GE Trans v1.43 serial key or number

COVID shows an increased number of cases and a greater risk of severe disease with increasing age^5,6, a feature shared with the SARS epidemics⁷. This age gradient in reported cases, which has been observed from the earliest stages of the pandemic¹, could result from children having decreased susceptibility to infection, a lower probability of showing disease on infection or a combination of both, compared with adults. Understanding the role of age in transmission and disease severity is critical for determining the likely impact of social-distancing interventions on SARS-CoV-2 transmission⁸, especially those aimed at schools, and for estimating the expected global disease burden.

Here, we disentangle the relative contributions of three potential drivers of the observed distribution of clinical cases by age. We present a summary of the main findings, limitations and implications of this work in Table 1.

First, age-varying susceptibility to infection by SARS-CoV-2, where children are less susceptible than adults to becoming infected on contact with an infectious person, would reduce cases among children. Decreased susceptibility could result from immune cross-protection from other coronaviruses^9,10,11, or from non-specific protection resulting from recent infection by other respiratory viruses¹², which children experience more frequently than adults^13,14. Direct evidence for decreased susceptibility to SARS-CoV-2 in children has been mixed^15,16, but if true could result in lower transmission in the population overall.

Second, children could experience mild or no symptoms on infection more frequently than adults. Clinical cases result from infections that cause noticeable symptoms, such that the person may seek clinical care. An infection that does not result in a clinical case may be truly asymptomatic, or may be paucisymptomatic—that is, resulting in mild symptoms that may not be noticed or reported even though they occur. We refer to both asymptomatic and paucisymptomatic infections as ‘subclinical infections’—which are more likely to remain undetected than clinical cases—and refer to the age-specific proportion of infections resulting in clinical symptoms as the ‘clinical fraction’. Age-dependent variation in severity has been observed for other respiratory virus infections¹⁷, including SARS^17,18. For COVID, there are strong indications of age dependence in severity^5,19 and mortality^18,19 among those cases that are reported, which could extend more generally to age-dependent severity and likelihood of clinically reportable symptoms upon infection. If infected children are less likely to show clinical symptoms, then the number of cases reported among children would be lower, but children with subclinical symptoms could still be capable of transmitting the virus to others, potentially at lower rates than fully symptomatic individuals, as has been shown for influenza²⁰.

Third, differences in contact patterns among individuals of different ages, and setting-specific differences in age distribution, themselves affect the expected number of cases in each age group. Children tend to make more social contacts than adults²¹ and hence, all else being equal, should contribute more to transmission than adults^22,23. If the number of infections or cases depends strongly on the role of children, countries with different age distributions could exhibit substantially different epidemic profiles and overall impact of COVID epidemics.

The higher contact rates in children are why school closures are considered a key intervention for epidemics of respiratory infections²², but the impact of school closure depends on the role of children in transmission. The particular context of SARS-CoV-2 in Wuhan, China, could have resulted in a skewed age distribution because early cases were concentrated in adults over 40 years of age²⁴, and assortative mixing between adults could have reduced transmission to children in the very early stages of the outbreak. Outside China, COVID outbreaks may have been initially seeded by working-age travelers entering the country^25,26, producing a similar excess of adults in early phases of local epidemics. In both cases, the school closures that occurred subsequently potentially further decreased transmission among children, but to what degree is unclear.

We developed an age-stratified transmission model with heterogeneous contact rates between age groups (Fig. 1a), and fitted three variants of this model to the COVID epidemic in Wuhan: one in which susceptibility to infection varies by age, one in which clinical fraction varies by age and one with no age-dependent variation in either susceptibility or clinical fraction (Fig. 1b,c and see Methods). We fitted to two data sources from the Wuhan epidemic: a time series of reported cases¹ and four snapshots of the age distribution of cases^1,27 (Fig. 1 and Extended Data Fig. 1). We assumed that initial cases were in adults, and accounted for school closures in the model by decreasing the school contacts of children starting on 12 January , when schools were closed for the Lunar New Year holiday. We also estimated the effect of the Lunar New Year holiday period on non-school contact rates from 12 January to 22 January , as well as the impact on transmission of travel and movement restrictions in Wuhan, which came into effect on 23 January (Fig. 1d). We found that, under each hypothesis, the basic reproduction number R₀ was initially –, was inflated – fold during the pre-Lunar New Year holiday period and then fell by 60–70% during restrictions in Wuhan (Fig. 1e).

All model variants fitted the daily incident number of confirmed cases equally well (Fig. 1f), but the model without age-varying susceptibility or clinical fraction could not reproduce the observed age distribution of cases. In this model, the number of cases in children was overestimated and cases in older adults were underestimated (Fig. 1g), suggesting that initial seeding among older individuals, together with the impact of school closures, did not explain the lack of observed cases among children. The other two model variants showed an improved fit to the observed age distribution of cases; both models suggested that 20% of all infections occurred in those aged over 70 years. However, the model that assumed no age variation in the clinical fraction implied that a large proportion (50%) of infections among the elderly would be mild or asymptomatic, compared with less than 25% when clinical fraction varied with age (Fig. 1h). Age-dependent severity has been demonstrated in hospitalized confirmed cases^16,28, which suggests that subclinical infection in individuals aged over 70 years is probably rare and supports that the clinical fraction increases with age. Comparison using the deviance information criterion⁶ (DIC) showed that the age-varying susceptibility (DIC, ) and age-varying clinical fraction (DIC, ) model variants were preferred over the model with neither (DIC, ).

