Northeast Historical Archaeology

Volume 13 Article 6

1984 Log to Light Rails: The volutE ion of Main and Transportation in Buffalo, Michael A. Cinquino

Marvin G. Keller

Carmine A. Tronolone

Charles E. Vandrei Jr.

Follow this and additional works at: http://orb.binghamton.edu/neha Part of the Archaeological Anthropology Commons

Recommended Citation Cinquino, Michael A.; Keller, Marvin G.; Tronolone, Carmine A.; and Vandrei, Charles E. Jr. (1984) "Log Roads to Light Rails: The Evolution of Main Street and Transportation in Buffalo, New York," Northeast Historical Archaeology: Vol. 13 13, Article 6. https://doi.org/10.22191/neha/vol13/iss1/6 Available at: http://orb.binghamton.edu/neha/vol13/iss1/6

This Article is brought to you for free and open access by The Open Repository @ Binghamton (The ORB). It has been accepted for inclusion in Northeast Historical Archaeology by an authorized editor of The Open Repository @ Binghamton (The ORB). For more information, please contact [email protected]. Log Roads to Light Rails: The volutE ion of Main Street and Transportation in Buffalo, New York

Cover Page Footnote This article is a condensed and edited version of a report submitted to the Niagara Frontier Transportation Authority (NFTA) in June 1981 (Ecology and Environment, Inc. 1981). Research on the Buffalo Log was conducted with public funds provided through PL 93-291. The NFTA and the engineering firm of Cordy, Carpenter, Dietz and Zack deserve special thanks for their cooperation and logistical support. Ecology and Environment, Inc. in particular Mr. Gerald Strobel, provided the corporate support required for the production of this paper; Patrica Kammer, who prepared the graphics for this paper; and Nancy Schiller, who edited the original report which was the basis of this paper. The uthora s also acknowledge the valuable comments provided by the anonymous reviewers. Of course, authors alone take full responsibility for the content of this paper.

This article is available in Northeast Historical Archaeology: http://orb.binghamton.edu/neha/vol13/iss1/6 53

LOG ROADS TO LIGHT aration for the first wave of Euro-American settlement in . RAILS: THE EVOLUTION In 1801, , agent of the Hol­ OF MAIN STREET AND land Land Company, comm.issioned the marking of a new on drier land along TRANSPORTATION IN the same line as the "Great Central Trail." BUFFALO, NEW YORK This became known as "Buffalo Road" (Main Street). Soon afterward, this trail was wid­ Michael A. Cinquino, ened to 40' by cutting all trees 12" or less in Marvin G. Keller, diameter to ground level. Although larger Carmine. A. Tronolone, trees were also cut down, they were cut to Charles E. Vandrei, Jr. ground level only if they were within 8' of the centerline. Therefore, only a part of this INTRODUCTION 40' section was actually used as a road, since In June 1980, during initial construction the remaining part was often obstructed by of Buffalo's Light Rail Rapid Transit Sys­ tree stumps and undergrowth. No attempt tem, parallel logs were discovered 1.5 was made to grade the . Com­ meters below the present street surface in pleted in 1803, Buffalo Road was an un­ downtown Buffalo, New York. Preliminary graded, stump-covered, wagon route extend­ investigations suggested that they were part ing from what is now Batavia, New York, to of a log or used sometime during the Village of New Amsterdam (now Buf­ the first part of the 19th century, and as falo) (Bingham 1937; Evans 1924:232; Cum­ such. represented a significant historic re­ mings 1907:6-11, Horton, et al. 1947:22-23; source (Vandrei and Nagel 1980). Subse­ Koszuta 1969:43-44, Chazanoff 1970). quently, the site was declared eligible for Although recognizing the importance of inclusion in the National Register. opening a roadway to encourage settlement In November 1980, when destruction of of western New York State, the Holland the log road was imminent, data recovery Land Company invested only a minimal investigations were performed by the au­ amount of money and effort in road con­ thors. Information was gathered on the age, struction, and considerably less in road construction, and function of the log road by maintenance. By 1810, Main Street and Buf­ examining both the physical remains and falo Road were in disrepair due to inade­ the historic record. In addition, available quate maintenance (Koszuta 1969:43-44). information afforded an opportunity to The construction of log causeways or cor­ study not only the log road but also the duroy roads was one inexpensive method general development of a particular road­ used to keep poorly-drained, swampy, or way, Main Street, over the span of nearly muddy road segments passable. Sidway 200 years. (1906:315) offers this account of the condi­ tion of Main Street during the War of 1812:

MAIN STREET: THE DOCUMENTARY RECORD The road then was in the usual condition of The first road located along the approxi­ most if not all roads, through a new and mate route of what is now Main Street was timbered country, there being very little done the "Great Central Trail." This trail crossed except to cut the timber out sufficiently for the New York State from the Hudson River to passage of teams and the making of causeways over the low and very wet grounds. The road ; it was a major aboriginal trail from the Three-mile Creek was at first called and remained in use during the colonial Four Miles, and was throughout the -entire period. Initially, the trail was used to bring distance from the creek to the "Cold Springs" supplies to the 's covered with a log . The road, from field crews who were surveying lots in prep- what was afterwards Walden's Hill to Chip- 54

pewa Street and to North Church (Dr. those constructed in the usual Way; because Johnson's lot) was covered with a log cause­ the Logs being surrounded with Earth will way. Between Chippewa Street and where the prevent the Air from having Access to them, North Church now stands was a log over and consequently will indure a great Number a ravine and low grounds, which extended of Years(Bingham 1937:337 -338). nearly if not the entire distance. Passage on a bed was satisfac­ From this eye witness account it can be con­ tory, although after a short time many logs cluded that log causeways and corduroy pave­ would sink into the soft soil, and constant ment were used extensively in early Buffalo. use would break apart the remaining logs, In the early 1800s, and throughout most of making passage difficult (Beers 1880:7; Dur­ that century, money for road construction renberger 1931:29; Koszuta 1969: 39-40, was scarce. The corduroy technique of road 45-46). repair was cheaper than building or An attempt was made in 1826 to improve constructing a more permanent road sur­ travel conditions on Main Street by paving face. Lumber was inexpensive and an abun­ it with gravel (Reeves 1957). Gravel paving dant local resource. Corduroy roads were was more expensive and entailed more labor made of any available trees including oak, than previous forms of road construction. cedar, and hemlock (Beers 1880:7; Koszuta Gravel or pebble size usually ranged from 1969:40,45-46). 3/4" to about 2". Soil was used to properly Causeway improvements undertaken on unite and bind the gravel surface since wa­ the Buffalo-Batavia Road c.1807 are dis­ ter worn pebbles were too clean and would cussed by Joseph Ellicott in the Holland not adequately bind the pavement (Gillespie Land Company's Papers: 1871:189, 193-194). In the second quarter of the 19th century, A usual mode of constructing Causeways is to the technique, an improvement on lay two Rows of timber parallel to each other, commonly called String Piece from nine to gravel roads, became popular in eastern twelve feet asunder, the whole Extent to be North America. This road building tech­ causewayed. This Method saves the Trouble nique was developed in 1815 in England by and Expense of leveling the ground on which John L. McAdam (Encyclopedia Brittancia the Logs would otherwise lie, and is much the 1974b:426, Levy 1980). Two types of mac­ most expenditious Method of making Cause­ adam pavement were used: waterbound ways; but in consequence of the underside of crushed rock and bituminous (tar and the logs being kept from the Earth by the ) macadam pavement (Byrne String-Pieces which support them, the Mois­ 1892:200-201). The Buffalo and Williamsv­ ture of the Ground keeping them always ille Macadam Road company was chartered swamp, and being exposed to the Air, in a few to build a macadam road from Buffalo to Years the Logs become rotten, and require further Expenses in repairing . . . Williamsville. This road was completed be­ tween 1839 and 1840 (Johnson 1876:412; The Manner I had the Causeways constructed Burr 1840). Waterbound crushed rock mac­ on this Road was as follows. First the several adam pavement was constructed by repeat­ Parts of the Road intended to be causewayed edly rolling layers of wet crushed stone into were made level by digging down the Hillocks a compacted base. Bituminous macadam and filling up the Cavities, and where Stumps pavement was applied in a similar manner, came in the Way they were dug up and re­ except that coal tar or were used as moved, or cut close to the Ground. This being done and bedded in the Earth close to each a binder for stone of slightly inferior quality. other, and required to be so nearly of a size as Another paving technique used in the to make a tolerable even Surface. They were 1840s and 1850s was the plank road. In then covered with Earth from 4 to 8 inches in some cases, plank roads were constructed Depth. Causeways constructed on this Plan over old road beds, such as macadam or must necessarily be more permanent than corduroy roads. Most often, plank roads 55 were constructed as branches connecting service in 1899. However, operating costs other systems of transportation (e.g., canals were too high to make this type of system or railroads) and, as a result, were not in profitable. Consequently, storage battery­ direct competition with these systems powered streetcars and buses remained in (Owen 1850:25; Durrenberger 1931: operation for a only short time (Dibble 144-146). 1980a:H-8). Construction of P.lank roads began with In 1890, the electric trolley streetcar be­ the preparation of the road surface. After gan operations. By 1894, the last horse­ grading, two stringers (base planks) were drawn streetcars were replaced by the elec­ laid parallel to the road bed. Wooden planks tric trolley, and the public transportation were then laid over the stringers (Owen system in Buffalo was completely driven by 1850:30, 66-67). The most durable and prof­ electric power. One of the premiere systems itable wood pavements were made of planks of its time, by 1892 there were 140 miles of with as few irregularities as possible (Ged­ electric trolley track in Buffalo. This system des 1850:10; Owen 1850:10, 67). Planks had increased to approximately 202 miles of were usually laid down in tracks 8' wide; two track by 1922 (Dibble 1980a:H-8). tracks were often constructed together These forms of public transportation re­ (Owen 1850:28-29). Normally, plank road quired additional improvements to the road surfaces lasted only about five years, al­ surfaces. , block stone, and though, in some cases, the road could be pavements were utilized in Buffalo during used for eight years before repairs were the last quarter of the 19th century. Cobble­ necessary (Gedpes 1850:12). stone pavements usually were embedded in Along with t~e development of better road a rather unstable sand base. As a result, surfaces came improvements .in public cobblestone pavements were difficult to transportation systems. By 1829, Main travel over and to maintain (Byrne 1917: Street was designated a stage coach road. 125). Block stone pavement is made of stone Shortly thereafter, the public transportation blocks of regular shape which remain in system began a major expansion. In 1835, a place better than cobblestone. Block stone horse-drawn omnibus was in operation, its pavement was laid on Main Street in con­ service extending north from the waterfront junction with the trolley car railroad and the for a distance of about two miles. This sys­ electric trolley tracks of the 1890s (Byrne tem was discontinued in 1837 due to finan­ 1917:151-152). cial problems. By 1843, an intra-city stage­ was used in some areas of the coach line was in operation. Between 184 7 country as a smooth paved surface, but it and 1860, another omnibus was in opera­ was asphalt that became the most popular tion, increasing the service from the water­ pavement in Buffalo. Asphalt was first used front to three miles (Dibble 1980:H-8). as a road pavement in 1838 in Paris; it was The first iron streetcar lines, those of the not used in the United States until about Niagara Streetcar Company, were in opera­ 1869 (Byrne f892: 114). By 1890, nearly 83 tion in 1860. In that same year, Buffalo miles of Buffalo's were paved with Street Railroad began horsecar service on asphalt (Byrne 1892:114). However, signifi­ Main Street. By 1884, the system included a cant portions of Main Street remained block total of approximately 40 miles of iron track stone until the electric trolley service ended and 120 cars drawn by 730 horses, and in the mid-20th century. employed approximately 350 men (Smith The gasoline driven bus, introduced in 1884b: 529-530; Dibble 1980a:H-8). 1923, eventually became the main form of In 1888, Buffalo's first electric streetcar public transportation in Buffalo. By 1951, system, powered by a system of storage all trolley service had ended, and most batteries, began operation. The first bus, tracks and block stone pavement were com­ also powered by storage batteries, began pletely covered with asphalt to offer a 56

