How The Railroads Defeated Winter
THE SEASON NO LONGER MEANS HUNGER, SHIVERING, AND ISOLATION FOR AMERICANS, AND THE MOST IMPORTANT AGENT IN THIS TRANSFORMATION WAS THE RAILROADS
WINTER WAS AN ANNUAL ORDEAL to Americans in the colonial era and early Republic, especially in the North. Food from the fall harvest had to be husbanded carefully and a large supply of firewood laid in to carry families through the cold months. Heavy snow cut communities off for weeks at a time and forced settlers to rely on their own resources. Isolated farms and villages could do little to help those who fell sick from pneumonia, grippe, scurvy, and other winter illnesses. Work, diet, dress, lighting, and social life all were restricted. Transportation was affected too: Bulk shipping by river, canal, and lake had to stop with the annual freeze. Hauling freight by road was possible using sleds drawn by horses or oxen, but it was so slow and cumbersome that few goods traveled far in wintertime.
Nothing did more to change the effects of winter! and reduce its annual impact and danger than the railroad. The American rail network, begun in the 1830s, linked the Atlantic and Pacific Coasts in 1869, and was largely completed by 1900. From the outset railroads were designed and built to overcome seasonal problems and operate throughout the year. They brought distant communities into contact, carried food, fuel, and medicine to formerly isolated places, kept industry supplied with raw materials, and enabled Americans to travel in any season. Before they could help the nation overcome winter, however, the railroads had to find ways to overcome winter themselves.
Like most other things in railroad history, the need to keep running in all weather conditions was determined by economics. Railroad builders were often burdened with heavy interest payments on their loans and could not afford to have their capital-intensive equipment lying idle, frozen, or rusting for several months of every year. To avoid winter shutdowns, railroads had to surmount obstacles at every stage, from planning to construction to operation. In the planning phase they had to map out routes that would not be closed down by winter conditions, or at least not often. During construction they had to lay track and build facilities in the face of extremely hostile weather conditions, especially in the Plains and mountain states. And when a line went into operation, they had to keep it running through the worst that nature could dish out. Finally, they had to teach their widely dispersed employees to use the right blend of discipline and initiative in responding to winter emergencies.
EARLY RAILROADS FACED AN ARRAY OF unfamiliar problems each winter. First and most basic was the expansion and contraction of the rails. In some parts of the country temperatures routinely vary between 100T in summer and -20T in winter. With no provision for expansion, rails would buckle in the heat, causing derailments and requiring laborious repairs. Through trial and error, line engineers learned to lay rails in short sections, bolted together but with expansion gaps of about a quarter-inch between the lengths. They left larger gaps, up to threequarters of an inch, when laying track in below-freezing temperatures. Even so, intense cold and frequent changes of temperature made early iron rails brittle before development of the Bessemer process allowed railroads to replace iron with steel after 1865.
A closely related source of problems was the track bed on which the rails sat. The earliest lines were built at ground level, but operators soon realized that with the annual thaw, or even sustained heavy rains, the weight of trains would press the track down into the soft ground. They learned to elevate the track bed a foot or more above ground level and cover it with crushed rock, earth, sand, or cinders for ballast. This ballast held the track ties in place and ensured a solid foundation and good drainage even at the wettest times of the year.
The first generation of railroaders lacked the money and expertise to build long-span bridges over major rivers and estuaries. Instead they relied on ferries to move passengers, freight, and sometimes whole cars over waterways. However, winter conditions, especially heavy concentrations of floating ice, could stop the ferries for days or weeks on end. The Philadelphia, Wilmington & Baltimore Railroad, for example, had to shut down for six weeks in 1849 because of fast-flowing water choked with ice floes at the mouth of the Susquehanna River. (In years when the river froze solid, passengers could walk across the 3,500-foot-wide ice sheet if they dared.)
Ice jeopardized early bridges too. Their piers, sunk into the riverbeds, were subjected to immense pressures from ice, and from swollen rivers when it melted. The Erie Railroad’s bridge over the Delaware at Port Jervis, New York, for example, was carried away by a fast-flowing icy river in February 1857. A replacement was almost completed when it was destroyed in the same way. In 1875 the Erie lost yet a third bridge in the same place when a natural ice dam, fifty feet high, formed upstream and broke at winter’s end, creating a flood that battered away everything in its path.