Both age-varying susceptibility and age-varying clinical fraction could contribute in part to the observed age patterns. There is evidence for both age-varying susceptibility to SARS-CoV-2 infection¹⁵ and age-varying severity^9,18,19 in COVID cases. A fourth model variant in which both susceptibility and clinical fraction vary by age was able to reproduce the epidemic in Wuhan, and was statistically preferred to any other model variant (DIC, ; Extended Data Fig. 2). However, because decreased susceptibility and decreased clinical fraction have a similar effect on the age distribution of cases, it is necessary to use additional sources of data to disentangle the relative contribution of each to the observed patterns.

We used age-specific case data from 32 settings in six countries (China^1,29, Japan^30,31, Italy³², Singapore²⁵, Canada³³ and South Korea²⁶) and data from six studies giving estimates of infection rates and symptom severity across ages^{16,19,34,35,36,37}, to simultaneously estimate susceptibility and clinical fraction by age (Fig. 2a and Extended Data Fig. 3). We fitted the stationary distribution of the next-generation matrix to these data sources, using setting-specific demographics, with measured contact matrices where possible and synthetic contact matrices otherwise (see Methods)³⁸. The age-dependent clinical fraction was markedly lower in younger age groups in all regions (Fig. 2b), with 21% (12–31%) of infections in those aged 10 to 19 years resulting in clinical cases, which increased to 69% (57–82%) in adults aged over 70 years in the consensus age distribution estimated across all regions. The age-specific susceptibility profile suggested that those aged under 20 years were half as susceptible to SARS-CoV-2 infection as those aged over 20 years (Extended Data Fig. 4). Specifically, relative susceptibility to infection was (–) in those aged 0 to 9 years, compared with (–) in those aged 60 to 69 years.

To determine whether this consensus age-specific profile of susceptibility and clinical fraction for COVID was capable of reproducing epidemic dynamics, we fitted our dynamic model to the incidence of clinical cases in Beijing, Shanghai, South Korea and Italy (Fig. 2c and Extended Data Fig. 5). The consensus age-specific susceptibility and clinical fraction were largely capable of reproducing the age distribution of cases, although there are some outliers, for example in the to year-old age group in South Korea. This could, however, be the result of clustered transmission within a church group in this country⁴. The predicted age distribution of cases for Italy is also less skewed toward adults, especially those over 70 years, than reported cases show, suggesting potential differences in age-specific testing in Italy³⁹. Locally estimated age-varying susceptibility and clinical fraction captured these patterns more precisely (Fig. 2c).

School closures during epidemics^40,41 and pandemics^42,43 aim to decrease transmission among children²² and might also have whole-population effects if children are major contributors to community transmission rates. The effect of school closures will depend on the fraction of the population that are children, the contacts they have with other age groups, their susceptibility to infection and their infectiousness if infected. Using schematic values for pandemic influenza⁴⁴ and our inferred values for COVID (Fig. 3a), we simulated epidemics in three cities with very different demographics: Milan, Italy (median age of 43 years), Birmingham, UK (median age of 30 years) and Bulawayo, Zimbabwe (median age of 15 years) (Fig. 3b), using measured contact matrices for each country. There were many more clinical cases for COVID than influenza in all cities (mean clinical case rate across the three cities: per 1, for COVID versus 23 per 1, for influenza), with more cases occurring in unders (67%) in the influenza-like scenario compared with COVID (17%) (Fig. 3c). More clinical cases were in adults aged over 20 years in Milan compared with the other cities, with a markedly younger age distribution of cases in the simulated epidemic in Bulawayo.

To explore the effect of school closure, we simulated three months of school closures with varying infectiousness of subclinical infections, at either 0%, 50% or % the infectiousness of clinical cases (Fig. 3d). For influenza-like infections we found that school closures decreased the peak incidence by 17–35% across settings, and delayed the peak by 10–89 days across settings (Fig. 3e). For COVID epidemics, the delay and decrease of the peak was smaller (10–19% decrease in peak incidence, 1–6-day delay in peak timing), reflecting findings that school closures in response to SARS-CoV-1 did not have a substantial effect on SARS cases⁴⁵. Among the three cities analyzed here, school closures had the least impact in Bulawayo, which has both the youngest population and the fewest contacts in school relative to the other cities (19% of contacts for 0- to year-olds occurring in school, compared with 39% in Birmingham and 48% in Milan). This pattern could be generalizable to other low-income settings. Because children have lower susceptibility and exhibit more mildly symptomatic cases for COVID, school closures were slightly more effective at reducing transmission of COVID when the infectiousness of subclinical infections was assumed to be high. School closures reduced median peak cases by 8–17% for 0% infectiousness, by 10–20% for 50% infectiousness and by 11–21% for % infectiousness of subclinical infections across each of the settings (Fig. 3f).

Age dependence in susceptibility and clinical fraction has implications for the projected global burden of COVID We simulated COVID epidemics in capital cities and found that the total expected number of clinical cases in an unmitigated epidemic varied between cities depending on the median age of the population, which is a proxy for the age structure of the population (Fig. 4). There were more clinical cases per capita projected in cities with older populations (Fig. 4a), and more subclinical infections projected in cities with younger populations (Fig. 4b). However, the mean estimated basic reproduction number, R₀, did not substantially differ by median age (Fig. 4c), because, across cities, the lower susceptibility and clinical fraction in children relative to adults was counteracted by greater contact rates among children relative to adults. Our finding that cities with younger populations are expected to show fewer cases than cities with older populations depends on all cities having the same age-dependent clinical fraction. However, the relationship between age and clinical symptoms could differ across settings because of a different distribution of comorbidities⁴⁶ or setting-specific comorbidities (such as human immunodeficiency virus (HIV)⁴⁷), for example. If children in low-income and lower–middle-income countries tend to show a higher clinical fraction than children in higher-income countries, then there could be higher numbers of clinical cases in these cities (Extended Data Fig. 6).

