Category: Adams Basin

Finally, the part I’ve long waited for (but with more than a little apprehension): Adding human figures to the Erie Canal scene. The setting is rural upstate New York, and the time is early morning, so we won’t need many. But the view would seem lifeless without a few human inhabitants.

I have little experience in creating human figures, but it is a skill that I’ve long wanted to learn. Now would be a good time.

And the first lesson: This is a complex, labor-intensive process. I tip my hat to all the modelers, riggers, and animators who work hard every day to bring 3D film characters to life. My plans are more modest (for one thing, they don’t involve animation). I simply want to make a collection of quality human figures that can be posed and used in the lift bridge scene and other scenes later on.

Of course, 3D human models of varying degrees of quality are available – for free or for sale – from many online sources. But what’s the fun in that? I much prefer working with assets that I create myself.

My process using involves four applications: Silo for creating the mesh, Maya for rigging and posing, Mudbox for sculpting and shading, and Terragen for rendering.

Here we go.

As with most things, we begin with a box to create a model of a man’s head.

The subdivided box is shaped and cut to rough in the skull, chin and positions for the ear and eyes.

More edges are added and details – nose, lips, eyes – begin to take shape. Edges are routed to follow the natural contours of the face.

Several iterations later, we have a man’s head.

Other body parts are also modeled separately. These can be adapted and reused as needed.

The torso, arms and legs are modeled, and then all of the pieces are put together.

Now we have a basic mesh that can also serve as a base for additional models later.

In Maya, the mesh is UV mapped and hair and eyebrows are added. A skeleton is built and bound to the mesh, and a control rig allows it to be moved into different poses like a puppet.

Binding the mesh to the skeleton is by far the most difficult step. Maya calculates the initial bind but it then must be fine tuned by hand – a time-consuming process that requires some practice and a lot of patience. But it’s the only way to make sure the character will move naturally.

Clothing is added in Maya and the figure is posed. Folds and wrinkles in the clothing are added in Mudbox.

One down, several more to go . . .

Almost all of the man-made objects and foliage are done, so it’s time to start on the final, most challenging part of the scene. Modeling organic shapes is not something I have a lot of experience at. Time to address that.

First up: The draft animals needed for our two-horsepower farm wagon.

We turn to Silo, my favorite subdivision modeling app and, beginning with a cube, begin roughing in the basic shape.

The reference images are from Dover’s An Atlas of Animal Anatomy for Artists, which includes fine horse drawings, including drawings of the bone and muscle structure, in each of the three dimensions. (The book also includes drawings of other animals.)

A few iterations later we’ve added a head and subdivided the mesh, reshaping the polygons and edges to follow the main muscle groups.

More edges are added and redirected where necessary. Ears, eyes, nostrils, mouth and hooves are finished up. The mesh is then saved out so it can be reopened in Maya, where the mane and tail will be added.

The horse collar is based on historical photographs and also a nice set of images from eBay, where I found a vintage 1900s collar for sale. In the process of modeling it and the rest of the harness I learned some new terminology and a little bit about how everything worked.

As with most things, there’s a lot more to this than meets the eye.

The model is shaded twice to create a pair of draft horses that are moved into position next to the tongue of the farm wagon. The rest of the harness is then modeled to connect all the pieces.

The Adams Basin scene is shaping up bit by bit, much of it unplanned. But if there is a theme, it might be “transportation.”

The juxtaposition of the Erie Canal, lift bridge, steam-powered and towed canal boats, and an interurban passenger car isn’t accidental. The original canal followed a natural corridor, and once the route was pioneered other modes of transportation soon followed.

Thousands of businesses also sprung up along the canal. In Adams Basin one of these was the E. F. Blackford Co., a fruit wholesaler that built a complex of warehouses and a railroad siding that allowed it to ship its produce on the New York Central Railroad, which paralleled the canal. (All of these structures, including the railroad, have long since disappeared.)

The warehouses are too far from the bridge to be included in the scene, but in a nod to the business, we’ll include a farm wagon on its way to deliver a load of produce.

Back in the day, the ubiquitous box farm wagon was manufactured by various suppliers, and nearly all followed the same basic pattern. We’ll base ours on the wagon built by the Studebaker Brothers Manufacturing Company in South Bend, Indiana. (Later it became the Studebaker Corporation and manufactured automobiles until the early 1960s.)