The Union Pacific in Nebraska lost nearly ten miles of bridges, track bed, and embankment in 1867 when a latewinter ice jam forced the Platte River out of its normal course and rerouted it across seemingly secure dry land. In Alaska, where every winter railroading problem appeared in extreme form, the Copper River & Northwestern Railroad came to expect an annual washout at its trestle bridge crossing of the Copper River when the snow melted. The company routinely sank new riverbed pilings and rebuilt the trestle each spring. The solution to this problem finally came from stronger steel and improved bridge-building technology, which enabled railroads to throw long single spans across ice-prone rivers rather than rely on trestles or piers on the riverbed.
ICE WAS NOT AL ways a menace, however; the railroads sometimes made it an ally. In the late nineteenth century Canadian and Alaskan railroad companies, assured of long freezes, regularly laid tracks on frozen rivers each winter. Their flat surfaces made ideal track beds so long as the ice stayed thick and hard. In 1852 the Philadelphia, Wilmington & Baltimore Railroad experimentally laid tracks across the frozen Susquehanna, to move freight cars. Locomotives shoved uncoupled cars down tracks on one riverbank, and they rolled part of the way across. Teams of horses pulled them the rest of the way, and a steam winch lifted them along temporary tracks up the other bank.
The PW&B never risked its much heavier locomotives on the ice. But the builders of the transcontinental railroad did take that risk in 1867 to get their first locomotive across the Missouri River. Aware that there was a wagon road on the ice every winter, wrote one eyewitness, “They graded on the Missouri side down to the river’s edge, and up on the Kansas side, got long telegraph poles to use as ties, and had all material ready for a hard freeze. Finally in January the weather became very cold and the ice was frozen 18 inches thick. They laid poles on the ice, spiked the rails on, fired her up, and started. As they approached the ice the engineer jumped off and let the locomotive go across alone. The ice groaned and cracked until we thought it would surely break. But it held, and when it reached the Kansas side another engineer jumped on and ran it up the bank. It was a tremendously exciting quarter hour.”
The first transcontinental railroad was the greatest engineering accomplishment of its era. Starting in 1866, the Union Pacific built westward from Omaha, Nebraska, while the Central Pacific built eastward from Sacramento, California. Each railroad moved as fast as possible for the sake of land grants and federal subsidies, which were paid in return for completed miles. The Central Pacific started slowly because it faced the daunting task of crossing the Sierra Nevada, whose heights are snow-covered almost all year and whose winters are merciless. It was impossible to build the line above ground all the way; deep cuts and tunnels were need too. Primitive hand tools and bitter winter temperatures delayed the tunneling, which took more than two years to complete.
The winter of 1866-67 was particularly severe, with forty-four snowstorms between November and May that dropped more than forty feet of snow on the area. It was still three feet deep at Donner Pass as late as June 1867. One storm alone, between February 18 and 22, dropped six feet of fresh snow onto an already inundated work crew. The 1,000 tunnel builders, most of them Chinese immigrants, lived in a village that was all but buried under the snowpack, and they burrowed through a warren of snow caves to get to their work each day. They suffered exposure, frostbite, and a horrifying succession of accidents and mishaps.
Despite immense winter hardships, Leland Stanford of the Central Pacific was able to hammer the final spike linking his line and the Union Pacific in May 1869 at Promontory Point, Utah. But winter problems on this first transcontinental line did not end there. The Central Pacific’s first season of operation was plagued by heavy snow and an avalanche that swept away a hundred feet of track. Westbound passengers determined to get to Sacramento rode the train as far as the snowslide and then trudged seven miles on foot to Emigrant Pass, often in icy winds, before boarding a second train to take them the rest of the way.
The company realized that it would have to build snow shelters to keep the line open and prevent recurrent avalanche damage. The snowsheds were of several types. Some were A-frames meant to be covered by snow while keeping the track open. Others, built across mountain slopes, were designed to withstand the impact of snowslides and to carry avalanches over the line. Both types were built of immense, sturdy timbers from the nearby redwood forests and were bolted to the mountainside, greatly increasing the company’s cost per mile of construction.
Maintaining these shelters was a big problem. The Aframes were finicky and tended to shift when snow packed on their lee sides melted in warmer weather, necessitating constant maintenance. In summer they would dry out and catch fire from locomotive sparks. Inside, smoke from trains accumulated and sometimes nearly asphyxiated the train crews, who took to wearing World War I-style masks. This problem, along with travelers’ complaints that they could not see the mountain scenery, induced the railway to leave gaps between snowsheds for scenery and fresh air. Even with its tunnels and snowsheds, though, the Central Pacific still required 4,000 workmen along the mountain sections each winter to maintain the shelters and shovel snow off track sections that were left exposed.