The expected age distribution of cases shifted substantially during the simulated epidemics. In the early phase there were more cases in the central age group (20–59 years) and after the peak a higher proportion of cases in those younger than 20 years and those older than 60 years (Fig. 4d). The magnitude of the shift was higher in those countries with a higher median age, which affects projections for likely healthcare burdens at different phases of the epidemic (Fig. 4e), particularly because older individuals, such as those over 60 years, tend to have high healthcare utilization if infected¹.

We have shown age dependence in susceptibility to infection and in the probability of having clinically symptomatic presentation of COVID, from ~20% in children to ~70% in older adults. For a number of other pathogens, there is evidence that children (except for the very youngest, 0–4 years of age) have lower rates of symptomatic disease¹² and mortality²⁶, so the variable age-specific clinical fraction for COVID we find here is consistent with other studies⁴⁸. We have quantified the age-specific susceptibility from available data, and other study types will be needed to build the evidence base for the role of children, including serological surveys and close follow-up of those in infected households.

The age-specific distribution of clinical infection we have found is similar in shape (but larger in scale) to that generally assumed for pandemic influenza, but the age-specific susceptibility is inverted. These differences have a large effect on how effective school closures could be in limiting transmission, delaying the peak of expected cases and decreasing the total and peak numbers of cases. For COVID, school closures are likely to be much less effective than for influenza-like infections.

It is critical to determine how infectious subclinical infections are compared with clinically apparent infections so as to properly assess predicted burdens both with and without interventions. It is biologically plausible that milder cases are less transmissible, for example, because of an absence of cough^16,28, but direct evidence is limited⁴⁹ and viral load is high in both clinical and subclinical cases³⁶. If those with no or mild symptoms are efficient transmitters of infection compared with those with fully symptomatic infections, the overall burden is higher than if they are not as infectious. At the same time, lower relative infectiousness would reduce the impact of interventions targeting children, such as school closure. By analyzing epidemic dynamics before and after school closures, or close follow-up in household studies, it might be possible to estimate the infectiousness of subclinical infections, but this analysis will rely on granular data by age and time.

A great deal of concern has been directed toward the expected burden of COVID in low- and middle-income countries (LMICs), which generally have a lower population median age than many high-income countries. Our results show that these demographic differences, coupled with a lower susceptibility and clinical fraction in younger ages, can result in proportionally fewer clinical cases than would be expected in high-income countries with flatter demographic pyramids. This finding should not be interpreted as fewer cases in LMICs, because the projected epidemics remain large. Moreover, the relationships found between age, susceptibility and clinical fraction are drawn from high-income and middle-income countries and might reflect not only age, but also the increasing frequency of comorbidities with age. This relationship could therefore differ in LMICs for two key reasons. First, the distribution of non-communicable comorbid conditions—which are already known to increase the risk of severe disease from COVID¹⁸—might be differently distributed by age⁵⁰, along with other risk factors such as undernutrition⁵¹. Second, communicable comorbidities such as HIV⁴⁷, tuberculosis co-infection (which has been suggested to increase risk⁵²) and others⁵³ could alter the distribution of severe outcomes by age. Observed severity and burden in LMICs might also be higher than in HICs due to a lack of health system capacity for intensive treatment of severe cases.

There are some limitations to the study. Information drawn from the early stages of the epidemic is subject to uncertainty; however, age-specific information in our study is drawn from several regions and countries, and clinical studies^1,54 support the hypothesis presented here. We assumed that clinical cases are reported at a fixed fraction throughout the time period, although there may have been changes in reporting and testing practices that affected case ascertainment by age. We assumed that subclinical infections are less infectious than clinically apparent infections. We tested the effects of differences in infectivity on our findings (Extended Data Figs. 7 and 8) but were not able to estimate how infectious subclinical cases were. The sensitivity analyses showed very similar clinical fraction and susceptibility with age, and we demonstrated the effect of this parameter on school closure and global projections (Fig. 3 and Extended Data Fig. 8). We used mixing matrices from the same country, but not the same location as the fitted data. We used contact matrices that combined physical and conversational contacts. We therefore implicitly assume that they are a good reflection of contact relevant for the transmission of SARS-CoV However, if fomite or fecal–oral routes are important contributors to transmission, these contact matrices might not be representative of overall transmission risk.

The role of age in transmission is critical to designing interventions aiming to decrease transmission in the population as a whole and to projecting the expected global burden. Our findings, together with early evidence¹⁶, suggest that there is age dependence in susceptibility and in the risk of clinical symptoms following infection with SARS-CoV Understanding if and by how much subclinical infections contribute to transmission has implications for predicted global burden and the effectiveness of control interventions. This question must be resolved to effectively forecast and control COVID epidemics.

