Navigation and Waterpower: Adaptation and Technology on Canadian Canals

Canals in Canadian industrial history are best known for their contribution to transportation. With the exception of the Lachine Canal in Montreal, however, their role as a source of power for driving industrial machinery is less well known.[1] Canal-based waterpower was an important catalyst for industrialization in several regions of Canada following the 1844 decision of the Board of Works of the United Canadas to lease surplus canal water for power generation. The board’s decision was later extended to all potential waterpower sites created by the construction of public works. Discussion centers on the extent of waterpower generation on the canals associated with the St. Lawrence River and Niagara escarpment, their influence, and the spatial organization of the hydraulic sites resulting from the modification of navigation canals for power generation, including the construction of weirs and races around the locks. How improvements in waterwheel efficiency enabled manufacturers to get the maximum power from canal-side waterpower locations is also covered.

Les canaux, dans l’histoire industrielle canadienne, sont bien connus en tant que voie de communication maritime. Cependant, à l’exception du canal de Lachine, leur rôle en tant que source d’énergie pour alimenter la machinerie industrielle, est beaucoup moins connu. En fait, l’énergie hydraulique produite par les canaux de navigation a été un puissant catalyseur pour le développement industriel de plusieurs régions canadiennes, principalement suite à la décision, en 1844, du gouvernement du Canada-Uni de louer des prises d’eau pour la production d’énergie mécanique le long des canaux. Cette décision va ensuite être généralisée à tous sites, susceptibles de générer de l’énergie hydraulique, adjacents à des écluses ou des barrages construits par le gouvernement. La discussion qui suit porte sur l’ampleur de la production d’énergie hydraulique sur les canaux associés avec le fleuve Saint-Laurent et l’escarpement du Niagara. On regardera premièrement l’influence de cette production et l’organisation spatiale des sites hydrauliques résultant de l’adaptation des canaux de navigation à la production d’énergie, incluant la construction des prises d’eau et des coursiers autour des écluses. La discussion portera ensuite sur comment l’évolution technologique des turbines a permis aux manufacturiers de tirer le maximum d’énergie de leurs lots hydrauliques.


The first Canadian canal built for both transportation and waterpower in the early-19th century was the Welland Canal (1824–1833), which joins Lake Erie to Lake Ontario. The Lachine Canal (1821–1825), which bypassed the rapids at Lachine and crossed the island of Montreal, had been built a few years earlier but for transportation only. When, in the 1840s, the new United Canadas government improved the St. Lawrence navigation corridor by enlarging existing transportation canals and by digging new ones to bypass the numerous rapids between Montreal, Lake Ontario, and Lake Erie, it decided to lease waterpower rights along both the enlarged canals and the new ones (Beauharnois, Cornwall, and Williamsburg) (figure 1).

Figure 1. Location of the St. Lawrence and Welland canals: (a) the St. Lawrence canals, (b) the Welland Canal, (c) the old Military canals.
Maps drawn by David Edwards-May, Euromapping, France.

Placing watermills along artificial waterways had a long history in Europe and even in Canada.[2] The Sulpician Order, seigneurs of the island of Montreal, planned a dual-purpose canal for milling and transportation between Lachine and Montreal as early as the 17th century. This canal was never officially used for transportation but served to bring water to the Sulpician flour mills near Ville-Marie during the 18th century.[3] Not until 1781 did the first Canadian navigation canal became a reality at Coteau du Lac (1779–1781). This pioneer canal used three locks to overcome a 2-meter rise. The Split Rock (Rocher Fendu), Faucille, and Trou du Moulin canals followed by 1783. The Cascades Canal replaced these three in 1804 (figure 1c). British authorities built these pioneer navigation canals principally to improve the movement of military troops and to protect the border following American independence. For this reason they were known as military canals. They were located between Lake St. Francis and Lake St. Louis and were later replaced by the Beauharnois Canal.

The first significant canal for commercial navigation built on the St. Lawrence was the Lachine Canal on the island of Montreal, constructed between 1821 and 1825. The Welland Canal, linking Lake Ontario and Lake Erie, constructed between 1824 and 1833, closely followed.[4] Initiated by private interests, the Lachine Canal was completed by the government of Lower Canada and designed solely for transit navigation. Other early government-constructed canals on the St. Lawrence were also built for navigation only. In contrast, a private company (the Welland Canal Company) built the first Welland Canal for the dual purposes of waterpower and transportation.

In 1844 the chairman of the Board of Works was authorized to sell or lease power on the St. Lawrence canals. Following the 1867 British North America Act, establishing the Dominion of Canada, the federal government acquired the right to lease power and to sell or rent lots. By this date, however, factories were already well established on the St. Lawrence canals.[5]

Canada’s decision to lease surplus canal water in the mid-1840s was closely tied to improvements in the St. Lawrence navigation system designed to open navigation routes to the Great Lakes for larger boats (schooners, sloops) and steamboats. As part of these improvements, the Board of Works fixed the dimensions of St. Lawrence navigation locks at 200 × 45 feet with a depth of 9 feet (compared to less than 2.5 feet previously). The new standards led to the construction of new canals and to the enlargement of existing ones. The Lachine Canal (1843–1848) doubled its width and depth. The Beauharnois Canal, an entirely new canal on the south shore of the St. Lawrence River, replaced the old military canals (Cascades, Split Rock, and Coteau). The Williamsburg Canals were built. The Cornwall Canal, whose construction had been initiated by Upper Canada’s administration in 1834, was finally completed. The Welland Canal was repaired, reconstructed, widened, and deepened. These modifications improved navigation, but they also increased the flow of water in the canals and thus their waterpower potential. Government acquisition of the Welland Canal in the 1840s also influenced matters. The Welland was already a dual navigation and power canal, and Canada’s government decided to extend the waterpower policy existing on this canal to the other canals and eventually to all installations where waterpower was created by the construction of any public work.

Figure 2. Map (by Chaussegros de Lery 1733) of the Sulpician canals showing the two sections of canal known as the Sulpicians or Lachine Canal and the St. Gabriel Canal. These canals fed the two water-powered communal gristmills.
Courtesy of Centre des archives d’outre-mer (France), 03DFC 483B.

Two leasing systems had developed along the Welland Canal. Under one system, a private company would lease an entire section at a price based on the quantity of water used. The company then paid for the installation of weirs and races but could sublease available waterpower to individual users. Under the second system, the canal company erected the weirs and races and leased lots directly to individual users who paid the canal authority on the basis of the amount of water consumed. As these leasing systems were expanded to other canals, the government, of course, insisted that the use of waterpower not be detrimental, in any regard, to navigation.[6]

In reality, each location used different rates and conditions, and the application of controls varied widely. The situation became so chaotic that in 1882 the Department of Railways and Canals published a series of reports on the use of waterpower on the St. Lawrence and Welland canals.[7] The purpose of these reports, authored by Robert Douglas, was to reform the regulations and standardize lease conditions. Douglas found both overuse of waterpower and discrepancies in rental prices. This is the only document providing detailed information about those who leased water for power from navigation canals, the conditions of their leases, and information about the hydraulic machinery in operation. While the conclusions of his reports are not discussed, this paper uses Douglas’s data to investigate the development of waterpower on Canadian navigation canals.

Role of Waterpower in Canada

Water has traditionally been an important source of power in Canada. Today, hydroelectricity still produces 60 percent of Canadian electricity.[8] Water, likewise, was the primary source of power before 1900. Felicity Leung notes,
Waterpower was the principal source of industrial power at the majority of mills in British North America until the last decade of the 19th century, and possibly into the first decade of the 20th century. Although the introduction of steam for heating, drying or processing in factories and plants might encourage the use of steam for power, evidence indicates that, when available, waterpower was used for driving machinery. These were regional variations.[9]

Table 1, for example, shows the location and the type of power indicated by manufacturers in the 1871 census.[10] Only 30 percent of the manufacturers declared any source of power, and, if one did not know otherwise, the returns could suggest that 70 percent of the manufacturers did not use any form of mechanical power.[11] Nonetheless, from this fragmentary data, it is clear that waterpower was the major source of power for almost all the mills in the Maritimes and Quebec and for almost half in Ontario.

Table 1. Establishments by Location and Type of Power 1871
Province Water Power Steam Both Horse Total
No. % No. % No. % No. % No.
Ontario 2,520
2,600 28.36
11.97 302
2,132 3.40
9.81 12
123 0.14
0.57 122
713 1.37
3.28 8,887
Lachine Canal*
Welland Sub-district
Welland canal**
Cornwall Sub-district
Cornwall Canal*
Willamsburgh Sub-districts
Willamsburgh Canals*
Beauharnois District
Beauharnois Canal*






























*in 1882
**in 1882, without feeder Data from CANIND71 (see n. 10).