When snow did not stop trains altogether, it caused endless operating problems. Snow accumulations on locomotives and cars weighed them down, creating drag and impairing efficiency. Snow could pack down between the rails and lift wheels off the track, derailing the train. Decades of effort went into developing effective snowremoval methods. The earliest and simplest one was to perch a group of workmen on the front of a locomotive, each holding a snow shovel, which they slid along the track to dislodge snow and ice. This primitive process, used by the pioneering Camden & Amboy Railroad of New Jersey in the 183Os, soon yielded to better techniques. Before 1840 the projecting V-shaped “pilot plow” attached to the front of every locomotive was in use on Eastern roads, designed to remove snow from between the rails as well as from on top.
Unfortunately, when snow fell it often came so fast that pilot plows alone could not cope; cuts in particular filled quickly. Then the railroad had to fall back on old-fashioned shoveling. Edgar A. Custer, an engineer, recalled that when he was a young apprentice on the Pennsylvania Railroad in the late 187Os, he and his coworkers were once taken out of their workshops in the middle of a January blizzard and told to keep a deeply cut section of the line over the Alleghenies clear at all costs. They had to “shovel or freeze to death.”
“Every four hours,” Custer continued, “two baggage cars equipped for making coffee and broiling beefsteak were hauled up the mountain. At each stop the men crowded into the warm cars and thawed out while devouring huge beefsteak sandwiches and cups of scalding coffee. The stops were short; there were many men to feed. The bitter wind howled down the mountain gaps, opaque with frozen particles that cut our faces and drifted the tracks almost as fast as we shoveled it away.” A sudden worsening of the storm left his group out overnight, causing them all to suffer from frostbite and exposure. When the company finally got a rescue train to them, it gave the men hot food and carried them to Altoona. Custer never forgot that his total pay for the twenty-eight-hour ordeal was $1.96—seven cents an hour, straight time.
As the railroads matured, track engineers learned how to moderate snowdrifts in cuts by appropriate siting of snow fences and hedges. They also brought improved plows into service, among them the giant Congdon “bucker plow,” which served on Western lines between the 186Os and the 188Os. It was a massive V-shaped blade- as high as the locomotives that pushed it in teams of five, ten, or even fourteen—that was used as a battering ram against snow-blocked cuts on the Great Plains. It was so high that the leading locomotive had to be fitted with an elevated cupola so that its fireman could climb up and describe to his engineer what lay ahead.
Drivers approached snowdrifted cuts warily, knowing that the gradual rise of snow at the entrance, if frozen hard, could derail plow and locomotive alike. A trainload of workmen would dig out the start of the cut, making a broad “face” that the plow could strike without rising off the rails. They would also excavate a series of trenches farther in to weaken the solid mass of snow. Then the train would back up for five miles or more so that it could take a run at the snowbank and try to smash its way through at sixty or seventy miles per hour. It was hazardous work, often done in temperatures far below zero.
A worker who took part in one such operation in North Dakota recorded its climax: “The plow rushes toward the snowdrift with a will. But the cut is deep and narrow, and the snow is hard packed. It is like running into a stone wall. The plow, moving at sixty-five miles per hour, stops with a mighty shock. Snow bursts through the cab windows and comes pouring in like an avalanche. Tons of coal in the tender surge forward, breaking the gate and sweeping against the boiler head. The cab is filled with escaping steam and falling glass. Wildly, you search for some means of escape.
“Again, the plow is dug out. The snow is shoveled out of the cab, boards nailed over the windows, and the engineer, tying a handkerchief around his forehead to stop the flow of blood from cuts made by the broken glass, sounds a retreat. But he backs the engine with a grim vow to ‘put her through this time or break a steam pipe.'” This practice, called “bucking,” worked better on the Plains, where curves were gentle enough to permit high speeds, than on steeply curved mountain sections, where drivers had to go slowly.