Canadian Pacific system map (does not include DM&E or CMQ trackage)
Overview
Reporting mark	CP, CPAA, MILW, SOO, DME, ICE, DH
Locale	Canada and the United States
Dates of operation	16 February &#;present
Technical
Track gauge	1,&#;mm (4&#;ft&#;8&#;1&#;2&#;in) standard gauge
Length	20, kilometres (12,&#;mi)
Other
Website	manicapital.com

An alternative CP logo, featuring a beaver, Canada's national animal

The Canadian Pacific Railway (CPR) (reporting marksCP, CPAA, MILW, SOO), known as CP Rail between and and simply Canadian Pacific, is a historic CanadianClass I railway incorporated in The railway is owned by Canadian Pacific Railway Limited, which began operations as legal owner in a corporate restructuring in ^[2]

Headquartered in Calgary, Alberta, it owns approximately 20, kilometres (12,&#;mi) of track in six provinces of Canada and into the United States,^[2] stretching from Montreal to Vancouver, and as far north as Edmonton. Its rail network also serves Minneapolis–St. Paul, Milwaukee, Detroit, Chicago, and Albany, New York in the United States.

The railway was first built between eastern Canada and British Columbia between and (connecting with Ottawa Valley and Georgian Bay area lines built earlier), fulfilling a commitment extended to British Columbia when it entered Confederation in It was Canada's first transcontinental railway, but no longer reaches the Atlantic coast. Primarily a freight railway, the CPR was for decades the only practical means of long-distance passenger transport in most regions of Canada, and was instrumental in the settlement and development of Western Canada. The CPR became one of the largest and most powerful companies in Canada, a position it held as late as ^[3] Its primary passenger services were eliminated in , after being assumed by Via Rail Canada in A beaver was chosen as the railway's logo in honour of Donald Smith, 1st Baron Strathcona and Mount Royal, who had risen from factor to governor of the Hudson's Bay Company over a lengthy career in the beaver fur trade. Smith was a principal financier of the CPR.^[4] staking much of his personal wealth. In , he drove the last spike to complete the transcontinental line.^[4]

The company acquired two American lines in the Dakota, Minnesota and Eastern Railroad and the Iowa, Chicago and Eastern Railroad. The trackage of the IC&E was at one time part of CP subsidiary Soo Line and predecessor line The Milwaukee Road. The combined DME/ICE system spanned North Dakota, South Dakota, Minnesota, Wisconsin, Nebraska and Iowa, as well as two short stretches into two other states, which included a line to Kansas City, Missouri, and a line to Chicago, Illinois, and regulatory approval to build a line into the Powder River Basin of Wyoming. It is publicly traded on both the Toronto Stock Exchange and the New York Stock Exchange under the ticker CP. Its U.S. headquarters are in Minneapolis.^[5]

History[edit]

Together with the Canadian Confederation, the creation of the Canadian Pacific Railway was a task originally undertaken as the National Dream by the Conservative government of Prime Minister Sir John A. Macdonald (1st Canadian Ministry).^[6] He was helped by Sir Alexander Tilloch Galt, who was the owner of the North Western Coal and Navigation Company. British Columbia, a four-month sea voyage away from the East Coast, had insisted upon a land transport link to the East as a condition for joining Confederation (initially requesting a wagon road).^[7] The government however proposed to build a railway linking the Pacificprovince to the Eastern provinces within 10 years of 20 July Macdonald saw it as essential to the creation of a unified Canadian nation that would stretch across the continent. Moreover, manufacturing interests in Quebec and Ontario wanted access to raw materials and markets in Western Canada.^{[citation needed]}

The first obstacle to its construction was political. The logical route went through the American Midwest and the city of Chicago, Illinois (via some Milwaukee Road and Soo Line Railroad trackage that would later be acquired by CP in the late 20th century). In addition to this was the difficulty of building a railway through the Canadian Rockies; an entirely Canadian route would require crossing 1,&#;km (&#;mi) of rugged terrain across the barren Canadian Shield and muskeg of Northern Ontario. To ensure this routing, the government offered huge incentives including vast grants of land in the West.^{[citation needed]}

In , Sir John A. Macdonald and other high-ranking politicians, bribed in the Pacific Scandal, granted federal contracts to Hugh Allan's Canada Pacific Railway Company (which was unrelated to the current company) rather than to David Lewis Macpherson's Inter-Ocean Railway Company which was thought to have connections to the American Northern Pacific Railway Company. Because of this scandal, the Conservative Party was removed from office in The new Liberal prime minister, Alexander Mackenzie, ordered construction of segments of the railway as a public enterprise under the supervision of the Department of Public Works led by Sandford Fleming. Surveying was carried out during the first years of a number of alternative routes in this virgin territory followed by construction of a telegraph along the lines that had been agreed upon.^[8] The Thunder Bay section linking Lake Superior to Winnipeg was commenced in By , around 1, kilometres (&#;mi) was nearly complete, mainly across the troublesome Canadian Shield terrain, with trains running on only kilometres (&#;mi) of track.^[9]

C.P.R. locomotive and employees

With Macdonald's return to power on 16 October , a more aggressive construction policy was adopted. Macdonald confirmed that Port Moody would be the terminus of the transcontinental railway, and announced that the railway would follow the Fraser and Thompson rivers between Port Moody and Kamloops. In , the federal government floated bonds in London and called for tenders to construct the &#;km (&#;mi) section of the railway from Yale, British Columbia, to Savona's Ferry, on Kamloops Lake. The contract was awarded to Andrew Onderdonk, whose men started work on 15 May After the completion of that section, Onderdonk received contracts to build between Yale and Port Moody, and between Savona's Ferry and Eagle Pass.^[10]

