested in au-tumn and winter and were piled near driv-able water bodies to be transported down-stream during the spring snowmelt. Instreams, rivers, and lakes, some portion ofthe logs sank during transit to and storage at

collection points called log booms. Accord-ing to Cayford and Scott (1964), “Time andeffort were not wasted on second-grade ma-terial or any logs sunk while floating to themill.” It is not clear how much pine sank;however, a pine salvage operation recovered2 million bd ft of sunken logs from a singlesite on the Penobscot River in the 1960s(Cayford and Scott 1973). There were fourpermanent and several temporary collectingbooms as well as many storage booms forindividual mills in and north of Bangor onthe Penobscot River (Hempstead 1931,Smith 1972). These log booms sometimesfailed, especially under pressure from flood-waters or ice flows, allowing a few or manylogs to float downriver and out to sea(Houpt 1964, Smith 1972).

Nineteenth century manufacturing oflumber generated another point of ineffi-ciency. According to Defenbaugh (1906),“As much pine was wasted in the flush daysas is cut now, and the slabs that were thrownaway were as thick as deal planks.” Dealplanks are squared timbers from which stan-dard dimension lumber is cut (Houpt1964). In addition, lengths of imperfectlumber called “scoots” were thrown into theriver, and their volume not recorded (Houpt1964). An 1868 address to the legislature byGovernor Chamberlain illustrates the vol-ume of lumbering waste thrown in the river.

A bed of obstructions from slabs, edgings,and sawdust extends from Crosby’s Nar-rows to Bangor, a distance of some threeand one-half miles, covering an area ofabout 320 acres, and of an average depth of10 feet, being in some localities more than18 feet deep, forming an entangled mass ofmore than 5,000,000 cubic yards.

Nineteenth century sawmills wereprone to fires, and frequent mill conflagra-tions consumed a considerable quantity ofsawn lumber (Smith 1972). Transportingthe manufactured lumber to Bangor fromupriver mills required building lumber raftsand additional river transport, resulting infurther losses (Hempstead 1931, Wood1935, Houpt 1964). Even when rafts oflumber arrived in Bangor, additional lossesoccurred. According to Defenbaugh,“When a crew of stevedores got down to themuddy bottom of a raft they would set itadrift sooner than handle it or take the trou-ble of washing it.” A report by the Harbor-master of Bangor in 1862 describes lumbertheft, yet another point of volume loss priorto surveying.

Figure 1. Log boom on the Penobscot River near Bangor, Maine (Bangor Public Library).

Journal of Forestry • July/August 2005 219

There is a practice extensively followed inour harbor by a class of persons called“wreckers” of pilfering lumber of variouskinds from rafts, which, in the aggregate,amounts to quite a large sum in value.

Volume Not Measured. The tally oflumber surveyed in Bangor does not includelumber volumes used locally or lumber man-ufactured outside of Penobscot County(e.g., below Bangor on the Penobscot). Fur-thermore, substantial harvesting of largepine in the Penobscot watershed took placein the decades before 1832 (estimates sug-gest 200 million bd ft) and continued foranother decade beyond 1872 (150 millionbd ft; Defenbaugh 1906, Smith 1972).There may have been some double countingin the Bangor lumber survey (Whitman1873). This could have occurred when lum-ber was surveyed but did not get shipped.

This lumber might be surveyed again—double counted—the next time it was pre-pared for shipment. Harvested volumes oflarge pines occurring before and after the1832 to 1872 period would balance the 10%to 15% double counting that has been esti-mated for the survey (Whitman 1873).