Despite these strenuous efforts, winter conditions could still be bad enough to stop the plows. In February and March of 1869, three months before the Promontory Point meeting, a ninety-mile stretch of the Union Pacific in Wyoming was overwhelmed by a massive blizzard, stranding several trains and two hundred eastbound passengers at Rawlins. They soon exhausted the meager local food supply. One locomotive exploded in its efforts to batter through the snow, killing its crew; and some of the passengers, desperate to get to Washington for President Grant’s inaugural, resolved to walk the ninety miles in sub-zero conditions to get to Laramie, where most arrived with frostbitten feet.
The invention of the rotary plow, which chewed its way into snow, powdered it, and then threw it aside, marked a great advance over the old digging and bludgeoning methods of snow clearance. Patented by Orange JuIl, a Canadian flour-mill owner, in 1884, it is still in use on American railroads today. The Cooke Locomotive and Machine Works, of Paterson, New Jersey, bought the patent and began to deliver rotary plows to Northern U.S. and Canadian railroads in 1888. Their main advantages over wedge plows were that they could overcome greater depths of accumulated snow without getting stuck, they worked well at low speeds, and they were less prone to derailment. Their big disadvantage was that the high-speed rotary mechanism could be damaged by objects other than snow. Hence they were of limited use in clearing lines smothered by mountain avalanches, where the snow could be mixed with telegraph poles, wires, rocks, and metal debris.
Without rotary plows, continuous railroad service in harsh-winter districts would never have been possible. On the Copper River line in Alaska, writes the historian William H. Wilson, “When winter winds of up to eighty miles per hour sometimes drove snow deep over the tracks along the delta, a rotary plow with two pusher locomotives would ram into the thick, white, swirling mass while the train dogged behind. … Looking back, the trainmen in the caboose watched the snow obliterate the cut” as soon as they had passed.
RAILROADS PIONEERED MODERN MANAGE -ment methods as well as machines, entrusting each section to a well-trained and effective superintendent. In 1897 E. W. Hadley, a retired division superintendent on the Northern Pacific Railroad in North Dakota and Minnesota, wrote an account of his duties in an area where winter lasted almost half of every year. He was charged by his employers to keep the trains running at all times and to do everything he could to prevent passenger trains, with their valuable human cargoes, from being stranded in the snow—a source of danger and great financial expense. The suddenness of weather changes and the great distances being covered by the transcontinental trains made periodic failures inevitable, so the railroad worked out procedures for protecting stranded passengers. The train crew was ordered to use coal from the locomotive’s tender, if necessary, to keep stoves burning in the coaches, and then to trudge to the nearest farmhouse or town to raise food supplies before organizing relief trains.
On routes close to major cities, superintendents found that the best way to keep the lines open was simply to run trains as often as possible, to prevent the buildup of snow on the tracks. Instead of sending out occasional long freight trains, the most economical method during mild weather, they would adapt to winter by running frequent short trains, often heading them with two locomotives instead of one to provide extra tractive effort in the event of snowdrifts across the track or ice on the line.
Another section superintendent told the journalist Edward Hungerford around the turn of the century that when heavy snow began, “we just fall back on our roundhouses. We cut our local freights down to 1500 tons, then to 1200, 900, 600, rather than send them into shelter … and while we are cutting off cars we are adding power. Everything that goes out of this yard will be double-headed as long as there is danger in the air. There will be two engines to a passenger train and ahead of each a rotary, with two or three locomotives to push her.” In this way the main lines, at least, could usually be kept open. Branch lines, which saw less traffic, were much more likely to become impassable. Another problem with winter operation was that steam locomotives became less efficient as temperatures dropped, more of their energy was diverted into heating their own massive boilers, and they used more fuel.
Winter work was an ordeal for train and track crews. The driver and fireman, who worked in a semi-open cab, and the brakemen, who operated the brake wheels manually, striding back and forth along the swaying roofs of the train, were exposed to freezing conditions for long periods and often suffered from exposure, which dulled their judgment. “The windows are frozen up or covered with snow,” wrote one witness, “and from innumerable cracks and crevices around the floor where it joins the boiler come draughts that bite and sting. The engine caws like a crow—'haugh, haugh,’ now fast, now slow, according as the drifts cover the track or uncover it for a brief space, and when it strikes a drift it throws the snow in blinding clouds all over itself, just as the spray flies over a vessel shipping a sea. The track is rough, for the frost has disturbed it, and the engine lurches ahead, staggering to and fro like a drunken man.”
No wonder owners and crewmen alike were delighted by the development first of electric locomotives just after the turn of the century (these were mostly used in or near major cities) and then of diesel-electrics, the kind of locomotives that have remained the workhorses of American railroads up to the present. These newer models, lacking the open furnaces of their older siblings, were able to have closed and heated cabs for their drivers’ comfort and could work more efficiently in cold weather.