On 21 October , a new syndicate, unrelated to Hugh Allan's, signed a contract with the Macdonald government. Fleming was dismissed and replaced with Sir Collingwood Schreiber as chief engineer and general manager of all government railways. They agreed to build the railway in exchange for $25 million (approximately $&#;million in modern Canadian dollars) in credit from the Canadian government and a grant of 25&#;million acres (,&#;km²) of land. The government transferred to the new company those sections of the railway it had constructed under government ownership, on which it had already spent at least $25&#;million. But its estimates of the cost of the Rocky Mountain section alone was over $60&#;million.^[11] The government also defrayed surveying costs and exempted the railway from property taxes for 20 years. The Montreal-based syndicate officially comprised five men: George Stephen, James J. Hill, Duncan McIntyre, Richard B. Angus and John Stewart Kennedy. Donald A. Smith^[4] and Norman Kittson were unofficial silent partners with a significant financial interest. On 15 February , legislation confirming the contract received royal assent, and the Canadian Pacific Railway Company was formally incorporated the next day.^[12] Critics claimed that the government gave too large a subsidy for the proposed project but this was to incorporate uncertainties of risk and irreversibility of insurance. The large subsidy also needed to compensate the CPR for not constructing the line in the future, but rather right away even though demand would not cover operational costs.^[13]

Building the railway, –[edit]

Canadian Pacific Railway Crew laying tracks at lower Fraser Valley,

Building the railway took over four years. James J. Hill in sent Alpheus Beede Stickney to be construction superintendent for the Canadian Pacific Railway. The Canadian Pacific Railway began its westward expansion from Bonfield, Ontario (previously called Callander Station), where the first spike was driven into a sunken railway tie. Bonfield was inducted into Canadian Railway Hall of Fame in as the CPR first spike location. That was the point where the Canada Central Railway extension ended.^[14] The CCR was owned by Duncan McIntyre, who amalgamated it with the CPR, and became one of the handful of officers of the newly formed CPR. The CCR started in Brockville and extended to Pembroke. It then followed a westward route along the Ottawa River passing through places like Cobden, Deux-Rivières and eventually to Mattawa at the confluence of the Mattawa and Ottawa rivers. It then proceeded cross-country towards its final destination of Bonfield. Duncan McIntyre and his contractor James Worthington piloted the CPR expansion. Worthington continued on as the construction superintendent for the CPR past Bonfield. He remained with the CPR for about a year after which he left the company. McIntyre was uncle to John Ferguson who staked out future North Bay and who became the town's wealthiest inhabitant and mayor for four successive terms.^[15]

It was presumed that the railway would travel through the rich "Fertile Belt" of the North Saskatchewan River Valley and cross the Rocky Mountains via the Yellowhead Pass, a route suggested by Sir Sandford Fleming based on a decade of work. However, the CPR quickly discarded this plan in favour of a more southerly route across the arid Palliser's Triangle in Saskatchewan and via Kicking Horse Pass and down the Field Hill to the Rocky Mountain Trench. This route was more direct and closer to the Canada–US border, making it easier for the CPR to keep American railways from encroaching on the Canadian market. However, this route also had several disadvantages.

One was that the CPR would need to find a route through the Selkirk Mountains in British Columbia while, at the time, it was not known whether a route even existed. The job of finding a pass was assigned to a surveyor named Major Albert Bowman Rogers. The CPR promised him a cheque for $5, and that the pass would be named in his honour. Rogers became obsessed with finding the pass that would immortalize his name. He discovered the pass in April ^[16][17] and, true to its word, the CPR named it "Rogers Pass" and gave him the cheque. However, he at first refused to cash it, preferring to frame it, saying he did not do it for the money. He later agreed to cash it with the promise of an engraved watch.^{[citation needed]}

Another obstacle was that the proposed route crossed land in Alberta that was controlled by the BlackfootFirst Nation. This difficulty was overcome when a missionary priest, Albert Lacombe, persuaded the Blackfoot chief Crowfoot that construction of the railway was inevitable. In return for his assent, Crowfoot was famously rewarded with a lifetime pass to ride the CPR.

A more lasting consequence of the choice of route was that, unlike the one proposed by Fleming, the land surrounding the railway often proved too arid for successful agriculture. The CPR may have placed too much reliance on a report from naturalist John Macoun, who had crossed the prairies at a time of very high rainfall and had reported that the area was fertile.^[18]

The greatest disadvantage of the route was in Kicking Horse Pass, at the Alberta-British Columbia border on the continental divide. In the first 6&#;km (&#;mi) west of the 1, metres (5, feet) high summit, the Kicking Horse River drops metres (1, feet). The steep drop would force the cash-strapped CPR to build a 7&#;km (&#;mi) long stretch of track with a very steep 4¹&#;₂ percent gradient once it reached the pass in This was over four times the maximum gradient recommended for railways of this era, and even modern railways rarely exceed a two-percent gradient. However, this route was far more direct than one through the Yellowhead Pass and saved hours for both passengers and freight. This section of track was the CPR's Big Hill. Safety switches were installed at several points, the speed limit for descending trains was set at 10&#;km per hour (6&#;mph), and special locomotives were ordered. Despite these measures, several serious runaways still occurred including the first locomotive, which belonged to the contractors, to descend the line. CPR officials insisted that this was a temporary expediency, but this state of affairs would last for 25 years until the completion of the Spiral Tunnels in the early 20th century.^[19]

In , construction progressed at a pace too slow for the railway's officials who, in , hired the renowned railway executive William Cornelius Van Horne to oversee construction with the inducement of a generous salary and the intriguing challenge of handling such a difficult railway project. Van Horne stated that he would have &#;km (&#;mi) of main line built in Floods delayed the start of the construction season, but over &#;km (&#;mi) of main line, as well as sidings and branch lines, were built that year. The Thunder Bay branch (west from Fort William) was completed in June by the Department of Railways and Canals and turned over to the company in May , permitting all-Canadian lake and railway traffic from Eastern Canada to Winnipeg, for the first time in Canada's history. By the end of , the railway had reached the Rocky Mountains, just eight kilometres (five miles) east of Kicking Horse Pass. The construction seasons of and would be spent in the mountains of British Columbia and on the north shore of Lake Superior.