Growth and Interwatershed LogTransfers. A complicating factor in using 40years of lumber records to evaluate standingwood volume across a large watershed is thevolume growth of trees that occurred duringthe recording period. There would certainlyhave been growth of white pine volume inthe watershed between 1832 and 1872;however, the early logging focused on re-moving large pines that were growing rela-tively slowly (Defenbaugh 1906, Coolidge1963). The net volume growth of these large

trees would have been greatly reduced by de-cay, mortality, and disturbance (e.g., wind,fire, and lightning) losses occurring over thesame period. Substantial net growth of whitepine volume could have occurred if therewere a considerable component of interme-diate-sized relatively fast-growing trees inthe pre-European settlement forest; how-ever, some of these trees would have neededto attain large size within the 40-year periodfor that volume growth to become mer-chantable. This scenario seems unlikely be-cause pine harvests peaked in the 1850s, andby the turn of the century (1903), the MaineForest Commissioner estimated that therewere only 153 million bd ft of pine standingin the Penobscot river system (Defenbaugh1907).

Another confounding factor is thetransfer of logs between watersheds inMaine. The Penobscot gained some logsfrom areas in the headwaters of the Allagashthrough the Telos Canal, but also lost logs tothe Kennebec (Defenbaugh 1906, Smith1972, Rolde 2002). Defenbaugh (1907) re-ports, “By means of a sluice at NorthwestCarry a large number of logs cut on thenorth and South branches of the Penobscotare conveyed into Moosehead Lake, towedacross the lake and thence sent down theKennebec.” Without more detailed infor-mation, a balanced interwatershed transfer isassumed.

A Minimum Benchmark and aRough Estimate. Estimates deriving howmuch standing volume of large pine wasneeded to survey shipments of more than3.36 bbf of pine lumber from Bangor be-tween 1832 and 1872 require multiple as-sumptions and are prone to alternative sub-jective interpretations. Recognizing theselimitations, rough estimates reflecting inef-ficiencies and losses associated with 19thcentury lumber production can providesome sense for the magnitude of differencebetween lumber surveys and standing vol-ume. For example, we could conservativelyassume that 50% of pine volume never leftthe woods (Coolidge 1963), 10% was lostduring transit, 10% was wasted in the saw-mills (above deductions assumed in the in-ternational one-quarter-inch log rule), and10% was used locally or manufactureddownriver. These estimates for inefficienciesand losses sum to 80% and generate a stand-ing pine volume of approximately 6 bbf.The surveyed lumber records from Bangorprovide a solid minimum benchmark—3.36 bbf of pine lumber. The rough estimate

Figure 2. Penobscot River watershed area (Maine Rivers 2003: www.mainerivers.org). ThePenobscot drains roughly 1.871 million hectares of forested land above Bangor. The totalarea of watershed is 2.219 million hectares, 2.057 million hectares above Bangor (NationalOceanic and Atmospheric Administration 2003: spo.nos.noaa.gov/projects/gomaine/gome wtshd.html), and forested land accounts for 91% of the total land area in northern andeastern Maine (Laustsen 2003).

220 Journal of Forestry • July/August 2005

of standing pine volume, 6 bbf, providessome sense for the magnitude of the pre-European Penobscot River watershed pineresource that was required to manufacturethat much lumber.

Comparison with 2001 InventoryLevels. Forest inventory information forMaine in 2001 is available for comparisonwith estimated historical levels of white pinein the Penobscot River watershed (Laustsen

and Griffith 2002). Large trees (50 cm dbh)dominated the white pine harvest in themiddle of the 19th century (Woods 1935,Stanley 1963, Houpt 1964, Smith 1972).The category of large sawtimber in the 2001forest inventory includes trees �38.1 cmdbh; therefore, the current large sawtimbercategories include some smaller trees thanthose typically harvested between 1832 and1872. Figure 4 provides a comparison of2001 standing volumes of white pine withthe 1832 to 1872 lumber survey volume andthe rough standing volume estimate for thePenobscot watershed. The 3.36 bbf of lum-ber surveyed in Bangor suggest that the vol-ume of large pine in the pre-European set-tlement forest of the Penobscot watershedwas at an absolute minimum more thanthree times greater than today’s large saw-timber pine volumes in the eastern andnorthern regions of Maine. The Penobscotwatershed includes roughly equal areas inboth of these regions. White pine currentlyrepresents 51% and 22% of conifer volumein large sawlogs in the eastern and northernregions, respectively (Laustsen and Griffith2002). The 6 bbf standing volume estimatesuggests that the pre-European settlementPenobscot forest had standing volumes oflarge pine that exceeded current forest con-ditions in southern Maine, where aban-doned agricultural fields and forest fires havegenerated considerable amounts of whitepine.