By the turn of the century, the country’s main railroad network was complete, and winter service was becoming highly dependable. Trouble spots remained, however, especially in mountain areas. The Great Northern’s problems in crossing the Cascade Mountains at Stevens Pass, Washington, were particularly severe. The area was subject to avalanches, especially when the heavy snows on the western slopes began to thaw in late winter. In 1910 two passenger trains snowed in by severe storms were unable to advance or retreat for six days. A forest fire the previous summer had laid bare the slope across which the two trains stood, increasing the avalanche hazard, and heavy rain following hard snow began to weaken the snowpack. Workmen dug desperately to clear the blocked line, and as a historian of the event recalls, “Many of the passengers were in a complete state of panic, and several on board had premonitions of impending disaster.” Their premonitions were justified. On March 1 a massive avalanche swept down the mountainside and carried away both trains, killing ninety-six people and destroying 2,000 feet of track and seven locomotives. It was, and remains, the worst avalanche disaster in American history.
It was impossible to clear away the tangle of trees, wagons, track, ties, electrical and telegraph wires, and other debris until the snow finally thawed, more than a month later. The company, facing a public relations disaster, strengthened its snowsheds and tightened up its winter safety precautions, then set about building a long tunnel (more than eight miles) at a lower altitude. When finally opened in 1929, it depended on electric locomotives and featured high-pressure air blowers to ensure that crews and passengers inside would be able to breathe normally. Its opening greatly enhanced the winter reliability and safety of the Great Northern and enabled it to abandon one of the most dangerous and labor-intensive sections of railroad line in all of the United States.
THE PROSAIC SIDE OF RAILROADING, THAT OF accountancy, also shows the importance of keeping going through the winter. Track, locomotives, cars, fuel, labor costs, and a thousand other expenses made railroading the most highly capitalized business in American history. Most railroads were heavily in debt by the time they started operating and had to make high interest payments in addition to covering daily operating costs. The owners knew that their ability to run through the winter, when most other transport stopped, presented an opportunity to recoup some of their costs.
In the Great Lakes area, for example, railroads were at a disadvantage in competition with lake steamers, which did not have to pay for the upkeep of their “road” and which could carry greater bulk loads than trains, though more slowly. But shipping on the Great Lakes and the Erie Canal had to stop soon after December 1 each year, and the routes stayed frozen until the following April. Freight that would have been waterborne at other times of the year had to be diverted onto the railroads, which responded to this temporary monopoly by charging higher rates. While they could not match ships in low-cost bulk capacity, they could get through to their destinations on all but a handful of the snowiest days. Industrial customers grumbled, but in the end they paid the higher rates. Without winter rail service, iron mines in Wisconsin and Minnesota would have had to shut down or maintain huge ore dumps, and steel mills in Ohio and Pennsylvania would have had to stockpile several months’ worth of raw materials.
Winter freight rates from Chicago to New York, the most important trade route, were often one and a half times the summer lake and canal rates in the second half of the nineteenth century. Road owners said that the higher rates reflected higher operating costs in winter, but as the historian John H. White, Jr., points out, “These were actually relatively minor costs compared with the railroad’s interest payments.” He speculates that “the reason for the stepped-up winter rates was to maintain revenue when traffic slumped after the fall harvest rush was over.” When the thaw came, rail charges fell sharply, to a rate only slightly higher than that of the steamships.
In general the railroads, then as today, preferred carrying freight to passengers, because freight required less careful handling, generated more revenue per mile, and was not as vulnerable in an emergency. Here again they found ways to take advantage of winter conditions by manipulating the winter freeze. Chicago meat-packers cut thousands of tons of ice from Wisconsin lakes and stored it in insulated warehouses. In summer they loaded the ice onto freight cars to refrigerate them and prevent meat from spoiling en route to East Coast cities. (They originally tried Chicago River ice, but it proved to be too polluted.) Fruit and vegetable shippers from California and Florida also adopted the idea, creating refrigerated trains with imported Northern ice thirty or forty years before the invention of electric refrigeration.