This article looks specifically at the hydraulic complexes located along the navigation canals of the St. Lawrence (Lachine, Beauharnois, Cornwall, Williamsburg) and the Welland Canal. The canals are situated in two provinces, Quebec and Ontario, where industrial development and regulations differ slightly. Some of the most important industrial complexes of the 19th century were along those canals. There were also water privileges on some other navigation routes (the Ottawa River, the Rideau Canal, the Chambly Canal, and the Trent River) where manufacturing developed, but these will not be dealt with here.

Very few publications document the use of waterpower to drive machinery in industries established along navigation canals. In terms of industrial objectives, a canal-side location was very valuable. Such locations provided a regular source of water as well as ready access to transportation. The most important New England waterpower industrial centers (Lowell, Holyoke, Lawrence, Manchester) were located on navigation routes improved with canals and were developed at about the same time. Were the waterpower sites on the St. Lawrence and the Welland canals as attractive for industries?

The second part of Table 1 indicates that waterpower sites along navigation canals in Canada were, indeed, attractive to industry. Most of the watermills of the Niagara region were concentrated along the Welland Canal. Most of those around Montreal were concentrated on the Lachine Canal. The regional dominance of canal-side mills was not as great along the other canals. Many factors may explain these differences, such as the role of the canal in the economy of the region, the economic importance of the region, the region’s type of industries, and the sizes of its industries. Nonetheless, the data demonstrate that navigation canals, at least in the case of Welland and Lachine, were preferred locations for watermills.

Location and Spatial Organization of Mills along Canals

As Canadian navigation canals were modified to provide a source of waterpower, additional modifications were necessary. Engineers had to redesign the canal infrastructures and construct the necessary hydraulic works (sluices, weirs, raceways, etc.) to allow for both transportation and waterpower (Table 2). For instance, rapid flow along a canal route can be advantageous for waterpower but disadvantageous for canal navigation. Engineers constantly had to make modifications to cope with the conflicting demands of dual-purpose canals. This would be a recurring problem on the Lachine Canal.[12] Power users almost always wanted more water. An 1887 report of the Canadian government on the use of waterpower on the Lachine Canal and a 1910 inventory made by G. Viger show that manufacturers, for example at Basin 2, improved their waterpower capacity by replacing their waterwheels, since they were limited in the amount of water they could drain from the canal without interfering with navigation.[13]

Table 2. Dates of Construction and Hydraulic Privileges on the St. Lawrence and Welland Canals
Canal Built No. of
locks Length
(mi) Rise
(ft) Date of
Privileges No. of Mills
in 1882
Total Mills
7 (5)
39 (26)
27 (27)
*We take in account just the first and second canal that served for waterpower.

Lachine Canal

Even though it opened to navigation in 1825, the Lachine Canal did not provide waterpower until 1846, though plans to lease hydraulic privileges had been initiated in 1844. Following improvements made in the early 1840s, the Lachine Canal had five locks rising 45 feet over a distance of 14 miles. Water privileges were concentrated around locks 2 (Basin 2, Montreal), 3 (St. Gabriel), and 4 (Côte St. Paul), but the waterpower installations and the types of ownership differed.[14]

The first water rights were auctioned off at Basin 2, at the head of lock 2, between 1846 and 1851. The canal at this location (figure 3a) ran parallel to the St. Lawrence River, creating a head of 22 feet during summer.[15] The hydraulic lots were located in a row on the artificial island created by the canal. A sluice in the canal wall at each lot allowed water to enter the factories, drop through their power-producing wheels or turbines, and exit back to the river below.[16] The government leased the land and the hydraulic rights on an individual basis for 99 years. Engineers erected weirs to regulate the flow of water and constructed the head gates or sluices of each race in the wall of the canal.

Figure 3. Plans of weirs, sluices, and races at Basin 2, St. Gabriel, and Côte St. Paul hydraulic sites along the Lachine Canal, Montreal. Also shown are photographs of the sluices at Basin 2 and St. Gabriel, in 1950 and 1920, respectively. An insurance drawing (Charles E. Goad) shows the John Frothingham Estate in 1880 near the Côte St. Paul locks.

The second waterpower site on the Lachine Canal was located one mile west of Basin 2, at lock 3, also named St. Gabriel. The head at this location was 8 feet, far less than at Basin 2, but had the advantage that it was not affected by flooding. This mill site was auctioned off in 1851 to John Young and Ira Gould, who later formed the St. Gabriel Hydraulic Company.[17] Unlike at Basin 2, a single group leased all water rights on lock no. 3. The lease was still in force in 1999. At this site, the engineer Samuel Keefer originally planned seven hydraulic lots. However, by 1856 (figure 3b) 17 hydraulic lots had been located on both sides of the canal; by the 1860s there were 20 lots. The spatial organization of this site was more complex than at Basin 2. The canal was realigned, and the curve of the old 1825 canal, on the north side, became the headrace. The sluices opened directly onto the canal. A tailrace was built behind the headrace, and a weir replaced the old lock. A race was also built on the island formed by the old and new canals. On the south side, some lots received water directly from the canal while others received water from a headrace. The water exited in the Priest’s Basin through a tailrace running parallel to the canal.

The third waterpower location was established at lock 4 in Côte St. Paul, about 2 miles from St. Gabriel, in a rural area. The head was 9 feet. Keefer originally planned five hydraulic lots. In 1853, the site was leased to William Parkyn, who took responsibility for constructing a 2,000-foot-long headrace parallel to the canal, on the south side. The hydraulic lots (figure 3c) were in a row, and the water exited into the canal. Parkyn was also responsible for erecting the buildings, installing equipment, and providing housing for the workmen. Parkyn’s interests were gradually sold to John Frottingham and William Workman, who later formed the St. Paul Land and Hydraulic Company.

The same canal had three different types of waterpower leases. At Basin 2, leases were granted to a large number of individuals for 99 years, and the government built the intake and races while the lessees constructed and equipped the buildings. At St. Gabriel, all hydraulic privileges were leased to a single company, which built all the hydraulic works and managed the site. At Côte St-Paul, an individual, later a company, leased all the hydraulic privileges, including the land south of the lock, with the authorization to develop the site and lease the water to a number of turnkey manufactures.

Welland Canal

The Welland Canal crossed the Niagara Peninsula, linking Lake Ontario to Lake Erie. The difference between the levels of those two lakes is around 327 feet. Today the canal links Port Colborne on Lake Erie to Port Weller on Lake Ontario. Its eight locks cover a distance of 27 miles. The lift of this canal is more than the total lift of all of the St. Lawrence canals combined.

The Welland Canal is the most complex of the artificial waterways on the St. Lawrence route in terms of navigation and waterpower (figure 1b). Work on this canal started in 1824, and later improvements created four canals by relocating some sections. The first two canals were planned dual-purpose: navigation and waterpower. In fact, planners chose the final route for the First Canal to maximize hydraulic advantages rather than to provide the shortest route for navigation. When the canal came under the jurisdiction of the United Province of Canada in 1841, its Board of Works completed work on the Second Canal and made provisions for leasing hydraulic privileges there. It appears to have been assumed, by both businessmen and even by some members of the government, that this practice would be followed when the Third Canal was built (1872–1881). The northern section of this canal, however, bypassed the industrialized canal city of St. Catharines, which could still, however, draw water from the Second Canal.
The decision not to lease water privileges on the Third Canal was arrived at only after long and acrimonious debate. Over a period of years [from late 1881 through the 1890s], a number of studies and reports on the pros and cons of doing so were produced. Thorold and Merritton businessmen, with their many industries near the Second Canal locks, were naturally eager to avail themselves of new opportunities and waged a vigorous, if unsuccessful, campaign.[18]

By the time the decision not to lease waterpower on the Third Canal was made, water was more likely to be used to produce steam, or even hydroelectricity, and even the “new” Welland was beginning to show its age amid calls for a “deep waterway” to the Atlantic. The northern section of the Fourth Canal was even further from the established industrial canal cities of St. Catharines and Thorold, although it did run through both Welland and Port Colborne. “Even so, the water rights and the supply of water for industrial purposes in the hydraulic raceway and at the locks in the valley bottom remained.”[19] A study of the waterpower of Canada, undertaken by Leo G. Denis and Arthur V. White for the Commission of Conservation and published in 1911, recorded 30 mills still using waterpower along the Welland.[20] These were located mainly in St. Catharines, Merritton, and Thorold, along the First and Second canals. In fact, the DeCew generating station still draws water from the Fourth Canal, and the General Motors plant near lock 4 continues to use canal water.