Comparison with Analysis of LandSurvey Information. Lorimer (1977)found that only 1.3% of the witness trees inland surveys of northeastern Maine werepine; however, he did note a discrepancy be-tween the proportion of pine witness treesand the frequency that pine was mentionedin the land surveyors’ descriptive lists of veg-etation. In these descriptive lists, pine wasthe third most commonly mentioned coni-fer after spruce and cedar. Lorimer (1977)downplayed the importance of pine in thesedescriptive lists because records from 1826to 1846 timber cruises of 21 townships in oradjacent to his study area produced an aver-age merchantable pine volume of 6.7 mil-lion bd. ft. per township. This correspondsto one large pine (1,000 bd ft per tree) forevery 1.4 hectares (Lorimer 1977). The Pe-nobscot watershed above Bangor encom-passes an area the size of roughly 220 town-ships. Using the average pine volumes fromthe 1826 to 1846 timber, cruises would pro-vide an estimate of standing pine volumesfor the entire watershed of roughly 1.4 bbf.

Figure 3. Lumber surveyed in Bangor, Maine, from 1832 to 1872 (City of Bangor variousdates, Whitman 1873, Roberts 1875, Defenbaugh 1907, Smith 1972). The logs for thismaterial were driven to the Bangor area on the Penobscot River. The first spruce logging onthe Penobscot River was reported in 1845 (Defenbaugh 1907, Hempstead 1931, Smith1972). Before 1851, records do not distinguish between species. Pine proportions havebeen estimated for 1845 to 1850 by extrapolating post-1851 trends in species proportions.

Figure 4. Standing volume of large pine (38.1 cm dbh) if current inventory estimates(average volume of large pine per hectare) from different regions in Maine are used topopulate a Penobscot River watershed-sized area (1.871 million hectares of timberland)with pine (Laustsen and Griffith 2002, Laustsen 2003). An asterisk indicates that standingpine volume estimate assumes that inefficiencies and losses during the harvest, transport,and processing of lumber amounted to 80% of total volume.

Journal of Forestry • July/August 2005 221

This is less than one-half of the 3.36 bbf ofpine lumber surveyed, yet still representsmore large pine sawtimber than is currentlyfound in a combination of eastern andnorthern Maine (Figure 4). These early 19thcentury timber cruises included some town-ships where harvesting had already takenplace (Lorimer 1977). In fact, by 1846, 1.38bbf of pine lumber had already been sur-veyed in Bangor. The timber cruises alsorepresent a small proportion of the water-shed, and were probably conservative esti-mates (Lorimer 1977). These factors couldhelp explain the difference between esti-mates from surveyed lumber records andearly timber cruises.

In his evaluation of the timber cruise in-formation, Lorimer (1977) described lownumbers of pine when average volume esti-mates were spread evenly across a township-sized area. It seems more reasonable to assumethat there would be some concentrations oftree species between or across townships givenpronounced variation in soil, topography, anddisturbance history found throughout the re-gion. In fact, Lorimer (1977) reported thatthe merchantable pine estimated in timbercruises ranged from 600,000 to 30 millionbd ft across the 21 townships he examined.The tremendous range in pine volumes re-ported across the township suggests thatconcentrations of pines occurred at a varietyof spatial scales. The prominence of pine inthe land surveyor’s descriptive lists would beconsistent with estimates from the Bangorlumber survey records if volumes of largepine were concentrated on some portion(e.g., 10%–30%) of a township or the entirewatershed area. Using the 3.36 bbf mini-mum benchmark to populate 15%, or the 6bbf standing volume estimate to populate27% of the Penobscot River watershed,white pine numbers per hectare would stillbe relatively low (i.e., 12 trees per hectare);however, 12 large pines (1,000 bd ft per treeor roughly 70 cm) could easily have a crownprojection area equivalent to 10–15% of ahectare, representing an important compo-nent of any forest canopy (Figure 5).