Many thousands of navvies worked on the railway. Many were European immigrants. In British Columbia, government contractors eventually hired workers from China, known as "coolies". A navvy received between $1 and $ per day, but had to pay for his own food, clothing, transport to the job site, mail and medical care. After 2¹&#;₂ months of hard labour, they could net as little as $ Chinese labourers in British Columbia made only between 75 cents and $ a day, paid in rice mats, and not including expenses, leaving barely anything to send home. They did the most dangerous construction jobs, such as working with explosives to clear tunnels through rock.^[20] The exact number of Chinese workers who died is unknown but historians estimate the number is between and The victims of sickness and accidents were not given proper funerals. Most of the remains were buried into the railway and the families of the Chinese who were killed received no compensation, or even notification of loss of life. Many of the men who survived did not have enough money to return to their families in China, although Chinese labour contractors had promised that as part of their responsibilities.^[21] Many spent years in isolated and often poor conditions. Yet the Chinese were hard working and played a key role in building the Western stretch of the railway; even some boys as young as twelve years old served as tea-boys. In , the Canadian government issued a formal apology to the Chinese population in Canada for their treatment both during and following the construction of the CPR.^[22]

By , railway construction was progressing rapidly, but the CPR was in danger of running out of funds. In response, on 31 January , the government passed the Railway Relief Bill, providing a further $&#;million in loans to the CPR. The bill received royal assent on 6 March ^[23]

In March , the North-West Rebellion broke out in the District of Saskatchewan. Van Horne, in Ottawa at the time, suggested to the government that the CPR could transport troops to Qu'Appelle, Saskatchewan (Assiniboia) in 10 days. Some sections of track were incomplete or had not been used before, but the trip to Winnipeg was made in nine days and the rebellion quickly suppressed. Perhaps because the government was grateful for this service, they subsequently reorganized the CPR's debt and provided a further $5&#;million loan. This money was desperately needed by the CPR. However, this government loan later became controversial. Even with Van Horne's support with moving troops to Qu'Appelle, the government still delayed in giving its support to CPR. This was due to Sir John A. Macdonald putting pressure on George Stephen for additional benefits. Stephen himself later did admit to spending $1&#;million between and to ensure government support. This money went to buying a £40, necklace for Lady MacDonald and numerous other "bonifications" to government members.^[24]

Telegram to Prime Minister John A. Macdonald announcing the completion of the Canadian Pacific Railway, 7 November

On 7 November , the last spike was driven at Craigellachie, British Columbia, making good on the original promise. Four days earlier, the last spike of the Lake Superior section was driven in just west of Jackfish, Ontario. While the railway was completed four years after the original deadline, it was completed more than five years ahead of the new date of that Macdonald gave in The successful construction of such a massive project, although troubled by delays and scandal, was considered an impressive feat of engineering and political will for a country with such a small population, limited capital, and difficult terrain. It was by far the longest railway ever constructed at the time. It had taken 12, men and 5, horses to construct the Lake section alone.^[25]

Meanwhile, in Eastern Canada, the CPR had created a network of lines reaching from Quebec City to St. Thomas, Ontario, by (mainly by buying the Quebec, Montreal, Ottawa & Occidental Railway from the Quebec government), and had launched a fleet of Great Lakes ships to link its terminals. The CPR had effected purchases and long-term leases of several railways through an associated railway company, the Ontario and Quebec Railway (O&Q). The O&Q built a line between Perth, Ontario, and Toronto (completed on 5 May ) to connect these acquisitions. The CPR obtained a year lease on the O&Q on 4 January In , it acquired a minority interest in the Toronto, Hamilton and Buffalo Railway, giving it a link to New York and the Northeast United States.^[26]

–[edit]

The last spike in the CPR was driven on 7 November , by one of its directors, Donald Smith,^[4] but so many cost-cutting shortcuts were taken in constructing the railway that regular transcontinental service could not start for another seven months while work was done to improve the railway's condition (part of this was because of snow in the mountains and lack of snowsheds to keep the line open). However, had these shortcuts not been taken, it is conceivable that the CPR might have had to default financially, leaving the railway unfinished.^{[citation needed]}

The first transcontinental passenger train departed from Montreal's Dalhousie Station, located at Berri Street and Notre Dame Street at 8&#;pm on 28 June , and arrived at Port Moody at noon on 4 July This train consisted of two baggage cars, a mail car, one second-class coach, two immigrant sleepers, two first-class coaches, two sleeping cars and a diner (several dining cars were used throughout the journey, as they were removed from the train during the night, with another one added the next morning).^{[citation needed]}