As a reference I’m using a fine set of plans of a Studebaker Wagon drawn by Bob Kuntz. The plans diagram the wagon’s running gear or wheel assemblies. The box and seat are based on historical photos.

The wagon, lightweight and rugged, featured dished wheels (to increase strength) that were cambered (tilted out to keep the bottom rung perpendicular to the ground). The rudimentary brake system was optional. Most wagons were shipped without brakes, and drivers counted on their skill and the ability of their teams to control the vehicle on grades.

It’s interesting how the lives of rural Americans had barely changed despite a century of innovation. Even as canals and railroads were built, steam harnessed to power boats and locomotives, and electricity generated to lift steel bridges and push interurban cars, the average farmer still depended on horse power and centuries-old technology to get his produce to market.

Besides the farm wagon I’ve also built a dray, a heavy-duty truck based on model plans from the Vilas County Lumber Company.

Now we just need horse teams, drivers, and produce for the wagons to haul.

The end of the animal-powered canal boat, after decades of encroachment by steam vessels, came quickly.

Primitive steamboats had been sighted on the canal since the 1840s but  would remain a novelty for most of the century. Canal enlargement in the mid-19th century and improved steamer designs gradually changed things, though, and around 1880 the first practical steamers were introduced. The advantage of steam power was immediately apparent. While animal-powered fleets were limited to two vessels – so-called doubleheaders – a single steamer could push one boat (a consort) and tow up to five or six more.

The final Barge Canal enlargement of 1908–1918 permanently settled things. As each new section came online the towpath itself began to disappear. When the enlargement was completed, only powered boats could navigate the entire distance between Buffalo and Albany.

In 1916, the date of my scene, most steam-powered boats on the canal were of two types, both built of wood:  tugboats and steam canal boats. The U.S. government’s 1884 Report on the Ship-Building Industry of the United States helpfully includes a diagram of a steam canal boat, a functional design (besides pushing and towing unpowered boats, it could carry its own full cargo) that remained in service well into the 20th century.

This was the design I wanted for my 3D model – new technology (steam power) grafted onto traditional design (the wooden laker hull). Now I just needed to find a specific vessel on which to base the model.

Unfortunately, period photographs of Erie Canal steamers are about as rare as hen’s teeth. Repeated online searches turned up nothing.

Eventually I learned to combine two sources: the List of Merchant Vessels of the United States, published annually by the Department of Commerce, and the Great Lakes Vessels Online Index, maintained by Bowling Green State University. The 1916 edition of the List included around 40 vessels designated as “Scb,” or steam canal boat. Searching for each of these on the Bowling Green site resulted in a handful of matches – some of which included photographs.

For various reasons, including the clarity of the photo and strong evidence that the boat operated in this area, I chose a steamer named the Ashford. It is pictured at the top of this post.

According to the Index, the Ashford was 95 feet long and 17 feet 8 inches wide. It was built in 1883 in Buffalo and sank in the Barge Canal near Schenectady in November 1919 – “a total loss.”

Filling in the blanks

Online searches (mostly through Google Books) turned up several references to the Ashford. The earliest was a notice from the Buffalo Express reprinted in the August 22, 1883 issue of The Coal Trade Journal:

The Rochester and Pittsburgh Railroad Company was founded in 1881 to haul coal from western Pennsylvania to Rochester and Buffalo. That was as far as its lines extended. The canal boats were to be used to ship the coal to the lucrative New York City market. The company’s 1883 annual report explains that land purchased in Buffalo will be developed into a railyard, but also that

. . . We have made two slips to accommodate eight canal boats . . . and have constructed coal shutes [sic] . . . so that coal can be shipped in canal boats at minimum cost.

The assumption that the Erie Canal was always in competition with the railroads, then, is an oversimplification. Things were more complicated, especially early on when hundreds of railways were being formed and struggling to grow and survive. In this case the R&P’s fledgling rail barons were merely attempting to leverage an existing resource – the canal – to give them a leg up on the competition – other railroads.

Unfortunately, things didn’t work out as planned. The Rochester and Pittsburgh went bankrupt and in 1887 all of its assets were sold. The rolling stock and track were converted into the Buffalo, Rochester and Pittsburgh Railroad, which would be absorbed into the Baltimore and Ohio Railroad in 1932.