A private company, the Welland Canal Co., incorporated in 1824 and said to be Canada’s first successful waterpower company, began construction of the Welland Canal.[21] Two years later, the St. Catharine’s Waterpower Company was created to attract industries between Merritton and St. Catharines. This company dug a 2.5-mile headrace above Lock 24 (Merritton) in order to channel water to a multilevel race system (figure 4) located between locks 3 and 6 (St. Catharines). This millrace system is the most sophisticated of all the systems examined in this article. Data from 1847 and 1882 indicate that the multilevel races supplied 10 users, which meant that this location reached its maximum capacity very early.[22] Even as the number of users of waterpower along the entire canal increased from 31 to 50, the number of users of the multilevel raceway remained the same. Without doubt, St. Catharines was one of the main water-powered industrial centers along the canal. Remains of the raceways can still be seen today behind the Canada Hair Cloth building (figure 4), and the name Race Street reminds us of the city’s hydraulic heritage. More traces of this heritage could probably be found archaeologically.[23]

Figure 4. Plan of the multilevel raceway of the St. Catharines Waterpower Company in St. Catharines, Welland Canal, in 1837 and 1852. Remains of a section of the raceways could still be seen behind the Canada Hair Cloth Building in St. Catharines in 2004.

The other Welland Canal water-powered industrial sites were located in Thorold, Merritton, and Dunnville (figure 5). In Thorold and Merritton short races and ponds between locks supplied mills. These were on the right side of the canal in Merritton and on both sides in Thorold. The construction of a dam in Dunnville to raise the level of Grand River to supply the feeder canal created a reservoir that provided head for mills located between the canal and the river.[24] The users leased directly from the government, as at Basin 2 on the Lachine Canal.

Figure 5. Along the Welland Canal, short races and ponds between the locks supplied the mills at Merritton and Thorold. The mills at Dunnville, at the head of the feeder canal, were supplied by a reservoir created by a dam across the Grand River. Maps courtesy of The Canadian County Atlas Digital Project, McGill University, Montreal, 2001 (accessed 6 August 2004).

Cornwall Canal

The Cornwall Canal was the first canal to be built according to the new navigation standards implemented by the commissioners of the Board of Works with 200 × 45 × 9 feet locks. Construction began in 1833, but the canal was not completed before 1843 after the Union Act. The canal circumvented one of the most important stretches of rapids on the upper St. Lawrence River, the Long Sault rapids. The canal was 11 miles long with seven locks (locks 15 through 21), rising 48 feet. An advertisement, published on 22 April 1846, announced a public auction on 11 May in Montreal regarding the lease of seven lots, concentrated around locks 15 through 18, for a 14-year period with weir and flume construction the responsibility of the lessee (figure 6). Lot one was on the north shore of the canal and required a waste weir at the head of lock 17, followed by a 230-foot millrace to bring water to this installation. It had a head of up to 23 feet. The other hydraulic lots sat in a row between the canal and the river and drew water directly from the canal, the greatest fall being 22 feet.

Figure 6. The location of weirs and millraces on the north and south sides of the lower entrance of the Cornwall Canal. Photos by author, 2004.

In 1847, the commissioners noted that it would be difficult to maintain a proper height of water for navigation while supplying water for industrial use in the quantities promised in the water leases.[25] They suggested building by-washes around the locks. On the south side of the canal, the lessees built a sluice gate for their mills as required in their leases. It was not before 1860 that the mills on the north side did the same. Six years later, an enlargement of the supply weir and the headrace of the guard lock and a new waste weir above lock 17 were required.

Douglas’s 1882 report indicated that seven companies occupied hydraulic lots on the Cornwall Canal.[26] Four companies had been installed north of locks 15, 16, and 17 since at least 1867: Canada Cotton Manufacturing Co., Lawrence Ballard Pottery, Cornwall Manufacturing Co. Woollen Mill, and William Mack Flour Mill. The Hodge & Bros’ Woollen, Planing, and Flour mills located on the south side of the upper portion of the reach, leased lots in 1879, along with the Stormont Cotton Manufacturing Co. They were using cumulatively about 1,940 hp.

The second phase enlargement of the St. Lawrence canals consisted, as at the Lachine Canal, in the construction of a second pair of locks alongside the old ones. Two locks (locks 15 and 17) on the south side paralleled the three first locks (locks 15, 16, and 17). In 1889, during enlargement operations, authorities agreed to an increase of 2 feet in the water level in response to an 1875 memorandum of mill owners and other interested parties.[27] The south raceway was maintained, but there were no installations of any kind remaining on this portion in 1929, based on photographs from that period. The old locks had been converted to dry docks. The north side raceway was still carrying water in 1958, but further research must be done to verify if waterpower was still being used.

The Cornwall Canal was abandoned in 1959, coinciding with the completion of the Wiley-Dundero Lock, part of the St. Lawrence Seaway, on the American side of the river. The canal was partly filled in 1971, but a section is still visible as well as some of the raceways. The Moses-Saunders Hydro Dam and plant replaced the old hydraulic lots, supplying hydroelectric power to regional industries. The new reservoir flooded the west section of the old canal, including locks 21 and 22 and the villages along it, up to Dickinson Landing at the upper entrance of the Cornwall Canal. No hydraulic lots were situated on that portion of the canal. The Domtar Paper Mill (figure 6), currently operating on one of the former hydraulic lots on the north side of lock 18, still has an intake. Most of the former industrial lots around locks 15 through 17 are now empty and await revitalization. The buildings of the Stormont Cotton Manufacturing Co., one of the most important cotton mills in Canada, have been demolished. A recreational park has been created over that section of the canal.

Williamsburg Canals

The Williamsburg Canals consist of three short canals: Farran’s Point, Rapide Plat, and Galops (figure 1a). The Farran’s Point Canal, built in 1847, was 1 mile long with one lock (lock 22) and a head of 31¼2 feet. A mill site beside the lock was granted in 1858. The Rapide Plat Canal, built the same year, was 4 miles long with two locks (locks 23 and 24) with heads of 10 and 8 feet, respectively. Mill sites were granted only at lock 23 at Morrisburg in 1852. The Galops Canal, built in 1846, was 71¼2 miles long after the completion of the Junction Canal in 1859. It had three locks (locks 25 through 27). The mill sites, auctioned off in 1849, were mainly located at Iroquois (lock 25) and Point Cardinal (lock 26), with heads of 8 and 10 feet, respectively. Altogether, the Williamsburg Canals had six locks rising around 32 feet over 30 miles.

According to the Douglas report, at least two hydraulic lots were granted free of charge, those at Farran’s Point (lock 22) and Point Cardinal (lock 26).[28] Felicity Leung indicated that free grants were given to landowners affected by canal construction.[29] In 1882, records indicate 11 mills on the canals, including grist, shingle, woolen, flour, saw, sash, and door mills, as well as starch and glucose factories. The Morrisburg, Iroquois, and Point Cardinal mills remained principally local industries. Only the Edwardsburgh Starch Company (Cardinal) and the Gibson’s flour mill (Morrisburg) attained wider notice.[30] The starch company, built in 1858 and now known as the Canada Starch company (CASCO), still exists, though a large part is below river level and is protected by a dike since the construction of the Seaway.[31]

The only substantially built-up areas along the St. Lawrence River in this region are those where the first canal-based water mills were located—Ingleside (Farran’s Point Canal), Morrisburg (Rapide Plat Canal), and Iroquois and Cardinal (Galop’s Canal).

Today, the Moses-Saunders Hydro Dam Reservoir covers most of the Williamsburg Canals and adjacent villages. The Farran’s Point and Rapide Plat canals are completely underwater except for a lock near Rapide Plat Point west of Morrisburg. All the villages along the shore were relocated.[32] The Old Galops Canal is partly visible, protected by a dike between Iroquois and west of Cardinal. The Iroquois Lock of the St. Lawrence Seaway replaced the old Iroquois Lock as well as all the others of the Williamsburg Canals. Even though these canals were designated as National Historic Sites of Canada in 1929, most are now underwater.

Beauharnois Canal

The second canal built between 1842 and 1845 in response to the new navigation standards was the Beauharnois Canal, connecting Lake St. Francis with Lake St. Louis, west of the Island of Montreal. Built solely for navigation, this canal was situated about 15 miles from the head of the Lachine Canal and replaced a number of old military canals (Cascades, Cedars, and Coteau). The Beauharnois Canal extended some 11 miles and had nine locks, numbered 6 to 14, with between 8 and 11 feet of head for each, for a total elevation of 82 feet.

A few years after the completion of the canal, problems appeared such as strong currents, low water, and flooding at the head of the canal at Lake St. Francis. Engineers proposed building a dam to control the flow of water, which was done in 1850. A December 1850 advertisement outlining the hydraulic privileges to be leased on the St. Lawrence canals, noted of the dam, “They have also created a splendid bay and an extensive water power, which, situated upon a line of navigation of such magnitude, cannot fail to become at some future day the seat of varied and important manufactures.”[33]

Improvements to facilitate the leasing of waterpower also included the construction of waste weirs at each lock, creating a total potential of 3,000 hp. Since most of this canal ran through a rural district, little demand for water privileges at the locks existed, the only lessee being at lock 7 (St. Timothée). The greatest power available was at the Valleyfield dam, estimated at 5,000 hp by promotional articles. This meant a total output of 8,000 hp. A 15-foot head was available at the dam on a year-round basis, according to the advertisements. In 1882, five industries shared 13 hydraulics lots east and west of the dam in addition to the lease at lock 7. The total power used was 2,387 hp, far below the estimated potential.