Surveyor Bias. Early land surveyorsmay have been biased against choosingwhite pines as witness trees for many rea-sons, including reluctance to impact high-value trees, alternate species preferences, un-equal sampling based on topography, andpreferences for mid-sized (20–40 cm dbh)trees (Lorimer 1977, Manies and Mladenoff2000, Black and Abrams 2001, Schulte andMladenoff 2001). More importantly, simple

proportions of different tree species, such asthose created from witness tree counts, maynot accurately reflect the significance of aspecies that is found in relatively low num-bers but includes many large dominant in-dividuals (Figure 5).

Why Was There Pine in thePre-European Settlement Forest?

The presence of white pine in foreststands has long been attributed to its regen-erative success and rapid growth rates inopenings created by major canopy distur-bances (Nichols 1935, Cline and Spurr1942). Disturbance events associated withhistoric pine regeneration include fires,windthrow, windthrow followed by fire, andabandoned agricultural fields or pastures(Lutz 1930, Henry and Swann 1974, Foster1988, Abrams 2001). White pine is consid-ered intermediate in shade tolerance and ismost often found to have originated aftermajor canopy disturbances (Lutz 1930,Cline and Spurr 1942, Wendell and Smith1990, Abrams 2001). In certain situations,two or more age cohorts of white pine havebeen found in a stand after more moderatecanopy disturbances; however, these tend tobe in areas where white pine is one of themore shade-tolerant species (Braun 1950,Holla and Knowles 1988, Abrams 2001). Innorthern and eastern Maine, many speciesthat occur with white pine are consideredvery shade tolerant, making them more suc-

cessful competitors in canopy gaps createdafter less intense disturbance events.

The volume of pine lumber surveyed inBangor during the mid-19th century, early19th century land surveyors’ descriptive listsof vegetation, and even timber cruise infor-mation from the first one-half of the 19thcentury suggest that pine was a substantialcomponent of the pre-European settlementforest in the Penobscot watershed. Estimatesof standing volume based on the 19th cen-tury lumber surveys produce white pine vol-umes that would easily represent an impor-tant constituent of the forest canopy overlarge portions of the watershed (15–30%).The presence of considerable white pine im-plies that major canopy disturbances were afactor influencing forest dynamics and shap-ing forest characteristics in pre-Europeansettlement in northern and eastern Maine.Calculating an average return frequency formajor canopy disturbances would be inaccu-rate based on the limited single-species vol-ume information available and misleadinggiven the varied topography and soils of theregion (Lorimer and White 2003). Compar-isons with current forest inventory estimatessuggest that the harvesting and managementpractices of the 19th and early 20th centuryregenerated substantially less pine relative tothe pre-European settlement disturbance re-gime. This implies that before European set-tlement, the watershed experienced pulses ofmore intense canopy disturbance eventsthan those related to early harvesting tech-

Figure 5. Stand visualization representing 0.4047 hectares. Density of emergent white pine(1,000 bd ft per tree) is equivalent to 12 trees per hectare. Minimum estimates of large whitepine from lumber survey records generate this amount of white pine across 15% of thePenobscot watershed. Pine volume estimates assuming 80% inefficiencies and losses inlumber manufacture generate this same amount of white pine across 27% of the watershed.

222 Journal of Forestry • July/August 2005

niques (high grading and then diameter-limit cutting) in these forests.