Since the location of the road had been established by preliminary investigations, the one meter square units that lay directly over the expected area of the log road were chosen for hand excavation. Ten units were hand excavated to the surface of the logs. Seven additional units were excavated in an attempt to locate any other features. Exca­ vation using trowels and brushes allowed for recovery of small artifacts and, at the same time, permitted careful and efficient expo­ Figure 1. Placement of Excavation Units at the sure of the log surface. All squares were Buffalo Log Road Site (UB 1682). then plan drawn and photographed, and the smooth surface for bus and automobile traf­ locations of the recovered artifacts were fic. mapped in situ. Artifacts recovered from the The construction of a 6.4 mile Light Rail excavation units were recorded within one of Rapid Transit Line that runs for a distance three general provenience categories: Level of 1.2 miles at ground surface and travels A: fill deposits 15 to 20 em. above the log below ground for a distance of 5.2 miles surface; Level B: directly on or in between began a new chapter of public transporta­ the logs; or Level C: below the logs. tion in Buffalo. The system became opera­ Two one meter squares (units IN 3E and tional in May 1985. ON 3E) were excavated to a sufficient depth below the logs to determine if any additional surfaces existed, or if any base preparation MAIN STREET: THE ARCHAEOLOGICAL had taken place prior to log placement (Fig­ REMAINS ures 2 and 3). The logs located within these Data recovery operations began as an "L"­ squares were cut in half and removed, sam­ shaped excavation unit which covered an ples were retained for identification of the area of about 38 m2 and was located approx­ wood. Profiles and cross sections were drawn imately in the middle of Main Street (Figure of the overall excavation, as well as of indi­ 1). Size and placement of the excavation vidual excavation units. Soil augering was were governed by the following objectives: to conducted throughout the excavation unit to uncover as complete a section of the log road verify the depth of deposits and to check for as possible; to open an area adjacent to the additional surfaces. log road surface in order to identify any additional features; and to place the excava­ Log Road Description tion unit in such a way as to allow free flow of two of at all times. All logs uncovered were in good condition. A backhoe was used to break through Fill material deposited over the top of the existing pavement and to excavate fill de­ logs, and the silty clay deposit in which they posits down to 1.35 meters below the present were embedded, served to keep them moist ground surface, or about 15 em. above the and well preserved. A chain saw was used to level of the log road. After completion of obtain a cross section of the road. The larg­ backhoe excavation, the side walls and floor est logs, which ranged from 29 to 40 em. in of the unit were straightened using shovels. diameter, were below an organic layer of A one meter square grid pattern was then rotted wood and earth measuring 2 to 5 em. established within the excavation. All mea­ in thickness (Figures 2 and 3). The cross surements were recorded in relation to this section of the logs indicated this organic grid pattern. The grid was tied into a nearby layer was not only present on the upper road USGS benchmark. surface, but also extended between the logs 57