Douglas reported that the waterpower at the dam
… is the most valuable power along the whole line of canals. It is not subject to navigation, the water is not drawn off every year for repairs, there is but little back water, and with slight improvements to the channel into which the tail water is discharged, there could be no finer power found in any locality. There is a never failing supply of water, constant head summer and winter, requiring no auxiliary engines as at other mills upon the canals.[34]

His estimate of the potential in 1882, subsequent to the improvements and the headraces built by the lessees (figure 7), was 6,000 hp on the west side and 7,000 hp on the east side. This new estimate was three times that of 1850. This illustrates how difficult it is to get an accurate picture of the waterpower potential of a site, since potential depends on water velocity, head, and the efficiency of the equipment. Considerable improvements made in the efficiency of waterwheels between 1850 and 1880 may have contributed to Douglas’s higher estimate.

Figure 7. Location of the races at the Valleyfield dam, located at the upper end of the Beauharnois Canal. Also shown are drawings of the two most important installations: The Valleyfield Paper Mills of Alexander Buntin and The Montreal Cotton Mills Co.

Construction of the Beauharnois Canal and leasing waterpower at the Valleyfield dam transformed a little village of 25 families in 1843 into a thriving town with a population of 3,000 people by 1845 when the canal was completed. With the installation of companies in the hydraulic lots, population increased to 6,000 in 1892 and to 11,000 in 1900, which led to the creation of the city of Valleyfield. Montreal Cottons, located there, alone employed more than 2,500 people in 1910.[35]

The Beauharnois canal was closed to navigation in 1907, eight years after the construction of the Soulanges Canal on the north shore of the river. Between 1911 and 1951, the portion of the Beauharnois Canal linking Valleyfield to St. Timothée became a feeder canal for the St. Timothée Hydroelectric Plant. In 1970, the old canal was filled in, except for three sections in downtown Valleyfield. The old canal was recognized as a Quebec Historic Landmark in 2000. The modern Beauharnois Canal, part of the St. Lawrence Seaway, was built in 1932, both to provide water to the Beauharnois Hydroelectric Plant and for navigation.

Impact of the Use of Waterpower on Canals

The preceding sections show that mills along the canals were primarily concentrated around the locks. This was especially true on the Welland and Lachine canals where the first water privileges were awarded, but it was not the case on the Beauharnois Canal where the concentration was near the upper dam in the modern city of Salaberry de Valleyfield. In fact, the only hydraulic lot leased at a lock on the Beauharnois Canal was at St. Timothée (lock 7), even though sites were offered at all the locks. Dunnville, at the upper entrance of the feeder on the Welland Canal, also developed near a dam. Along the Cornwall and Williamsburg canals, as well as the main line of the Welland Canal, mills were concentrated around certain locks, which generated the creation of communities.

Figures 3 to 7 illustrate the spatial organization of the mill sites, which reflected physical environment and management conditions. Development was more extensive at St. Catharines and St. Gabriel, where a specialized corporation managed the sites. The Welland Canal had quite an intricate network of manmade races and ponds, particularly across the Niagara escarpment between Thorold and St. Catharines. These installations ensured a regular flow of water to the mills and probably contributed to a more constant level in the canal, particularly important for a summit canal.

Leung shows that even though the Welland Canal was longer, had a greater total fall with more locks, was begun earlier, and drove more plants than any of the other canals, larger industrial establishments emerged on the other St. Lawrence canals.[36] It also appears that industry developed most rapidly at Lachine, mainly due to its proximity to Canada’s largest city and port (Montreal), even if the Welland Canal’s hydraulic potential was far greater. The competition of nearby large industrial centers such as Hamilton and Toronto was probably another factor that limited the development of large industrial centers along the Welland Canal.

In 1882, the principal types of water-powered industries along navigation canals were silk, wood, flour, paper, wool, iron and steel, and cotton (Table 3). The 19th century saw not only diversification in the types of production, especially on the Welland and Lachine canals, but also growth in productivity. “Flour mills on the Lachine Canal were more productive from the 1860s, and cotton and paper mills on the Beauharnois and Cornwall canals were larger from their beginnings.”[37] In 1866, the total annual production of the five flour mills along the Lachine Canal was 261,801 barrels, while the eight flour mills along the Welland Canal produced 177,203 barrels or about 30 percent less production from 62 percent more mills. The Valleyfield Paper Mill, property of Alexander Buntin, was the largest paper mill in Canada in the 1860s. The Toronto Paper Manufacturing Company of Cornwall replaced it in 1888. The Montreal Cotton Company’s mill on the Beauharnois Canal was the largest water-powered cotton mill in the Dominion in 1882, while the Stormont Manufacturing Company, which used both water and steam power, was second in size.[38]Table 3 shows that manufacturers of flour, iron and steel, and wood were the most numerous along the St. Lawrence and Welland canals but that cotton mills had the largest labor force, which had a greater impact on the urbanization of the neighboring areas.[39]

Table 3. Types of Water-Powered Industries along the St. Lawrence and Welland Canals 1882
Type of Manufacture Number Employees
Iron and Steel
Total 1
115 50
Data from Douglas 1882:138 (see n. 7).

The most difficult aspect to document is the amount of waterpower used by canal-side water-powered plants. As mentioned previously, censuses do not generally give that information. Alain Gelly attempted to compile total waterpower production for the Lachine Canal and came to the conclusion that even the annual reports of the Public Works Department, which annually recorded the amount of power used along the canals, lacked accuracy.[40] In all cases, the recorder relied on the honesty and the knowledge of lease owners. Since lease fees were based on the amount of water used, lessees saw no advantage in declaring the real amount of water used to representatives of public works.

In the 1871 census the power declared by industries along the Lachine Canal totaled 2,203 hp, twice the amount officially leased.[41]Table 4 shows that 50 percent more water was used than leased in 1882. On the Lachine Canal and the Williamsburg Canals, by this time, the amount of waterpower used was more than three times greater than that officially leased. Douglas noted that, even for his inventory, he did not really measure the amount of water used but often relied on the declarations of the occupants. This situation makes it difficult to determine the amount of waterpower really used on Canadian navigation canals and to make accurate comparisons between sites and periods. It also indicates a major problem: lack of good governmental regulation.

Table 4. Waterpower Leased and Used along the St. Lawrence and Welland Canals
Location 1831 1851 1882 1882 1882
Leased HP % Leased HP % Leased HP % Used HP % Wheels No. Head
Lachine Canal
Bassin 2
Côte St. Paul
Beauharnois Canal
Cornwall Canal
Williamsburgh Canals
Welland Canal
Main Line
Dunnville and feeder
St. Catharines
Water Power Co.




100 1,000


2,630 38.02


100 1,000

6,747 14.82


100 3,630

1,3083 27.75


100 102

290 9.6 to 19
9 to 13
8 to 22
3.6 to 10

7 to 18
5 to 8

9 to 22
Data from Douglas 1882:135 (see n. 7).

The amount of water leased increased 90 percent from 1831 to 1851 and 61 percent from 1851 to 1882. The first period reflects the initial transformation of navigation canals into dual-purpose canals. The second period indicates a consolidation of hydraulic power along those canals. Only the Lachine Canal showed no change, suggesting that all its hydraulic lots were already leased in 1851. Larry McNally points out that the manufacturers along the Lachine Canal were a closed society, transferring the leases among themselves.[42] It is also true that waterpower development was limited and that new installations also used water to produce steam and for certain manufacturing processes. In 1871, on the Lachine, approximately 40 industries used waterpower and 20 used steam power.[43] Even if it was not possible to increase the number of water-powered mills, industrialization and urbanization continued to grow along the canal.

Waterwheels and Their Improvement[44]

One of the Douglas report’s most valuable contributions to the history of technology is its compilation of the type and the number of wheels and turbines used and the water discharged by canal-side plants. One of the major problems noted by Douglas was the inefficient use of water by the industries along the canals.[45] He argued that more than 50 percent of the waterwheels in operation should be prohibited on navigation canals. For example, he maintained that 85 percent of the wheels at Basin 2 of the Lachine Canal and 81 percent of those at Dunnville on the Welland Canal were inadequate. Those at Basin 2 of Lachine, he stated, were “of a description unworthy of a primitive saw-mill in the backwoods of Canada.”