First Transcontinental Train arrives in Port Arthur on 30 June

By that time, however, the CPR had decided to move its western terminus from Port Moody to Granville, which was renamed "Vancouver" later that year. The first official train destined for Vancouver arrived on 23 May , although the line had already been in use for three months. The CPR quickly became profitable, and all loans from the Federal government were repaid years ahead of time. In , a branch line was opened between Sudbury and Sault Ste. Marie where the CPR connected with the American railway system and its own steamships. That same year, work was started on a line from London, Ontario, to the Canada–US border at Windsor, Ontario. That line opened on 12 June ^[27]

The CPR also leased the New Brunswick Railway in for years,^[28] and built the International Railway of Maine, connecting Montreal with Saint John, New Brunswick, in The connection with Saint John on the Atlantic coast made the CPR the first truly transcontinental railway company in Canada and permitted trans-Atlantic cargo and passenger services to continue year-round when sea ice in the Gulf of St. Lawrence closed the port of Montreal during the winter months. By , competition with the Great Northern Railway for traffic in southern British Columbia forced the CPR to construct a second line across the province, south of the original line. Van Horne, now president of the CPR, asked for government aid, and the government agreed to provide around $&#;million to construct a railway from Lethbridge, Alberta, through Crowsnest Pass to the south shore of Kootenay Lake, in exchange for the CPR agreeing to reduce freight rates in perpetuity for key commodities shipped in Western Canada.^{[citation needed]}

The controversial Crowsnest Pass Agreement effectively locked the eastbound rate on grain products and westbound rates on certain "settlers' effects" at the level. Although temporarily suspended during the First World War, it was not until that the "Crow Rate" was permanently replaced by the Western Grain Transportation Act which allowed for the gradual increase of grain shipping prices. The Crowsnest Pass line opened on 18 June , and followed a complicated route through the maze of valleys and passes in southern British Columbia, rejoining the original mainline at Hope after crossing the Cascade Mountains via Coquihalla Pass.^[29]

The Southern Mainline, generally known as the Kettle Valley Railway in British Columbia, was built in response to the booming mining and smelting economy in southern British Columbia, and the tendency of the local geography to encourage and enable easier access from neighbouring US states than from Vancouver or the rest of Canada, which was viewed to be as much of a threat to national security as it was to the province's control of its own resources. The local passenger service was re-routed to this new southerly line, which connected numerous emergent small cities across the region. Independent railways and subsidiaries that were eventually merged into the CPR in connection with this route were the Shuswap and Okanagan Railway, the Kaslo and Slocan Railway, the Columbia and Kootenay Railway, the Columbia and Western Railway and various others.^[28]

CPR and the settlement of western Canada[edit]

One of the CPR's land offerings,

The CPR had built a railway that operated mostly in the wilderness. The usefulness of the prairies was questionable in the minds of many. The thinking prevailed that the prairies had great potential. Under the initial contract with the Canadian government to build the railway, the CPR was granted 25&#;million acres (,&#;km²). Proving already to be a very resourceful organization, Canadian Pacific began an intense campaign to bring immigrants to Canada. Canadian Pacific agents operated in many overseas locations. Immigrants were often sold a package that included passage on a CP ship, travel on a CP train and land sold by the CP railway. Land was priced at $ an acre and up but required cultivation.^[30] To transport immigrants, Canadian Pacific developed a fleet of over a thousand Colonist cars, low-budget sleeper cars designed to transport immigrant families from eastern Canadian seaports to the west.^[31]

–[edit]

CPR advertisement highlighting "Free Farms for the Million" in western Canada, circa

During the first decade of the 20th century, the CPR continued to build more lines. In , the CPR opened a line connecting Toronto with Sudbury. Previously, westbound traffic originating in southern Ontario took a circuitous route through eastern Ontario. Several operational improvements were also made to the railway in Western Canada. In the CPR completed two significant engineering accomplishments. The most significant was the replacement of the Big Hill, which had become a major bottleneck in the CPR's main line, with the Spiral Tunnels, reducing the grade to percent from percent. The Spiral Tunnels opened in August. In April , the CPR started work to replace the Old Calgary-Edmonton Rail Bridge across the Red Deer River with a new standard steel bridge that was completed by March ^[32]

On 3 November , the Lethbridge Viaduct over the Oldman River valley at Lethbridge, Alberta, was opened. It is 1, metres (5, feet) long and, at its maximum, 96 metres ( feet) high, making it one of the longest railway bridges in Canada. In , the CPR replaced its line through Rogers Pass, which was prone to avalanches (the most serious of which killed 62 men in ) with the Connaught Tunnel, an eight-kilometre-long (5-mile) tunnel under Mount Macdonald^[33] that was, at the time of its opening, the longest railway tunnel in the Western Hemisphere.^[34][35]

C.P.R. railway locomotive

On 21 January , a passenger train derailed on the CPR line at the Spanish River bridge at Nairn, Ontario (near Sudbury), killing at least ^[36][37]

The CPR acquired several smaller railways via long-term leases in On 3 January , the CPR acquired the Dominion Atlantic Railway, a railway that ran in western Nova Scotia. This acquisition gave the CPR a connection to Halifax, a significant port on the Atlantic Ocean. The Dominion Atlantic was isolated from the rest of the CPR network and used the CNR to facilitate interchange; the DAR also operated ferry services across the Bay of Fundy for passengers and cargo (but not rail cars) from the port of Digby, Nova Scotia, to the CPR at Saint John, New Brunswick. DAR steamships also provided connections for passengers and cargo between Yarmouth, Boston and New York. On 1 July , the CPR acquired the Esquimalt and Nanaimo Railway, a railway on Vancouver Island that connected to the CPR using a railcar ferry. The CPR acquired the Quebec Central Railway on 14 December ^[28]