The Penobscot River watershed pineharvest is not unique in the region; substan-tial white pine harvests in the 18th and 19thcenturies occurred on every large river sys-tems in Maine, including the Saco, An-droscoggin, Kennebec, St. John, and theMachais (Defenbaugh 1906, Wood 1935,Coolidge 1963, Smith 1972, Judd 1989).

Increasingly in Maine and many otherregions, there are calls for adjusting forestmanagement to more closely emulate thepre-European settlement disturbance re-gime. Current interpretations of the pre-Eu-ropean settlement disturbance regime innorthern and eastern Maine suggest long re-turn intervals between major fire and windevents, and that forest dynamics and charac-teristics were driven by minor to moderatecanopy disturbances and subsequent gapcolonization (Lorimer 1977, Lorimer 1980,Seymour 1992, Seymour et al. 2002,Lorimer and White 2003). These current in-terpretations of the pre-European settle-ment disturbance regime for northernMaine should be re-evaluated to ensure thatthey do not contradict the presence of sub-stantial white pine in the pre-European set-tlement forest. The large volume of whitepine in the mid-19th century lumber surveyrecords for Bangor, Maine, suggest that be-fore European settlement, major canopy dis-turbance events influenced forest dynamicsand shaped forest characteristics across asizeable portion of the Penobscot River wa-tershed.

Literature CitedABRAMS, M.D. 2001. Eastern White Pine versa-

tility in the presettlement forest. Bioscience51(11):967–979.

BLACK, B.A, AND M.D. ABRAMS. 2001. Influencesof Native Americans and surveyor biases onmetes and bounds witness-tree distribution.Ecology 82(9):2574–2586.

BRAUN, E.L. 1950. Deciduous forests of easternNorth America. Hafner Press, New York. 596p.

CHAMBERLAIN, J.L. 1868. Governor Chamber-lain’s address to the legislature. Stevens and Say-word Printers to the State. Augusta, ME. 46 p.

CAYFORD, J.E., AND R.E. SCOTT. 1964. Underwa-ter logging. Cornell Maritime Press Inc., Cam-bridge, MD. 81 p.

CAYFORD J.E., AND R.E. SCOTT. 1973. Underwa-ter logging. N. Logger 21(12):42–46.

CITY OF BANGOR. 19th century various dates. Theannual reports. Various printers to the city.

CLINE, A.C., AND S.H. SPURR. 1942. The virginupland forest of central New England. Har-vard For. Bull. 21:1–58.

COOLIDGE, P.T. 1963. History of the Mainewoods. Furbush Roberts Printing Co. Inc.,Bangor, ME. 805 p.

DEFENBAUGH, J.E. 1907. History of the lumber in-dustry of America, volume 2. The AmericanLumberman, Chicago, IL. 655 p.

FOSTER, D.R. 1988. Disturbance history, com-munity organization and vegetation dynamicsof the old-growth Pisgah Forest, southwesternNew Hampshire, USA. J. Ecol. 76:105–134.

HEMPSTEAD, A.G. 1931. The Penobscot boom andthe development of the west branch of the Penob-scot River for log diving. University Press,Orono, ME. 187 p.

HENRY, J.D., AND J.M.A. SWANN. 1974. Recon-structing forest history from live and deadplant material—an approach to the study offorest succession in southwest New Hamp-shire. Ecology 55:772–783.

HOLLA, T.A., AND P. KNOWLES. 1988. Age struc-ture of a virgin white pine, Pinus strobus, pop-ulation. Can. Field-Nat. 102:221–226.

HOUPT, W.P. 1964. Maine long logging and itsreflection in the works of Holman Francis Day.PhD dissertation, University of Pennsylvania,Philadelphia, PA. 384 p.

JUDD, R.W. 1989. Aroostook: A century of loggingin northern Maine. University of Maine Press,Orono, ME. 351 p.