The company’s steamers apparently went their separate ways. The next mention of the Ashford that I can find is in the 1899 edition of The Blue Book of American Shipping, which gives the owner as one William Barker of Buffalo. The following year the Ashford is listed as one of two steamers owned by William J. Warwick, also of Buffalo. Ten years later the Ashford would be sold to Joseph A. Hutton of Rondout, on the Hudson River. Hutton may have been the final owner – I could find no others listed – and since this was the only steamer he owned, perhaps he was the skipper as well. Was he on board in November 1919 when the Ashford sank near Schenectady? It would be nice to know, but here the trail grows cold.

Here’s an interesting postscript:

In September, 1889, the Bishop, a canal boat towed by the Ashford, collided with the F.  W. Whalen, towed by the steamer Lizzie Crandal, while the two fleets passed each other a few miles west of the location pictured in my scene. Both the Bishop and the Whalen went down, and the owners of the Whalen sued the owners of the Ashford and the Bishop for damages.

The following year the District Court of the Southern District of New York dismissed the suit based on evidence that the Whalen had been on the wrong side of the canal at the time of the collision.

Rebuilding the Ashford

My sources for the model are a high-resolution photo of the Ashford (provided by Bowling Green University), and Robert E. Hagar’s excellent drawings of an 1870s-era laker, held by the Chittenango Landing Canal Boat Museum.

The photograph isn’t the best quality, unfortunately, and some detail (especially behind the wheelhouse) is impossible to figure out. As usual, my rule is to leave out details rather than make them up. (The model will be seen at a distance and the stern won’t be prominent, anyway.)

Once it’s finished the model is placed in the scene along with the three lakers. Our mighty fleet is at last under way.

Our Erie Canal laker model, shunted into digital dry dock last spring, will be expanded into a small fleet for our scene.

Of all the types of boats on the Erie Canal in 1916, the laker was the most common. Unpowered (or, in the jargon of the time, “unrigged”), it was the workhorse of the Erie Canal from the late 19th to the early 20th century. According to a Report on the Ship-Building Industry of the United States, published in 1884 by the U.S. Department of the Interior:

The laker is a regularly framed boat, with perfectly flat bottom, square bilge, perpendicular sides, straight body, round bow and stern, and decked entirely over. It is made of oak and white pine . . . and it is 97 or 98 feet long, 17½ or 17⅔ beam, and 10 or 10½ feet in depth of side. . . .

These boats . . . weigh from 65 to 72 tons, and carry 225 to 240 tons of cargo. They are usually painted white, and the stern is often profusely ornamented by a sign-painter, the name and home port being conspicuously painted in gilt and red and blue letters. About one-half of the boats on the Erie Canal are of this type. Being good for almost any service, these boats . . . are popularly used in forwarding grain from Buffalo and Oswego to New York city [sic] via the canal and the Hudson River.

In 1883 a newly built laker cost an average of $3,800, roughly equivalent to $95,000 today. It would see service for 10 to 15 years, perhaps longer if it were well maintained.

For much of this period most canal boats were animal-powered. A team of three mules, walking along the canal’s towpath, could pull two fully loaded lakers lashed together. These doubleheaders would allow the owner to maximize his or her carrying capacity (and profit) with a minimum crew.

That crew, which often included family members, lived in the aft cabins. The animals were housed in the bow cabin or bowstable. Two teams of three mules were switched every six hours and kept the boats moving 24 hours a day. The speed limit was 4 miles per hour – a brisk walking pace – but most boats averaged less, covering the 360 miles between Buffalo and Albany in five to seven days.

From 1880, steam-powered boats began to replace animal power. By 1916 – the date of our scene – the transition was virtually complete. Steel barges, another innovation, had made an appearance but their adoption would be interrupted by World War I, when all steel production would be diverted to the war effort.

So for the time being wooden lakers would be pushed and towed by wooden steamers. (There will be more about steamers in the next post.)

When they reached the end of their useful lives, wooden canal boats were often simply abandoned by their owners at “graveyards” where the canal was wide enough to accommodate their decaying wrecks.

One old graveyard, in Pittsford, New York, can still be seen on Google Maps, where the ghostly outlines of 100-year-old hulks are visible just below the water’s surface or jutting above it along the shore.

The sad, old frames are fully exposed when the canal is drained for winter. It’s amazing that they have survived for so long.

Now that the terrain and major human-built objects are in place, it’s time to work on the scene’s natural elements: grass, wildflowers, weeds, and trees.