Douglas identified more than 24 different types of waterwheels (Table 5), representative of the more popular wheels on the market from 1840 to 1880. The Leffel and Tyler waterwheels, which comprised 44 percent of the total, were the most popular. Those models had been put on the market, according to Arthur Safford and Edward Hamilton, at the beginning of the 1870s, so they would have been around 10 years old at the time of the Douglas inventory.[46] The maximum efficiency of these turbines varied from 74 percent to 81 percent (Table 6). Their capacity was twice that of central discharge turbines, an old Francis model of the 1850s. In fact, turbine efficiency and capacity increased significantly during the 1870s, so that newer turbine installations could potentially produce three times more horsepower.
Speed was increased by reducing the wheel diameter. Greater capacity was obtained by increasing the depth and later the width and openings of the buckets and extending them toward the wheel center—for the inward-flow principle was early adopted. This last change increased the volume of flow and the wheel’s capacity, but left little room to accommodate the increased discharge at the center. It was necessary, therefore, to turn the buckets downward and then outward to dispose of the water. The end result was the movement of water through the wheel in a continuously and smoothly curving spiral path. Then it was found that reducing the number of buckets and widening the distance between them both increased the wheel’s capacity and minimized its clogging from trash.[47]

Table 5. Types of Waterwheels and Turbines on the St. Lawrence and Welland Canals 1882
Wheel or Turbine Lachine Beauharnois Cornwall Williamsburg Wekkand Total
Leftel (1871)
Tyler (1873)
Wooden Central Discharge
Little Giant
Hercules (1876)
Turbine Class
New American (1884)
Fourneyron (1850)
Sampson (Cole, Samson) (1870)*
Wooden Wheel
Risdon (1873)
American (1859)
Jonval (1850)
Victor (1878)
Slayter Central Discharge
Butterfly Turbine Class
Inglis and Hunter Make
New Dominion
To be put in
Total 34




103 11




34 4



24 8



28 19





102 76
*All the Sampson, Samson, and Cole models are combined (see text).

Data from Douglas 1882: 62–65, 80–81, 90–91, 104–105, 142–153 (n, 7); date introduced in parentheses (Hunter 1979, [n. 47], Safford and Hamilton 1922 [n. 46]).

Table 6. Efficiency and Market Dates of Some American Turbines 1847–1900
introduced Type RPMs at
efficiency HP
maximum Maximum
1899 First Francis Model
Tyler Scroll
Special Hercules
Smith-McCormick 44
53 0.19
1.65 71.6
Data based on a 30-inch wheel under a 1-ft. head.
Data from Safford and Hamilton 1922: 1272 (n. 46).

In addition to the Leffel and Tyler turbines, the Douglas inventory recorded vertical wheels, undefined wooden wheels at the Williamsburg Canal, overshot wheels at Dunnville on the Welland Canal, reaction wheels also on the Welland Canal, transition models, and late model turbines.

For the Welland Canal, Douglas had both 1847 and 1882 data (Tables 5 and 7). Comparison illustrates the transition from traditional vertical wheels to water turbines. Traditional vertical wheels comprised 18 of 39 wheels on the canal in 1847, but only 6 of 102 by 1882. Douglas reported no vertical wheels on the Lachine Canal in 1882, but the basement of the former Livingston Linseed Oil Co. (a branch of Dominion Linseed in Toronto) has two division walls creating three spaces, suggesting the use of vertical wheels there at some time (figure 8). Douglas lists Tyler and Leffel turbines there in 1882.

Figure 8. Basement of the Meunerie Rozon (formerly Livingston Linseed Oil Co.) at Basin 2 (Lachine Canal), after demolition of the building, showing the stonewalls probably related to the installation of former vertical breast waterwheels.

Table 7. Types of Vertical and Horizontal Waterwheels on the Welland Canal 1847
Welland 1847 No. of Wheels
Total 1
Data from Douglas 1882: 154–57 (n. 7).

The improvements that turbine replacement could bring can be easily seen in the difference between Tyler, center vent, and other older turbine models and the newer Hercules models (figure 9). The Tyler wheel had concave buckets. Water entered from the side and discharged outward. John Tyler patented this turbine in the United States on 22 May 1855 (no. 12,927). Two Tyler wheels were installed in August of the same year at the Lachine Canal in Montreal. Alexander Fleck patented an improvement to the Tyler wheel in Canada on 31 March 1863 (series 1, no. 1,511), and T. H. Risdon and Tyler did the same in 7 May 1878 (no. 8,750).[48] The improved Canadian models were preferred in Montreal and account for two-thirds of the Tylers used along the St. Lawrence and Welland canals.

Figure 9. Details of the Tyler and Hercules turbines. The Tyler turbine (top) was copied from the John Tyler trade catalog (Old Sturbridge Village Research Library). The Hercules turbine (bottom) was copied from the Holyoke Machine Company trade catalog (National Museum of American History). Also shown is a detail of the runner (Emerson 1892: 316). On the right, bucket outlines of runners showing the evolution of the mixed-flow turbines (Safford and Hamilton 1922: 787).

All the center-vent turbines in Douglas’s inventory were located at Basin 2, 80 percent in the flour mills of James McDougall and Ira Gould & Co. The wooden central discharges were mostly concentrated at Dunnville (Welland Canal), where they represented with the reaction wheel 81 percent (13/16) of the total number of waterwheels at the site. Douglas’s evaluation of the percentage of inadequate waterwheels at Dunnville and at Basin 2 corresponds exactly to the number of aging center-vent, central discharge, reaction, Cole, and Tyler wheels at those sites.

The newer Hercules turbines were very popular and the first, among the mixed-flow turbines, to use elongated buckets with three additional blades (figure 9), considerably increasing the action of the water on the wheel. According to the Safford and Hamilton data (Table 6), the maximum horsepower produced by a Hercules was 3.5 times that of a Tyler or a Leffel turbine and 7.5 times that of a central discharge, due to a combination of its ability to use a greater amount of water and its higher efficiency.[49] Hercules models were used by Buntin & Co. and Montreal Cotton Co. at Valleyfield (Beauharnois) as well as Stormont Cotton Manufacturing Co. at Cornwall. These companies were the largest of their type in Canada. The last two were new installations, but the Buntin Paper Mill was established in 1855.

In the 1880s (Table 8) Hercules wheels replaced almost all of the central discharges and the center vents at Basin 2 (Lachine Canal). Based on the potential horsepower developed (Table 4), seven Hercules produced an average of 8.97 hp, replacing 24 center vents, which produced an average of only 3.23 hp each.[50] James Shearer, a lumber dealer located at St. Gabriel Lock, Lachine Canal, wrote the Holyoke Machine Company on 5 March 1885:
I am pleased in expressing to you my entire satisfaction with your 45-inch “Hercules” wheel, put into my factory now near two years ago,—and I thank our esteemed master millwright Thomas Pringle, of this city, for recommending and erecting it in my place. I may say I have had many wheels, but none have given the power this has, and I have been using water power for thirty years. I have great pleasure in recommending it to any in need of a powerful and perfect wheel.[51]

Table 8. Evolution in the Types of Turbines Used at Basin 2, Lachine Canal
Basin 2
Central Discharge
Little Giant
Total 1882


53 1887

38 1910



*or Sampson or Cole
Data from Douglas, Confidential Reports, 1882 (n. 7);
Canada Government, Report of Royal, 1887 (n. 13);
Viger, “Croquis et détails,” 1910 (n. 13).

The Leffel and the Little Giant illustrate another type of turbine modification occurring on Canadian navigation canals (figure 10). The Leffel turbine had a double runner, “the upper half being a radial inflow runner of the Francis type and the lower half consisting of a runner with inward radial admission and axial discharge.”[52] This wheel was one of the archetypes from which the American turbine or mixed-flow turbine developed. It was designed and built by the James Leffel Company of Springfield, Ohio, starting around 1860. The Little Giant waterwheel was also a double wheel but, in this case, the two runners were mounted with their backs together on the same shaft, one discharging downward and the other upward. This type of wheel appeared on the market by about 1875, as indicated by an advertisement of the Little Giant Water Wheel Works of J. C. Wilson & Co. The Munson Mill Machinery Co. Inc., Utica, and George H. Jones, Auburn, N.Y., also made this type of turbine.[53]

Figure 10. The Leffel and the Little Giant double-arrangement turbines used on the St. Lawrence and the Welland canals: the Leffel, as illustrated in the Joseph Hall Manufacturing Co. trade catalog (National Library of Canada) and the Little Giant turbine manufactured by J. C. Wilson & Co. (trade catalog, Old Sturbridge Village Research Library). One of the Little Giant turbines found in the basement of the Meunerie Rozon building at Basin 2, Lachine Canal is also shown (bottom right). Photo by Johanne Murray for the Old Port of Montreal.