Most of the logs showed no evidence of having been intentionally hewn to make a smoother road surface. A single exception was a hewn plank uncovered in excavation unit 2S 2E (Figure 2). Since this plank was the only one of its kind recovered, it is believed that its use as part of the log road was secondary. All logs were placed perpen­ dicular to the flow of traffic, and were roughly parallel to each other. There was no evidence that the logs were tied together; they were simply held in place by their own Figure 2. Plan Drawing of the Log Road. weight. The logs varied in size and diameter. Those excavated in preliminary investiga­ tions were fairly consistent in size, ranging in diameter from 15 to 30 em. (Vandrei and Nagel1980:9). However, the diameter of the logs and branches excavated during data recovery ranged from 2 to 40 em. The diameter of the logs generally decreased · from north to south across the overall 7 •• excavation unit. This size differential may --·""-n'~'""" D~.oaa1""'w best be explained if the portion of the road 0 D

CA~I I~CfNTIMfTI ... 0 • ' J •• Metal o £ Hi U n ICAliiNINO::.. U Metal artifacts comprise the largest cate­ gory of indentifiable materials, and provide the most reliable chronological information since they were the artifact type most fre­ quently found in association with the log pavement. Nine metal artifacts were recov­ ered; all but two appear to be iron. Six of these items are nails or spikes, two of which were recovered from the backhoe spoil during early excavation. These two spikes are mod­ ern steel railroad spikes, probably associated Figure 4. Handwrought Nails Found in Association with the electric railway which operated with Log Road along the center of Main Street. Of the re­ maining four nails, three were found in direct eastern side had been cut by another modern association with the log pavement. The one trench (Figure 3). As a result, only 2.1 nail not associated with the road appears to meters of the log feature remained undis­ be a farrier's shoeing nail, possibly hand­ turbed. Test units placed west of the modern wrought. All of the remaining nails are trench failed to locate any additional evi­ handwrought iron types of various sizes and dence of the road. No excavation was possi­ configurations (Figure 4). One is a large ble east of this line. wrought spike. With the onset of cut nail The length of the road or causeway also production, the manufacture of such wrought could not be precisely determined by the nails began to decline in the early 19th cen­ excavations. However, as a result of moni­ tury; thus the nails may be dated generally toring of construction and previous investi­ to this period of time (Noel-Hume gations in the area, it is known that the logs 1969:252-254). extend north-south along Main Street ap­ The remaining metal artifacts consist of a proximately 60 meters. Monitoring both drilled or punched metal plate, a fragment of north and south of the data recovery exca­ a horseshoe, and an unidentifiable iron frag­ vation unit turned up additional evidence of ment. The horseshoe fragment, recovered in possible road surfaces in the form of a dark backhoe spoil directly above the log layer, is organic stain or isolated logs. However, no about one-half of an iron shoe with one nail. evidence of a continuous road surface was The horseshoe is of a type in use c.1800 encountered. (Noel-Hume 1969:237-239; Figure 74 #6). ARTIFACT DESCRIPTION AND ANALYSIS Data recovery excavations on the log road Brass recovered a total of 49 artifacts. These were A single brass artifact was recovered. This recovered from three areas: the backhoe artifact, a civilian coat button, was dated spoil resulting from fill removal above the from about 1790 to the early 1800s and road, the area above the road surface exca­ suggests the log road was constructed and/or vated by hand (Level A), and the area in used around this time period (Vandrei and direct association with the log pavement Nagel1981:15). (Level B). No artifacts were recovered from beneath the logs (Level C). Artifactual ma­ Leather terial recovered included metal, leather, ce­ ramic, and glass. Nonartifactual materials Most of the leather was found in the hand recovered included wood, shell, stone, and excavated layer (level A) above the log pave­ soil. ment. Occasional fragments also were found 59

small fragment of smoky clear bottle glass found in direct association with the logs. The small size of each of these items makes further identification and chronologiCal placement impossible.