LAUSTSEN, K.M., AND D.M. GRIFFITH. 2002.Third annual inventory report on Maine’s forests.Maine Department of Conservation ForestHealth and Monitoring Division and U.S. De-partment of Agriculture Northeastern Re-search Station. Augusta, ME. 34 p.

LORIMER, C.G. 1977. The presettlement forestand natural disturbance cycle of northeasternMaine. Ecology 58(1):139–148.

LORIMER, C.G. 1980. The use of land surveyrecords in estimating presettlement fire fre-quency. Proc. P. 57–62 in Fire History Work-shop, October 20–24, 1980. Tucson, AZ.Stokes, M.A., and J.H. Dieterich (eds.).USDA Forest Service GTR, RM-81. 142 p.

LORIMER C.G., AND A.S. WHITE. 2003. Scale andfrequency of natural disturbances in the north-eastern US: implications for early successionalforest habitats and regional age distributions.For. Ecol. Manage. 185:41–64.

LUTZ, H.J. 1930. The vegetation of Heart’s Con-tent, a virgin forest in northwest Pennsylvania.Ecology 11:1–29.

MANIES, K.L., AND D.J. MLADENOFF. 2000. Test-ing methods to produce landscape-scale pre-settlement vegetation maps from the U.S. pub-

lic land survey records. Land Ecol.15:741–754.

NICHOLS, G.E. 1935. The hemlock-white pine-northern hardwood region of eastern NorthAmerica. Ecology 16:403–422.

ROBERTS, C.P. ED. 1875. Bangor Business Alma-nac for 1875. O.F. Knowles and Co., Bangor,ME. 84 p.

ROLDE, N. 2001. The interrupted forest: A historyof Maine’s wildlands. Tilbury House Publish-ers, Gardiner, ME. 402 p.

SCHULTE, L.A., AND D.J. MLADENOFF. 2001. Theoriginal US public land survey records. J. For.10:5–10.

SEYMOUR, R.S. 1992. The red spruce-balsam firforest of Maine: evolution of silvicultural prac-tice in response to stand development patternsand disturbances. In The ecology and silvicul-ture of mixed-species forests: A festschrift forDavid M. Smith, Kelty, M.J., B.C. Larson, andC.D. Oliver (eds.). Kluwer Academic Publish-ers, Boston, MA. 287 p.

SEYMOUR, R.S., A.S. WHITE, AND P.G. DE-MAYNADIER. 2002. Natural disturbance re-gimes in northeastern North America: evaluat-ing silvicultural systems using natural scalesand frequencies. For. Ecol. Manage. 155:357–367.

SMITH, D.C. 1972. A history of lumbering inMaine 1861–1960. University of Maine studiesno. 93. University of Maine Press, Orono, ME.469 p.

SPRINGER, J.S. 1851. Forest life and forest trees:comprising winter camp-life among the loggers,and wild-wood adventure. With descriptions oflumbering operations on the various rivers ofMaine and New Brunswick. New York, Harperand Brothers. 259 p.

STANLEY, R.D. 1963. The rise of the Penobscotlumber industry. Master’s thesis, University ofMaine, Orono, ME. 81 p.

WENDELL, G.W., AND H.C. SMITH. 1990. East-ern white pine. Silvics of North America, vol 1:Conifers. USDA Agricultural Handbook 654,Burns, R.M., and B.H. Honkala (eds.). Wash-ington, D.C.

WHITMAN, W.E.S. 1873. The wealth and industryof Maine for the year 1873: First annual report.Sprague, Owen and Nash Printers to the State.Augusta, ME. 454 p.

WOOD, R.G. 1935. A history of lumbering inMaine 1820–1861. University Press. Orono,ME. 267 p.

Jeremy S. Wilson (Jeremy Wilson@umenfa.maine.edu) is assistant professor and Irvingchair for forest ecosystem management, Depart-ment of Forest Management, University ofMaine, Orono, ME 04469-5755.

Journal of Forestry • July/August 2005 223