Emerson, a leading American turbine expert, criticized the use of multiple runners on the same shaft, claiming, “years of experience and demonstrations by decisive tests prove beyond chance for dispute that the double arrangements are less effective than simple turbines.”[54] Since Leffel and Little Giant turbines often used this arrangement and represented one-third of the total turbines in use in 1882, and since, as Hunter notes, more than 8,000 Leffels were sold from 1862 to 1880, Emerson’s judgment must have been flawed.[55] At least five Ontario companies (McGill Manufacturing Co., Oshawa; Joseph Hall Manufacturing Co., Oshawa; Thornton & Ewart, Oshawa; Nicholls & Co., Port Hope; and Paxton, Tate & Co., Port Perry) and one Quebec company (the St. Lawrence Engine Works, Montreal) distributed the Leffel. All the Little Giants came from J. C. Wilson & Co., Picton, Ontario. The ready availability of these models in Canada may have influenced their popularity on Canadian navigation canals.

One of the difficulties in working with Douglas’s list of waterwheels is that different models use the same name. For example, Wm. Dolan & Co. of Logansport, Indiana, manufactured a model with the name “Little Giant.”[56] It was a single runner, radial, inward-flow turbine, which was later modified to a mixed flow of a modern type (figure 9). It is not clear, from the list, whether the Little Giants used at Lachine Canal were Dolan or Wilson models. The Little Giants found in the basement of the former Livingston Linseed Oil Co. were double-wheel models manufactured by Wilson & Co. (figure 10).[57] The Little Giants set up on the other canals were all of the double model made by Wilson & Co.

The Samson model is also confusing. The James Leffel Company manufactured the best-known Samson model, regularly mentioned in the literature, introducing it around 1897.[58] The Samsons or “Sampsons” listed by Douglas, however, cannot be the Leffel models since his inventory dates from 1882, almost 15 years before the introduction of the Leffel Samson. The Samson or Sampson turbines installed along the Welland and St. Lawrence canals were probably a Canadian design patented by Ashley Dodge Cole. He obtained his first Canadian patent for a waterwheel in 1855 (no. 524), when living in Sherbrooke, Quebec. He moved to Toronto and obtained other patents for improvements in 1870 (no. 498, extended for 5 more years in 1875 no. 4,966), 1871 (no. 1,210), 1873 (no. 2,241), 1874 (no. 3,370, extended for 5 years in 1879 no. 9,898), and 1887 (no. 25,749). He also patented his turbine waterwheel in the United States on 24 October 1876 (no. 183,490). A. W. Cole from Oswego, New York, manufactured this Samson waterwheel. Is there a family link between these two Coles? Is there some continuity between this model and the later Leffel one? More research has to be done to answer those questions.

At least three Canadian manufacturers sold the Cole Samson wheel: St. Lawrence Engine Works in Montreal; Dickey, Neill & Co., Soho Foundry, in Toronto (1871); and John Abell, Woodbridge Agricultural Works, County York, Ontario (1872).[59] Moreover, William Hamilton Manufacturing Co. Ltd. in Peterborough advertised a Sampson turbine in 1884.[60] This company was the official Canadian distributor for James Leffel & Co., adding to the confusion.

A 1910 report documents the presence of both older and newer Samson models along the Lachine Canal.[61] One Sampson was listed at the Record Foundry, noted as “40 years old.” A Sampson was also recorded at that site in 1882 and 1887, probably the same turbine. The lot where the Sampson was located had been formerly occupied by the Iron Foundry and Machine shop of W. P. Bartley, proprietors of the St. Lawrence Engine Works, agents for the Cole or Samson turbine. The other five reported Sampsons had replaced earlier turbines such as the Leffel, the center vent, and the Hercules. These turbines are likely the modern Samsons developed by Leffel in the 1890s. Their output certainly suggests this, for the 56- and 68-inch Sampsons installed at the Royal Flour Mills produced 300 and 600 hp, respectively, under a head of 17 feet. Another model name, “Dominion,” replaced (in 1887) the Cole or Samson model recorded in 1882 at the G. & W. Tate Dry Dock. The same A. W. Cole, mentioned above, manufactured this Dominion turbine. Was it an improvement to the former Cole model patented by A. D. Cole or a new model without any relation to A. D. Cole?

The presence of a “New American” turbine at Cornwall in 1882 is also very surprising. According to Daniel Mead, this turbine was brought out only in 1894.[62] A “Standard New American” is listed in 1884, which is still too early. The capacity of the initial American was doubled by the Standard New American and tripled by the New American (Table 9).

Table 9. The Development of American Turbines in Capacity, Speed, and Power
Model Date
introduced Discharge
in cu. ft. RPM HP
Standard New American
New American
Special New American
Improved New American 1859
1903 3,271
13,234 102
139 79.1
Note: Calculations of a 48-inch turbine under a 16-foot head. Data from Mead, Water Power, 1920: table 22 (n. 62).

Persistence of Direct-Drive Waterpower

Unfortunately, comparative information has not been found regarding the persistence of direct-drive waterpower generation during the 20th century on most Canadian navigation canals. On the Lachine Canal, however, surviving records indicate that it persisted right up to the canal’s closing in the 1960s. For example, insurance plans clearly indicate that some industries continued to use water for power, even as they also began to make use of electricity for lighting and steam for heating.[63] The Lachine Canal industries used all kinds of combinations from direct connection between waterwheels and machinery, to adding an alternator and an electric motor, to using electricity from a general distribution grid.

Table 10 illustrates the decline of waterpower along the Lachine Canal during the 20th century. Because the sources employed to compile the table used a variety of different measures for waterpower (runs of stone, number of inches over the wheel, inches used, cubic feet per second discharged), part two of the table had to be separately calculated. Mead’s formula was used, where the theoretical power is equivalent to the product of the flow of water in cubic feet per second and the head in feet, divided by the constant 8.8.[64] The turbines were assumed to have an efficiency of 75 percent, based on averages from Douglas’s report and taking into account that older turbines were still in use. These results must be taken as approximations. The resulting table shows a continuous decrease in the number of companies using waterpower along the canal from 1867 to 1950, reflecting the concentration of control over hydraulic lots. On the other hand, the quantity of power developed increased from 1851 to 1910, reflecting the installation of progressively newer and more efficient turbines. Only after 1910 did the quantity of waterpower developed along the Lachine Canal decline, impacted, no doubt, by the emergence of hydroelectric power and long-distance distribution networks.

Table 10. Companies Using Waterpower and Estimated Quantity of Horsepower Produced along the Lachine Canal (1867–1960)
Location 1867 1871 1882 1908 1930s 1948 1950s
Basin 2
St. Gabriel
Cote St. Paul
Total 13
43 10
38 9
37 7
19 4
12 2
7 1
1882 data from Douglas, Confidential Reports, 1882: 62–65 (n. 7);
1908 to 1948 data from Gelly, “De l’eau,” 2001: 241–273 (n. 1);
1950 dala from Cureton, “The Lachine Canal,” 1957 (n. 65).

Esthnated quantity of horse power developed
Years 1851 1871 1882 1887 1910 1948

Total 600

600 2203

2203 3630

3630 4390
4489 4000
4642 875
Calculation based on HP=Qh/8.8. Q represents the flow of water in cubic feet/second, h represents the effective fall in feet. Mead, Water Power, 1920: 147 (n. 62).

Nonetheless, some companies continued to use direct-drive waterpower. In the 1950s, from a survey, Allan Cureton reported five companies still used waterpower for their machinery at Côte St. Paul (Lachine Canal), even though they also used steam for heating and electricity for lighting.[65] The 1955 underwriters insurance plans for Montreal, which show the races and location of the turbines, confirm this.[66] The 1964 insurance plans indicate that waterpower was still used at one of the buildings of Bancroft Industries Ltd, next to the lot where the Little Giant turbines were located.[67] Dominion Linseed Oil, owner of that lot, had indicated that it used waterpower in 1948, discharging 100.8 feet per second.[68] It is not possible to determine if the power transmission came directly from the waterwheel shaft or was assisted by an alternator and an electric motor.

The City of Montreal demolished the Côte St. Paul industries in 1967 to enlarge St. Patrick Street. Recent archaeological studies focused on trying to map the infrastructure relating to waterpower use in this sector. Archaeological work is also underway at St. Gabriel, where the City of Montreal has purchased some of the most important hydraulic lots and is planning an archaeological park dedicated to the history of waterpower and industry along the Lachine Canal.

Table 11 shows that direct-drive waterpower persisted at the other navigation canals in 1911. The horsepower developed by canal-based industries, in fact, increased by more than 40 percent between 1882 and 1911, suggesting that waterpower continued to be preferred by canal-based industries over steam. Only with the coming of hydroelectricity would local generation of power on navigation canals drop abruptly.