During the late 19th century, the railway undertook an ambitious programme of hotel construction, building Glacier House in Glacier National Park, Mount Stephen House at Field, British Columbia, the Château Frontenac in Quebec City and the Banff Springs Hotel. By then, the CPR had competition from three other transcontinental lines, all of them money-losers. In , these lines were consolidated, along with the track of the old Intercolonial Railway and its spurs, into the government-owned Canadian National Railways. The CPR suffered its greatest loss of life when one of its steamships, the Empress of Ireland, sank after a collision with the Norwegian collier SS Storstad. On 29 May , the Empress (operated by the CPR's Canadian Pacific Steamship Company) went down in the St. Lawrence River with the loss of 1, lives, of which were passengers.^[38]

First World War[edit]

During the First World War CPR put the entire resources of the "world's greatest travel system" at the disposal of the British Empire, not only trains and tracks, but also its ships, shops, hotels, telegraphs and, above all, its people. Aiding the war effort meant transporting and billeting troops; building and supplying arms and munitions; arming, lending and selling ships. Fifty-two CPR ships were pressed into service during World War I, carrying more than a million troops and passengers and four million tons of cargo. Twenty seven survived and returned to CPR. CPR also helped the war effort with money and jobs. CPR made loans and guarantees to the Allies of some $&#;million. As a lasting tribute, CPR commissioned three statues and 23 memorial tablets to commemorate the efforts of those who fought and those who died in the war.^[39] After the war, the Federal government created Canadian National Railways (CNR, later CN) out of several bankrupt railways that fell into government hands during and after the war. CNR would become the main competitor to the CPR in Canada. In , Henry Worth Thornton replaced David Blyth Hanna becoming the second president of the CNR, and his competition spurred Edward Wentworth Beatty, the first Canadian-born president of the CPR, to action.^[40] During this time the railway land grants were formalized.^[41]

Great Depression and the Second World War, –[edit]

Strikers from unemployment relief camps climbing on boxcars as part of the On-to-Ottawa Trek,

The Great Depression, which lasted from until , hit many companies heavily. While the CPR was affected, it was not affected to the extent of its rival CNR because it, unlike the CNR, was debt-free. The CPR scaled back on some of its passenger and freight services, and stopped issuing dividends to its shareholders after Hard times led to the creation of new political parties such as the Social Credit movement and the Cooperative Commonwealth Federation, as well as popular protest in the form of the On-to-Ottawa Trek.^[42]

One highlight of the late s, both for the railway and for Canada, was the visit of King George VI and Queen Elizabeth during their royal tour of Canada, the first time that the reigning monarch had visited the country. The CPR and the CNR shared the honours of pulling the royal train across the country, with the CPR undertaking the westbound journey from Quebec City to Vancouver. Later that year, the Second World War began. As it had done in World War I, the CPR devoted much of its resources to the war effort. It retooled its Angus Shops in Montreal to produce Valentine tanks and other armoured vehicles, and transported troops and resources across the country. As well, 22 of the CPR's ships went to war, 12 of which were sunk.^[43]

–[edit]

The Multimark logo was used from to , when it fell out of favour. It was sometimes referred to as the 'Pac-Man' logo, after the popular s video game of the same name.

CPR train step stool (Calgary station) c.&#;

After the Second World War, the transportation industry in Canada changed. Where railways had previously provided almost universal freight and passenger services, cars, trucks and airplanes started to take traffic away from railways. This naturally helped the CPR's air and trucking operations, and the railway's freight operations continued to thrive hauling resource traffic and bulk commodities. However, passenger trains quickly became unprofitable. During the s, the railway introduced new innovations in passenger service. In , it introduced The Canadian, a new luxury transcontinental train. However, in the s, the company started to pull out of passenger services, ending services on many of its branch lines. It also discontinued its secondary transcontinental train The Dominion in , and in , unsuccessfully applied to discontinue The Canadian. For the next eight years, it continued to apply to discontinue the service, and service on The Canadian declined markedly. On 29 October , CP Rail transferred its passenger services to Via Rail, a new federal Crown corporation that is responsible for managing all intercity passenger service formerly handled by both CP Rail and CN. Via eventually took almost all of its passenger trains, including The Canadian, off CP's lines.^[44]

In , as part of a corporate reorganization, each of the major operations, including its rail operations, were organized as separate subsidiaries. The name of the railway was changed to CP Rail, and the parent company changed its name to Canadian Pacific Limited in Its air, express, telecommunications, hotel and real estate holdings were spun off, and ownership of all of the companies transferred to Canadian Pacific Investments. The slogan was: "TO THE FOUR CORNERS OF THE WORLD" The company discarded its beaver logo, adopting the new Multimark {which, when mirrored by an adjacent "multi-mark" creates a diamond appearance on a globe} that was used—with a different colour background—for each of its operations.^[45]

–[edit]

On 10 November , a derailment of a hazardous materials train in Mississauga, Ontario, led to the evacuation of , people; there were no fatalities.^[46][47]

In , CP Rail commenced construction of the Mount Macdonald Tunnel to augment the Connaught Tunnel under the Selkirk Mountains. The first revenue train passed through the tunnel in At &#;km (nine miles), it is the longest tunnel in the Americas. During the s, the Soo Line Railroad, in which CP Rail still owned a controlling interest, underwent several changes. It acquired the Minneapolis, Northfield and Southern Railway in Then on 21 February , the Soo Line obtained a controlling interest in the bankrupt Milwaukee Road