But before I could start modeling, I had to familiarize myself with the species that are native to this area – and would have been a hundred years ago. The National Wildlife Federation Field Guide to Trees of North America and the National Wildlife Federation Field Guide to Wildflowers of North America became my constant companions this fall. Both are densely packed with photos, maps and detailed descriptions of hundreds of species. Great for browsing at home as well as using in the field.

Sometimes it was possible to identify trees in historical photos. I settled on a mix of northern red oak, black walnut, and sugar maple. Some trees that are common here now, such as black locust and honey locust, were omitted because they are not native and I was not able to find out when they would have been introduced.

I collected and scanned leaves for identification and to use for modeling. Scanned leaves can be processed to create the different textures needed for 3D model construction.

All of the foliage was created in SpeedTree, an industrial-strength tree and foliage 3D modeling tool that’s widely used in the motion picture and gaming industries. SpeedTree is one of those applications that does only one thing, but it does it very, very well. It has a short (but steep) learning curve. But once you begin to master it you can build just about anything.

I used SpeedTree to make several models of each tree species, in various stages of growth. I also made about a half-dozen types of wildflowers, all of which should have been present in late September-early October 1916.

As a test, many of the species were mixed and placed in a new Terragen scene. Terragen shines when it comes to foliage because of its ability to place populations of thousands of models, its sophisticated noise and fractal functions that can be used to create natural-looking species distributions, and because of its renderer. This scene consists of billions of polygons, yet Terragen produced a rendering in a couple of hours without breaking a sweat. Impressive indeed.

Besides foliage, I’ve also started building an assortment of small objects that will add detail to the scene. Here’s a selection of barrels that will be placed around the terminal area, on horse-drawn vehicles, and on the canal barges.

Some care has to be taken in placing the trees in the working scene. It’s easy to overdo it. A hundred years ago this area would have been nearly clearcut to make space for farming and canal construction. In the decades since, unchecked second-growth woods have filled in much of the area depicted in our scene.

The new scene rendering shows the foliage in place with a scattering of cirrus clouds in the sky. It’s beginning to look like a picture.

The past few weeks have been spent modeling and placing the buildings in the scene, just as they would have appeared in 1916. Enough historical photos of the area survive to allow us to do this with some accuracy.

The largest building is the hotel, known then as the Ryan House, located on the north side of the canal. The front section of the building dates from 1815, prior to the construction of the original Erie Canal. The back half was a separate home, moved into position and attached in 1850.

To make room for the canal enlargement in 1912, the entire structure was moved again – on rollers, 50 yards due north, where it remains today as the Adams Basin Inn, a local bed and breakfast.

The second group of buildings is on the south side of the canal. The main building served as the post office and telegraph office as well as, apparently, the telephone office for Adams Basin. A square, rather nondescript frame building, it was also moved to make room for the canal enlargement and turned 90 degrees to face Washington Street, the road that the bridge carries over the canal.

The purpose of the small shed behind the post office is a puzzle. It shows up in every historical photo of this location that I’ve seen. All of those images date from the period of the canal enlargement and bridge construction, so there may be some connection. A tool shed, perhaps, or a field kitchen to feed hungry workers? The chimney vent and side windows, which open vertically like the panels on a street vendor’s cart, suggest the latter. By 1916 it that service would no longer be needed, but the shed will remain in our scene.

The final building was the most difficult to track down. The Buffalo, Lockport & Rochester railway station is clearly marked on the 1924 plat map of the area. But no trace of it survives. No photographs, either. At least, so it seemed.

But after several months of searching I came across a reference, via Google, to a small box of photos housed in the rare books section of a local library. The photos documented the enlargement of this section of the canal from 1908 through 1912. A visit to the library and perusal of the box turned up three images that included – yes – the interurban railway station.

The photos confirmed the station’s location. The design was similar to those used in other small communities, such as Spencerport and Gasport, though this one was smaller than those.

In placing the buildings I discovered that the interurban railway track was located in the wrong place. It was moved a few meters south and “regraded” to fix the road crossing. Other changes to the base terrain followed so that now, overall, it’s more accurate than before.

Besides the buildings, I’ve also placed the utility poles that carried the electrical power lines for the railway, the telegraph lines for the steam road, and telegraph and telephone lines along Washington Street.

It seems our ancestors were getting used to living beneath an expanding web of wires that, today, we take for granted and rarely notice. I’ve often wondered what they must have thought.