STRATIGRAPHY Figure 5 illustrates the profile ofthe north wall of the excavation from the present surface of Main Street to beneath the log road. Descriptions of the eleven strata iden­ Figure 5. Profile of North Wall of the Excavation. tified in Figure 5 are included in Table 1. See Table I for stratigraphic descriptions The profile reveals the successive layers of in the backhoe spoil before hand excavation. both earlier road surfaces and fill deposits. Of the leather fragments which are identifi­ Analysis of sediments taken from these lay­ able, seven are probable strap fragments ers show well-sorted, beachlike sands above and three may be shoe parts. The remaining the logs, and glacial till below. The mineral unidentifiable pieces range in size from 1 content of these sediments is typical of sand­ 2 2 cm to 80cm . The thicknesses of the pieces sized sediments in western New York. It is also vary considerably. All leather pieces believed that the log road was constructed had one smooth finished side and one rough on naturally deposited glacial sediment and flesh side. The fragmented nature, wide dis­ later overlain with successive layers of sand tribution, and lack of association with the obtained locally, perhaps from Lake Erie road suggest that the leather was deposited beach deposits (Cazeau 1981:3). with later fill placed on top of the road The stratigraphy illustrates a general his­ surface. torical sequence of road const~uction on Main Street over the last 180 years. At least Ceramic nine distinct strata were recognized in the ,. north wall profile. These strata, identified Five ceramic fragments were recovered. by Roman numerals in Figure 5, represent Two are brick fragments. Two are untyped separate chronologically distinct deposits. pieces of unglazed, smooth, white paste Separate components recorded within two of . earthenware which could not be further the strata (Strata II and VIII) are identified identified. The earthenware fragments may by lower case letters. be a form of refined stucco and, therefore, Stratum I. Stratum I consists of asphalt not vessel sherds. None of these items was pavement, which is the present surface of· found in direct association with the log pave­ Main Street. This pavement was placed as ment. late as the 1950s. The stratigraphic profile The fifth ceramic sherd, a small piece of shows that the asphalt, which is 5 to 10 em. wheel-made, refined redware glazed on both thick, was laid directly over the block stone interior and exterior surfaces, was found in and rails of the previous road surface. direct association with the log road. Because Stratum II. This stratum represents the of the small size of this sherd, further iden­ road surface and remnants of the trolley car tification was not possible. system which operated on Main Street from about 1890 to 1950. Five separate compo­ Glass nents of this stratum are identified in the During excavation, two glass artifacts profile. These components illustrate how the were recovered: a clear window glass frag­ road surface of Stratum II was constructed. ment found in level A above the road and a A 20-cm. layer of stone ballast (Stratum 60

TABLE I STRATIGRAPHIC DESCRIPTION

Strata Description of Material Interpretation

Asphalt pavement Present road surface lia Stone and mortar Block stone pavement lib Quarry crush Sand and gravel deposit placed below block stone lie Concrete Concrete pavement, associated with trolley system lid Railroad tie and rails Trolley tracks lie Quarry crush Stone ballast to support trolley tracks III SAND: gray/tan, bulk/fine sand; pebbles, rotted FILL: fairly well-sorted beach sand wood, and glass. IV SAND: yellow/tan, uniform, well-sorted; peb­ FILL: fairly well-sorted beach sand bles. v SAND: dark gray/brown, dries to pale FILL: fairly well-sorted beach sand gray/brown; organic fragments common; peb­ bles. VI SAND: medium gray/brown, dries to pale FILL: fairly well-sorted beach sand gray/brown; organics and some grit present; predominantly fine sand. VII CLAY SAND: unlike other sands, this may FILL: not well-sorted sand of glacial origin have a glacio-lacustrine origin; pale brown when dry; contains some shell fragments and pebbles, but is predominantly reworked clay and very fine sand/silt; sieving yielded a sand distribution, but further examination indi­ cates that this contains a large proportion of aggregated lumps (sand/silt bonded by clay); overall behavior is that of a coarse pebbly sand, but composition is dominated by the aggregate clasts, not the primary components; hence, a poorly-sorted clay sand reworked by latter events seems to be indi­ cated VIlla ORGANICS: Predominantly rotted wood debris. FILL: earth mantel applied above and between logs VIIIb Log Log road pavement IX, X SILTY SAND: pale brown; coarse sizes absent; NATURAL DEPOSIT: not well sorted sand of exhibits little evidence of disturbance in its glacial origin behavior of composition. Source: Cazeau 1981.