Table 11. Use of Waterpower for Industries and for Hydroelectricity along the St. Lawrence and Welland Canals 1911
Power Site Minimum Electrical Energy Used, 1910 (HP) Developed (HP) Remarks
(in feet) HP Power Light Electrical
Energy Paper
and Pulp Other
Industries Total
Mill Street 11 332 332 Electricity used on canal
Mill Street 17 2,600 2,600 Used for flour mills, rolling mills, etc.
St. Gabriel Locks 8.5
Côte St.Paul 8.5 310 310 310 Electricity used on canal
1,400 1,400
Total 0 0 642 310 0 4,000 4,642
St. Timothée 50 21,000 21,000 In construction, current to be used in Montreal
Valleyfield 10 135 135 7,000 7,135 Utilized by Montreal Cotton
Total 0 0 135 21,135 0 7,000 28,135
Mille Roches 20 1,000 300 2,700 2,700
Lock 20 8 1,000
Lock 19 6 760
Lock 18 7.5 950 800 800 Toronto Paper Mfg. Co.
50 50 Municipal pumping plant
Lock 17 20 2,540 1,500 1,500 Canada Coloured Cotton. Co.
1,200 1,200 Canada Coloured Cotton. Co.
200 200 Canada Coloured Cotton. Co.
80 50 130 Cornwall Elect. Lt. & Street Ry. Co., and Wm. Hodge flour mill
Total 6,250 300 2,700 80 800 3,000 6,580
Cardinal (26) 6 200 200 Edwardsburg Starch Works
Iroquois (25) 14 40 40 M.F.Beach
20 70 90 90 Town of Iroquois pumping and electric plants
Morrisburg (23) 11 1,410 Gibson lease, 60 H.P
50 300 250 250 Town of Morrisburg pumping and electric plants
1,100 1,100 New municipal lease for electric power
Total 1,410 70 370 1,440 0 240 1,680
Port Dalhousie 12 500 500 Rubber mfg.
St. Catharines 9.5 325 325 Welland Vale Mfg. Co.
11 150 150 Lincoln Electric Light and Power Co.
12 320 320 Whitman & Barnes
12 2,000 2,000 2,000 Carbide works
21 600 600
19 260 260 Milling Co.
13 50 50 Hair cloth mfg.
19 100 100 Black & Son
19 150 150 Electric Mfg. Co.
Merritton 12 100 100 Wheel works
15 200 200
11 25 25 James Wilson
12 1,350 1,350 McLeary & McLean
14 190 190 McLeary & McLean
Thorold 14 150 150 Penman Mfg. Co.
14 300 300
12 340 340
12 30 30 Monroe & Roantree
12 150 150 Town of Thorold
12 300 300 Milling Co.
23 625 625 James Davy
23 400 400 Foley-Reiger Co.
11 10 10 Est. J. Battle
Port Robinson 10 25 25 Hyslop Bros.
Welland 8 100 100 Town of Welland
16 60 60 Hyslop Bros.
Cataract Co., De Cew 260 50,000 50,000 Hamilton Cataract P. L. & T. Co.;
estimated volume used 1800 c.f.s.
Marshville 10 25 25 D. Piggott
Dunnville 7 100 100 400 500 Grist and woollen mills
Total 0 0 2,100 52,250 2,790 4,295 59,335
Grand Total 7,660 370 5,947 75,215 3,590 18,535 100,372
Data from Denis and White, Water-Powers 1911 (n. 5).


Waterpower from canals was very attractive to 19th-century industries because it was reliable and cheap and brought with it ready availability to a means of transportation for raw materials and finished products. The industries that took advantage of the water privileges offered, beginning in the 1840s, by the government contributed to the urbanization of selected areas (Ste-Anne, St. Henry, Côte St. Paul, St. Catharines, Thorold, Cornwall, Valleyfield, etc.). With the evolution of technology, the use of direct-drive waterpower diminished as steam and then hydroelectricity replaced it. The reorganization of the canals with the opening of the St. Lawrence Seaway in 1959 and changes in the industrial economy marked the end of industrial activities along many of the old dual-purpose canals. They must now be revitalized for a different purpose: preserving their role in our history.

Some projects are currently being developed, most related to tourism and recreation, such as trails, green space, and ruins parks. Some projects reuse old mills for restaurants, housing, or offices. More must be done in that direction, but good inventories and archaeological investigations have to be included in all new development projects to insure the preservation of the knowledge retained by the remains. Canal-based waterpower sites were extended sites and economically interrelated. Redevelopment thus must be done with a broad view and necessitates a close partnership among all the local authorities concerned with these sites.


A part of that research was sponsored by the Fonds pour la Formation de Chercheurs et l’Aide à la Recherche (FCAR). The author wishes to thank the reviewers and the editor for their judicious comments as well as Roberta M. Styran, Gisèle Piédalue, Terry Reynolds, Jacques Lecours, Dennis Carter-Edwards, Aimée Belmore, and John Fisher for their help on content and form.


1. The most comprehensive study on direct-drive waterpower in Canada is Felicity L. Leung, “Direct Drive Waterpower in Canada: 1607–1910” (Ottawa: Parks Canada, Microfiche Report Series 271, 1986), which contains a specific chapter on the St. Lawrence and Welland canals. There are also studies by John Willis, “The Power of Water: Hydraulic Power and Hydro-Electric Power in the History of Ontario, From the Late-Eighteenth Century to 1960, ” report for National Museum of Science and Technology, Ottawa, 1990; and Catherine MacDonald, “Water Power and the Transformation of Canada, 1600–1960,” in Historical Assessments, report for National Museum of Science and Technology, Ottawa,1992. Other specific studies are of interest, such as Roberta M. Styran and Robert R. Taylor, Mr. Merritt’s Ditch—A Welland Canals Album (Erin: Boston Mills, 1992); and John N. Jackson, The Welland Canals and Their Communities: Engineering, Industrial, and Urban Transformation (Toronto: Univ. of Toronto Press, 1997). For the Lachine Canal, see Larry McNally, “Water Power on the Lachine Canal 1846–1900” (Ottawa: Parks Canada, Microfiche Report Series 54, 1982); John Willis, “The Process of Hydraulic Industrialization on the Lachine Canal 1840–1880: Origins, Rise and Fall” (Ottawa: Parks Canada, Microfiche Report Series 322, 1987); Pauline Desjardins, “L’organisation spatiale du corridor du canal de Lachine au 19e siècle” (doctoral diss., Univ. of Montreal, 1999); and Alain Gelly, “De l’eau et de la fumée: Forces motrices au canal de Lachine 1846–1940,” report for Parcs Canada, Québec, 2001.

2. Terry S. Reynolds, Stronger Than a Hundred Men: A History of the Vertical Water Wheel (Baltimore: Johns Hopkins, 1983).

3. Desjardins, “L’organisation spatiale,” 67–87 (see n. 1); Desjardins, “From the Warehouses to the Canal by Rail ca.1830: The Lachine Canal, Montreal, Quebec,” Northeast Historical Archaeology 28 (1999): 58; and Leung, “Direct Drive,” 131 (see n. 1).

4. Jackson, Welland Canals (see n. 1).

5. Leung, “Direct Drive,” 42–47 (see n. 1); Leo G. Denis and Arthur V. White, Water-Powers of Canada (Ottawa: The Mortimer Co., Ltd., 1911).

6. About conflicts resulting from the use of canals for waterpower and navigation, see Larry McNally, “Engineers and Waterpower on the Lachine Canal, 1843–1871,” Canadian Journal of Civil Engineering 20 (1993): 343–48.

7. 7. Robert C. Douglas, Confidential Reports to the Hon. Sir Charles Tupper, K.C.M.G., C.B., Minister of Railways and Canals, on the Hydraulic Powers Situated upon the St. Lawrence and Welland Canals (Ottawa: MacLean, Roger & Co., 1882).

8. For 1999 Hydro Québec, La production d’électricité et les émissions atmosphériques au Canada et aux États-Unis (Montreal: Carto-Media, 2000).

9. Leung, “Direct Drive,” 96 (see n. 1).

10. “Canadian Industry in 1871 Project (CANIND71)” (MS, Ontario: University of Guelph, 1991). All the 1871 data mentioned in this paper comes from this database. I only used data related to water-powered industries established along the canals studied in the present article. As shown by Table 1, this data is not really representative of all industries.

11. See these authors for more specific critiques about the use of censuses: Leung, “Direct Drive” (n. 1); Robert Lewis, Manufacturing Montreal: The Making of an Industrial Landscape, 1850 to 1930 (Baltimore: Johns Hopkins, 2000), 273; G. T. Bloomfield and Elizabeth Bloomfield, “Water Wheels and Steam Engines,” Canadian Industry in 1871, 2 (1989): 3. The 1871 census is the most informative, as demonstrated by CANIND71, and we will use it more heavily than others in this paper (see n. 10).

12. McNally, “Engineers and Waterpower” (see n. 6).

13. Canada Government, Report of Royal Commission on the Leasing of Water-Power Lachine Canal (Ottawa: MacLean, Roger & Co., 1887); G. Viger, “Croquis et détails des turbines des lots hydrauliques du bassin 2, écluses de Montréal” (Ottawa: Ministère des chemins de fer et des canaux, 1910).