Winter passenger service presented particular difficulties because every passenger car had to be heated. With all six sides exposed to the cold exterior air, poorly insulated and with loose-fitting doors and windows that were opened frequently by brakemen and the conductor, an unheated car rushing through freezing air tended to be only a few degrees warmer inside than out. The most obvious heating method was to supply passenger cars with wood- or coal-burning stoves. This was the practice on most roads from 1840 onwards: A group of seats near the center of the car would be removed in the fall and a stove bolted to the floor in their place. Passengers discovered that those sitting near the stove, which was stoked regularly by the brakeman, would be roasting hot, while those remote from it suffered drafts and icy cold. Experienced travelers sought out places halfway between the stove and the doors.
Charles Dickens, touring America in 1842, reported that “in the centre of the carriage there is usually a stove, fed with charcoal or anthracite coal; which is for the most part red-hot. It is insufferably close; and you see the hot air fluttering between yourself and any other object you may happen to look at, like the ghost of smoke.” The lurching and jolting of early trains sometimes threw passengers against the stove and burned them. Others had clothes and baggage scorched by glowing embers thrown from the fire. Worst of all, stoves were a leading cause of fire when there was a collision. Overturned stoves often set light to the splintered wooden wreckage, incinerating passengers who had survived the initial impact. To quote White again: “The combination of stoves and wooden cars was unquestionably dangerous. Tinder-dry wood heavily coated with varnish, oil cloth head linings, upholstered seats, lamp reservoirs sloshing with kerosene—it was a rich flammable mixture waiting only for the stove to topple over and spew out its cherry-red coals.”
Throughout the nineteenth century dozens of inventors searched for ways to reduce the fire hazard. They tried putting the stove underneath the car rather than inside, or setting it in a tank of water, or placing it on a fire-resistant zinc base. None of these methods proved satisfactory, and by 1880 fires in passenger trains were a public scandal. The American Railroad Journal announced the end of winter in 1882 with the remark, “The season for railroad barbecues is over, the annual sacrifice of baked and roasted victims has been offered up.” Between 1887 and 1890 ninety cars burned in stove accidents, claiming twenty victims. As late as December 1921 an accident on a branch of the Reading Railroad near Philadelphia led to twenty-seven deaths in wreckage ignited by a coal stove.
As a result, in the 188Os, railroads began experimenting with steam heating. Steam, taken from the locomotive itself or from an auxiliary boiler, was distributed through metal pipes and radiators along the train, piped between cars through coupling joints. The method had its drawbacks: Taking steam from the locomotive decreased the engine’s efficiency, which was already impaired in cold weather. Low-pressure steam systems kept the occupants of the first car toasty warm while those in the back cars froze. But high-pressure steam systems created a hazard potentially as bad as the old stoves, since ruptured highpressure steam hoses in an accident were also likely to be lethal. Early thermostats proved unreliable, and managing comfortable heating remained a challenge to the railroads into the twentieth century, until Egbert Gold’s “vapor system,” patented in 1904, circumvented the old problems and permitted safe and effective heating throughout the train with a series of pressure-regulating valves.
IN THE FIRST FOUR DECADES OF THE TWENTIETH century, despite residual problems, railroad operations continued to improve in safety and reliability, whatever the weather. When the sociologists Robert and Helen Lynd researched their study of a typical Midwestern city, “Middletown” (Muncie, Indiana), in the 1920s, they found dozens of ways in which railroads had transformed the city. Standardized goods and clothes made in all parts of the country were now advertised in Middletown and shipped in by rail. Older residents who remembered growing and canning their own food, weaving their own cloth, making and mending their own wagons and tack, and building their own furniture were now parts of a nationwide market in specialized goods.
The transformation of their winter diet was particularly striking. In the late nineteenth century, said the oldtimers, everyone ate meat three times a day, along with potatoes, turnips, cabbage, and masses of pickles. “We never thought of having fresh fruit or green vegetables and could not have got them if we had,” with the result that by springtime “nearly everybody used to be sick because of lack of green stuff to eat.” Newspapers carried advertisements for sarsaparilla “to cure boils, sluggishness, thick blood, and other ailments resulting from heavy winter food.” But by the 1920s fresh produce from Florida and Southern California, grown throughout the winter months and shipped in refrigerated trains, had transformed Middletown’s diet, making it uniform throughout the year and driving “spring sickness” into decline. By then the railroads in their turn were under threat from road transport. The car and highway builders borrowed from the railroads the once-revolutionary promise that they could operate whatever the season, and so continued the work the railroads had begun, of reducing the hardships that had once been an inevitable part of winter.