Ile) was first laid over the ground surface. these ties. Between the rails, mortared block As seen in the profile, a portion of the stone (IIa) was laid over a small bed of ground may have been excavated in order to crushed stone (IIb). Early photographs of provide a level bed for the route. The eastern Main Street taken during the late 19th end of Stratum III extends above the ballast century show block stones as part of the on the east side of the trench wall, suggest­ trolley system pavement (Dibble 1980a: ing that a portion of this deposit was re­ H-8). moved when the rails were laid. Concrete (Stratum Ilc) also was identified Wooden ties (IId) were set into the stone as part of the trolley system pavement. Con­ ballast, and the rails were placed on top of crete was widely used in the latter part of 61 the 19th century. Only the eastern edge of during the leveling of the roadway for the the profile contains concrete. Cobbles were trolley system. Since a portion of Stratum III recorded only between the rails. The use of has been disturbed, it is likely that any road concrete is restricted to the area outside of surface associated with this level also were the trolley tracks, and is contemporaneous removed. with the block stone pavement. Disturbances from utility trench excava­ Strata III through VII. These five strata tions are illustrated in Figure 5. The most represent successive layers of sand fill which recent trench is shown on the east side of the were deposited over the log road. Each stra­ wall profile. This trench, which was exca­ tum is between 10 and 20 em thick and vated during the summer of 1980, cuts represents a distinct fill deposit. The depos­ through all of the stratigraphic levels. A its were placed over a 70 year period from portion of the log road was removed during when the log road was first in use excavation of this trench. The trench on the (1807-1820) to when the trolley system be­ west side is much older and is the result of gan operation (1890). It could not be deter­ excavation of a waterline which runs down mined from either the profile or from docu­ the center of Main Street. An examination of mentary evidence exactly when the fill the profile reveals that this waterline was materials were deposited, or whether these placed in the ground prior to the construc­ deposits were laid consecutively over a pe­ tion of the trolley line and the block stone riod of weeks or years. The only diagonostic road surface in the late 19th century. The artifact recovered from the fill deposits was trench cuts through Strata V through IX, a horseshoe fragment, dating to the early but is overlaid by the undisturbed ballast part of the 19th century. Since the horseshoe (Stratum IV) of the trolley system. was uncovered during the initial backhoe excavation, exact placement of the artifact could not be determined. SUMMARY Stratum VIII. Stratum VIII represents the The field investigation along with the original log road surface. It is divided into historical research enabled questions to be two components: a 5- to 12-cm. thick organic answered regarding the age, function, and deposit of wood fragments and bark (Stra­ construction of the log road. Although the tum VIlla) and the logs themselves (Stra­ number of artifacts recovered was low, the tum VIIIb). dates for those that were recovered correlate Stratum IX. The deepest level excavated, well with the documentary evidence. A hand­ Stratum IX is the original ground surface. It wrought iron horseshoe was recovered from is a deposit of unsorted, silty sand, and lies the fill layers overlying the log road. Al­ at a depth of 1.8 m. below the present street though no specific provenience is available surface. for this artifact, the manufacture and use of Only three distinct road surfaces were this type of,horseshoe dates to c.1800 (Noel­ identified. Although other pavements, such Hume 1969:237-239; Figure 74 #6). Three as gravel, macadam, and plank, have been handwrought nails, most likely dating from recorded elsewhere on Main Street, none of no later than the early 19th century, also these surfaces appear in the excavation. were recovered from the excavation units However, during monitoring of construction directly over the log road surface. Also, a along other areas of Main Street, portions of brass coat button dated from the late 18th what may be a macadam pavement were century was recovered (Vandrei and Nagel recorded at a depth of 80 em. in one of the fill 1981:15). layers above the log road. No macadam Available documentary evidence supports pavement was found during data recovery. the dates obtained for the button and the It is possible that any previously existing wrought iron artifacts. It is unlikely that the macadam and plank roads were removed logs were in place prior to 1798. Before that 62 date, Main Street was an Indian trail one case where an odd "V" -shaped tree (Koszuta 1969:27-30). In 1798, the trail was trunk was used, another log was closely fit widened to accommodate wagon traffic. Log next to it (Unit ON 2E; Figure 2). Smaller causeways were employed shortly after branches were placed between larger logs to the turn of the 19th century (Bingham make as smooth a surface as possible before 1937:336). Documentary evidence places an organic layer was added. Excavation construction between 1800 and 1810. showed that a 2 to Scm. thick deposit of Evidence supports the final date for use of organic material was then placed over the the log road as 1839 or 1840, when a mac­ log surface. The nearest available trees were adam road was constructed on Main Street used to construct the causeways. Prelimi­ (Burr 1840; Johnson 1876:412). Although no nary identification of log samples indicates evidence of the macadam road was uncov­ that cedars and other hardwoods indigenous ered during data recovery excavations, mon­ to poorly-drained areas were used. A com­ itoring of construction activities located plete stratigraphic sequence of road building what may be a macadam surface at a depth was not uncovered during the field investi­ of approximately 80 to 100 em. The log and gations. The historic research indicates that probable macadam surfaces were separated gravel, macadam, and plank surfaces were by 5 to 20 em. of earth fill. The early date of used also