14. The numbering of locks for St. Lawrence Canals was standardized after the improvements in the 1840s. The new canal authority numbered them from east to west. Lock one is on the Lachine Canal in the Old Port of Montreal and lock 27 is the westernmost one on the Williamsburg Canals.

15. This drop could diminish to 13 feet in winter. There is no tide at this point on the St. Lawrence, but the level of water increased considerably in the winter and was responsible for some memorable floods in the lowlands of Montreal.

16. This system was ideal according to Robert B. Gordon and Patrick M. Malone, The Texture of Industry: An Archaeological View of the Industrialization of North America (New York: Oxford Univ. Press, 1994), 97.

17. There were some irregularities and many complaints around this sale, enough to require setting up a Royal Commission in 1887. See Canada Government, Report of Royal (n. 13); Douglas, Confidential Reports, appendix Memoranda (n. 7); Desjardins, “L’organisation spatiale,” 158 (n. 1).

18. See Roberta M. Styran and R. R. Taylor, The “Great Swivel Link”: Canada’s Welland Canal (Toronto: Univ. of Toronto Press for The Champlain Society, 2001), cxiii.

19. Jackson, Welland Canals, 148 (see n. 1).

20. Denis and White, Water-Powers, 113–15 (see n. 5).

21. A private company, The Proprietors of the Lachine Canal planned the Lachine Canal, but they could not sell enough shares to cover the cost of construction in 1819. Finally, the Government of Lower Canada took over the project and completed the construction with the help of the British Government.

22. Douglas, Confidential Reports, 142–57 (see n. 7).

23. See Christopher Andreae, “Archaeological Excavation of Lock 24, First Welland Canal,” report for Historical Research Limited, London, 1988; Michelle Greenwald and Alan Levitt, “The Welland Canals: Historical Resource Analysis and Preservation Alternatives,” in Heritage Planning Study 1, report for Historical Planning and Research Branch, Ontario Ministry of Culture and Recreation, Toronto, 1979.

24. Dunnville was named after J. H. Dunn, receiver general of Upper Canada (1820–43) and president of the Welland Canal Co. (1825–32).

25. “Annual Reports by the Board of Commissioners for 1847,” compiled by Dennis Carter-Edwards 1998 for Parks Canada, Cornwall, Ontario.

26. Douglas, Confidential Reports, 90, 135 (see n. 7).

27. “Annual Reports by the Board of Commissioners for 1889,” compiled by Dennis Carter-Edwards 1998 for Parks Canada, Cornwall, Ontario.

28. Douglas, Confidential Reports, 104 (see n. 7).

29. Leung, “Direct Drive,” 133 (see n. 1).

30. ibid., 137.

31. Municipality of Cardinal (accessed July 2004).

32. There is a museum and a Web site about the history of these “Lost Villages,” The Lost Villages Historical Society (accessed May 2003).

33. “Reports of the Engineer of Public Works” in Douglas, Confidential Reports, 73 (see n. 7).

34. Douglas, Confidential Reports, 78 (see n. 7).

35. Roland Viau. “Vie et mort d’une route d’eau: Patrimoine historique et potentiel archéologique de l’ancien canal de Beauharnois,” (MS, Ministère des Transports du Québec, Montréal, 1988), 30.

36. Leung, “Direct Drive,” 152–53 (see n. 1).

37. ibid., 152.

38. ibid., 153–63.

39. Note that this paper focuses on water-powered mills so it does not include manufacturers using other sources of power such as steam. For example, the Redpath Sugar Refinery at St. Gabriel is not included in this study even though it was very important in the development of the sugar industry because it was steam powered.

40. Gelly, “De l’eau,” 111 (see n. 1).

41. Data drawn from CANIND71 (see n. 10). Gelly, “De l’eau,” 152, gives a total of 2,255 hp (see n. 1), and McNally “Water Power,” 89 (see n. 1), gives 2,122 hp.

42. McNally, “Water Power” (see n. 1).

43. Yvon Desloges and Alain Gelly, The Lachine Canal: Riding the Waves of Industrial and Urban Development 1860–1950 (Quebec, Septentrion, 2002), 111.

44. This study uses the terms wheel and waterwheel as general terms to designate vertical wheels and reaction wheels, as well as turbines. This is in keeping with historical usage.

45. Douglas, Confidential Reports, 134 (see n. 7).

46. Arthur T. Safford and Edouard P. Hamilton, “The American Mixed-Flow Turbine and Its Setting,” Transactions of the American Society of Civil Engineers 48, no. 4 (1922): 758–808. We used their standard table, but the dates must be revised using patent data. The Tyler, for example, was first patented in 1855 and improvements were made in 1856, 1858, 1864, 1866, and 1872. A new scroll wheel was patented in 1874. So, depending upon the model, the date could change as well as the capacity and the efficiency.

47. Louis C. Hunter, A History of Industrial Power in the United States, 1780–1930, vol. 1 in Waterpower in the Century of the Steam Engine (Charlottesville: Eleutherian Mills-Hagley, 1979), 359–60.

48. A new patent act was issued in 1869 with a new series of numbers, so I added “series 1” to patents issued before 1869 but no special indication for those following 1869.

49. Some prudence must be exercised in using Safford’s and Hamilton’s generalizations, for they depend in part on James Emerson, who, in his Treatise Relative to the Testing of Water Wheels and Machinery (Williamsett, Mass.: Holyoke Testing Flume, 1892), 306, admitted approximating the performance of a 60-inch Tyler wheel in 1879: “By comparing its cost, capacity of transmission, and general efficiency with the Hercules, Victor or New American, its relative value may be approximated.”

50. This calculation is based on a 30-inch wheel under a 1 ft. head. These Hercules turbines were from 36 to 54 inches under 13 to 18 feet of head.

51. Holyoke Machine Co., The “Hercules” Turbine (Lowell, Mass.: Holyoke Machine Co., 1885), 47.

52. Emerson, Treatise Relative, 223 (see n. 49).

53. Felicity Leung, “Grist and Flour Mills in Ontario: From Millstones to Rollers, 1780–1880” (Ottawa: Parks Canada, Microfiche Report Series 201, 1976), 279; J. C. Wilson Company, Illustrated Catalogue of The Little Giant Turbine Water Wheel and Mill Machinery (Toronto: Imrie, Graham & Harrap, 1924); Munson Mill Machinery Company, Little Giant Turbine Water Wheel (Auburn: Fenton Press, Catalog No. 91, 1933); Merrell, Wilder and Company, Geo. H. Jones’ Little Giant Turbine Water Wheel (Auburn: Merrell, Wilder & Co., 1870).

54. Emerson, Treatise Relative, 214 (see n. 49).

55. Hunter, History of Industrial, 366 (see n. 47).

56. William Dolan and Company, Catalogue of the Improved “Little Giant” Cylinder Gate Turbine Water Wheel (Logansport, Ind.: Longwell & Summings, 1895) conserved at the Hagley Museum and Library, Wilmington, Delaware.

57. They are still in place in the basement, covered with sand after the demolition of the building located at 835–837 Mill Street, Montreal. Possibly others could be recovered in this area.

58. James Leffel and Company, New Pamphlet 1894–1895 of the James Leffel Water Wheels, Standard, Special and Samson (Springfield, Ohio: The James Leffel & Co., 1894). What adds to the confusion is that many listings of waterwheel indifferently use the words “Sampson” and “Samson” for this model.

59. Lovell’s province of Quebec directory for 1871 [microform] (Montreal: J. Lovell [1871?]), 56, 59; also found at Library and Archives Canada ; John Abell, Illustrated Catalogue of Agricultural Implements & Machines (Toronto: Globe Printing, 1872), 57–58.

60. Leung, “Grist and Flour,” 291 (see n. 53). We reproduce here the spelling used by Leung.

61. Viger, “Croquis et détails” (see n. 13).

62. Daniel W. Mead, Water Power Engineering (New York: McGraw-Hill, 1920), table 22: 222.

63. See G. T. Bloomfield, “Canadian Fire Insurance Plans and Industrial Archeology,” IA: The Journal of the Society for Industrial Archeology 8, no. 1 (1982): 67–80.

64. Mead, Water Power, 147 (see n. 62).

65. Allan Cureton, “The Lachine Canal” (master’s thesis, McGill University, 1957).

66. Underwriters’ Survey Bureau, Insurance Plan of the City of Montreal, vol. 7, sheet 710, December 1955.

67. Underwriters’ Survey Bureau, Insurance Plan of the City of Montreal, vol. 1, sheet 35, January 1964. Most of the complex is now demolished, but the foundations and the turbines remain underground at 835–837 Mill Street, Montréal.

68. Gelly, “De l’eau,” 273 (see n. 1).