on parts of Main Street. Although 1798 and the later date of 1840 correlate portions of what may have been a macadam well with the probable use-span of the but­ surface have been observed while monitor­ ton and wrought iron artifacts. ing construction activities in other areas of The purpose of the logs was to fill in a low Main Street, no evidence of a plank or gravel wet area to prevent vehicles becoming mired surface was noted. However, despite the in mud. Excavation uncovered logs either nature of urban development, historic re­ embedded in, or resting on top of, fine grain sources can still be found in an undisturbed deposits of silty clay sand and sandy clay, setting. The stratigraphic profile serves to which suggest a wet area. Although addi­ graphically illustrate extensive fill se­ tional logs were periodically noted during quences that can occur in many urban areas. subway excavations, there was no evidence Historic fill not only serves to conceal his­ of a continuous log surface. In every case, toric resources but to protect and preserve the logs were found embedded in a similar them. silty layer and overlain by various layers of fill. The archaeological evidence, as inter­ preted, correlates with the documentary ev­ ACKNOWLEDGMENTS idence. In most cases, the logs were used as This article is a condensed and edited causeways and were not part of the contin­ version of a report submitted to the Niagara uous pavement. Construction methods for Frontier Transportation Authority (NFTA) the log road were not complicated. Since in June 1981 (Ecology and Environment, trees were plentiful, they were used as the Inc. 1981). Research on the Buffalo Log Road most expedient material to cover poorly was conducted with public funds provided drained areas. Log causeways were often through PL 93-291. The NFTA and the en­ used while leveling hillock areas and filling gineering firm of Cordry, Carpenter, Dietz in low areas of the road. Soil was then placed and Zack deserve special thanks for their between and on top of the logs to level the cooperation and logistical support. Ecology surface of the logs and to provide a more and Environment, Inc., in particular Mr. permanent road (Bingham 1937:337-338). Gerald Strobel, provided the corporate sup­ Excavations at UB 1682 demonstrated port required for the production of this pa­ that logs were placed directly over the wet per; Patrica Kammer, who prepared the area, perpendicular to the traffic flow, and graphics for this paper; and Nancy Schiller, were fitted together as closely as possible. In who edited the original report which was the 63 basis of this paper. The authors also ac­ Dibble, Ralph knowledge the valuable comments provided 1980 Story of Mass Transit is Filled with Irony. by anonymous reviewers. Of course, authors The Buffalo News: Sunday, June 1, 1980, p. H-8. alone take full responsibility for the content Dibble, Ralph of this paper. 1980b Buffalo Led in Rail Transit-Nearly a Cen­ tury Ago. The Buffalo News, Sunday, June 1, BIBLIOGRAPHY 1980, p. H-8. Durrenberger, Joseph Austin Anonymous 1931 Turnpikes: A Study of the Move­ 1951 The Buffalo Evening News, March 16, p. 47. ment in the Middle Atlantic States and.Mary­ Beers, Frederick W. land. Southern Stationery and Printing Com­ 1880 Illustrated Historic Atlas ofErie County, New pany. Baldosta, . York. F.W. Beers and Company. New York. Ecology and Environment, Inc. Bingham, Robert Warwick, Editor 1981 Archaeological Investigations of the Buffalo 1937 Reports of Joseph Ellicott as Chief of Survey Log Road Site (UB 1682 Buffalo, New York: (1791-1800) and as Agent (1800-1821) of the Data Recovery Program. Report submitted to Holland Company's Purchase in Western New the Niagara Frontier Transportation Author­ York. Holland Land Company's Papers. Vol­ ity. Buffalo, New York. ume 1. The Buffalo Historical Society. Buf­ Encyclopedia Britannia falo. 1974a Roads and Highways. Volume 15, pp. Bingham, Robert Warwick, Editor 892-906. Encyclopedia Britannica, Inc. Chi­ 1941 Reports of Joseph Ellicott as Chief of Survey cago. (1797-1821) of the Holland Land Company's 1974b John Loudon McAdam. Volume 6, p. 426: Purchase" in Western New York. Holland Land Encyclopedia Britannica, Inc. Chicago. Company's Papers. Volume II. The Buffalo Evans, Paul D. Historical Society. Buffalo. 1924 The Holland Land Company. Buffalo Histor­ Bloodgood, S. Dewitt ical Society. Buffalo. 1838 Treatise on Roads, Their History, Character, Geddes, George and Utility. Oliver Stetle. Albany. 1850 Observations upon Plank Roads Together Bureau of Engineering with the General Plank Road Law of the State 1897 Index to Records of Streets, Public Grounds, of New York. L.W. Hall, Syracuse; and A.S. Waterways, Railroads, Gas Companies, Water Barnes and Company. New York. Works, etc. of the City of Buffalo from 1814 to Gillespie, William M. 1896. The Wenborne-Sumner Company, 1871 A Manual of the Principles and Practice of Printers. Buffalo. Road-Making. A.S. Barnes and Company. Burr, David H., (Ed.) New York and Chicago. , 1840 Map of Erie County. Burr's Atlas ofNew York Hammersley, Margaret State. David H. Burr and Company. New 1969 Niagara Frontier Roads that Grew from In­ York. dian . The Buffalo Evening News, Sep­ Byrne, Austin T. tember 20, 1969. 1892 A Treatise on Construction. John Horton, John T., Edward T. Williams, Wiley and Sons. New York. and Harry S. Douglas Byrne, Austin T. 1947 History of Northwestern New York, Erie, 1917 Modern Road Construction. American Tech­ Niagara, Wyoming, Genesee, and Orleans nical Society. Chicago. Counties. Volume I. J. Horton, editor. Lewis Cazeau, Charles J. Historical .Publishing Company, Inc. New 1981 Buffalo Log Road Site CUB 1682): A Geologi­ York. cal Analysis. CPF Associates. Buffalo. Johnson, Crisfield Chazanoff, William 1876 Centennial History ofErie County, New York. 1970 Joseph Ellicott and the Holland Land Com­ Printing House of Matthews and Warren. pany: The Opening of Western New York. Buffalo. Syracuse University Press. Syracuse, New Kingsford, W., F.G., Skinner, and York. O.E. Clarke Cummings, Uriah 1851 History, Structure and Statistics of Plank 1907 Indian Trails of Western New York. Akron Roads in the United Siates and Canada with Herald. Akron, New York. Remarks on Roads in General and a Letter on 64

Plank Roads. Hart, Carey, and Hart; Phila­ Vandrei, Charles E., and Brian Lee delphia. Nagel Koszuta, Joanne Barbara 1980 Archaeological and Documentary Research on 1969 An Interpretation of the Sequent Patterns of UB 1682 Buffalo Plank Road. Reports of the Trails, Roads, and Highways on the Niagara Archaeological Survey. Volume 12, Number Frontier to 1950. Masters of Arts Thesis, 15. Department of Anthropology, State Uni­ State University of New York at Buffalo. versity of New York at Buffalo. Levy, Michael White, Truman C., (Ed.) 1980 "City Fathers Saw its Roads as the Path to 1898 Our Country and its People: A Descriptive Future Growth." The Buffalo Evening News, Work on Erie County, New York. Volume I. Sunday, June 1, 1980, p. H-9. The Boston History Company, Publishers. Boston. Noel-Hume, lvor 1969 A Guide to Artifacts of Colonial America. Knopf; New York. Owen, Robert Dale 1850 A Brief Practical Treatise on the Construction Authors: and Management of Plank Roads. Kent and Norman, Publishers; New Albany. Michael A. Cinquino Reeves, Jean Carmine A. Tronolone 1957 Main St.-Road of Dreams; Progress. The Ecology and Environment, Inc. Buffalo News, March 23, 1957. P.O. Box D Sidway, Parnell Buffalo, New York 14225 1906 Recollection of the Burning of Buffalo. In Publications of the Buffalo Historic Society. Marvin G. Keller F.H. Severance, editor. Volume IX, pp. Bureau of Reclamation 311-336. P.O. Box 2553 Smith, Henry Perry Billings, Montana 59103 1884a History of Buffalo and Erie County. Volume I. D. Mason and Company, Publishers. Syracuse. Charles E. Vandrei, Jr. Smith, Henry Perry New York State Department of 1884b History of Buffalo and Erie County, New Environmental Conservation York. Volume II. D. Mason and Company, 150 Wolf Road Publishers. Syracuse. Albany, New York