Every Transit User is Also a Pedestrian
Walking and Public Transport: Competitors or Allies?
Increasing the use of public transport is not enough to make urban transport environmentally sustainable. Trains, buses and trams produce greenhouse gases and other pollutants, even if these are emitted from the power station instead of the tailpipe. Increased patronage only offers environmental benefits when it reduces car travel – provided passengers are carried at sufficiently high occupancy rates. If patronage gains come at the expense of walking or cycling, or because overall travel has risen, then environmental problems are increased, not reduced. Some free travel and park-and-ride schemes have produced increases in ridership while worsening environmental outcomes, as we saw in the previous chapter.
The only completely sustainable forms of travel are walking and cycling. Increases in cycling can impose small financial costs for infrastructure and parking, and increased risk to pedestrians (as discussed below, cyclists occasionally injure and even kill pedestrians). But more cycling will not increase pollution or the demand for oil, and will even improve overall fitness levels.
Walking produces similar benefits and no costs at all, whether it comes at the expense of the car or not. Indeed, medical practitioners and researchers have been largely responsible for the recent revival of interest in walking: transport planners have traditionally paid even less attention to walking than to public transport and cycling. Walking is now seen as an important part of the response to the declines in fitness, and associated health risks, resulting from sedentary lifestyles. Increased walking can improve health as well as the urban and global environment.
Unfortunately, in most places the trend has been to less walking, not more. The UK’s National Travel Survey found that walking trip rates fell by 16 per cent in the ten years to 2005, following steady declines in previous decades.1 Even in Zurich, as we saw in Chapter 8, walking is declining for trips to work, although it remains very high for school travel.
Precise rates and trends are difficult to estimate, because most travel surveys either ignore walking or understate it. Some explicitly exclude walking, while others omit short trips (e.g. under a kilometre), most of which are made on foot. Travel surveys that count all travel often miss short walking trips because respondents don’t remember them. Careful use of travel diaries is needed to record all these trips. When this was done in Melbourne in the mid-1990s, it appeared that there was twice as much walking as in other Australian cities: in fact, the difference really showed that the other cities’ surveys were missing half of all walking trips. Many surveys categorize trips on a ‘main mode’ basis rather than counting each stage separately: a walk to the tram stop, followed by a tram ride, then another walk to the destination will be recorded as ‘tram’. Finally, many surveys combine the number and length of trips to produce ‘passenger-kilometre’ figures – making walking trips, which are much shorter than car, public transport and even cycle trips, appear insignificant.
The importance of walking is usually underestimated, but the level of understatement varies from one survey to another, making it difficult to track trends over time and to compare different cities and countries. In this book, I have focused on trips to work and school, which people usually remember, and surveys usually report on a comparable basis. These figures reduce the difficulties of cross-city comparison, but need to be studied closely to reveal the true importance of walking. For example, as Chapter 8 reported, 66 per cent of people living and working in the City of Zurich travel to work by public transport; only 10 per cent walk, while another 6 per cent cycle. But virtually all the public transport users walk to the tram, train or bus stop, so the share of workers walking for at least part of their work journey is more like 76 per cent.
There has been little research into the relationship between public transport and walking, the main exception being two reports from the British-based academic and consultant Carmen Hass-Klau.2 The two modes can interact in different ways.
The first is as competitors: high-quality public transport will attract some people who would otherwise walk, which means an increase in pollution and greenhouse emissions. Mediocre public transport and low car ownership are significant factors behind the relatively high rates of walking in many UK cities, but this is hardly the best way to promote walking. It is inequitable and unsustainable, since poor public transport will drive increases in car ownership as incomes rise, ultimately producing declines in walking.
Walking can support public transport, since as Ruedi Ott, head of traffic planning at the Zurich City Council says, ‘every public transport user is also a pedestrian’. Therefore, ‘if people no longer wish to leave their homes on foot, whether because of unattractive footpaths, pointless detours, pollution caused by emissions, security problems, or because they feel that, as pedestrians, they are not treated as participants in traffic in their own right, it will not be possible to promote the use of public transport instead of the car.’3 Making life easier for pedestrians will also encourage public transport use.
Public transport can in turn support walking. Because every transit user is a pedestrian, public transport trips generate walking trips – except in systems that rely heavily on park-and-ride – as we saw above in the case of Zurich. But high quality public transport also promotes walking indirectly.
The first way it does this is by reducing car ownership. Hass-Klau’s research in Europe found that even when income, family type and other factors are corrected for, high quality public transport lowered car ownership levels by about 9 per cent on average, with bigger falls in some cities. The second way public transport influences walking is more subtle, but it may be more important. As far as I am aware, it has been the subject of no systematic research at all.
Walking and Habit Formation
A very high share of trips, even in car-dominated cities, are short enough to be made on foot. This applies especially to non-work trips such as the journey to school. But in most urban areas in English-speaking countries, only a small share of these short trips are in fact made on foot. As concern about pollution, greenhouse emissions and sedentary lifestyles has grown, a range of programmes has been developed to deal with the problem of short car trips. Social marketing campaigns with names like Travelsmart and school-based programmes such as the ‘walking school bus’ attack the problem by persuading urban dwellers to change their behaviour.
The assumption behind the ‘behaviour change’ approach is that people would stop driving short distances if only they were better informed. If this sounds a little patronizing, that may be part of the appeal: there is a long history of campaigns to encourage virtuous behaviour among the masses, dating back at least to the temperance movement of a century ago. Rather than change the reality, we should change people’s attitudes. Naturally, the same approach is attractive to governments that don’t want to change the substantive realities of transport policy. Is it a coincidence that behaviour change is generously funded in Australia and the UK, but plays little or no role in Scandinavia and Switzerland?
Encouraging results have been reported from behaviour change trials. Residents who have accepted Travelsmart literature, and schools with walking buses, all say that there is now much more walking and less driving than before. Unfortunately, these reports cannot be relied on as indicators that genuine, lasting change has taken place, for two reasons. First, it is rare to see follow-up surveys taken to see whether changes in behaviour outlast the original programme. One of the few cases where this was done is the British city of York. Follow-up surveys showed that ‘the intervention was successful initially in inducing behavioural change but behaviour returned back to normal levels of travel 12 months later.’4
The second problem is that behaviour change studies are set up in ways that almost guarantee inaccurate results. It is a commonplace of the medical and social scientific literature that results obtained under such circumstances will be unreliable because of the ‘good subject effect’. Participants in experiments want to help the experiment succeed, by giving the ‘right’ answer. If it is made clear at the outset that a particular result is not just correct but morally and environmentally so, then this result will be reported whether or not it really occurred. The participants aren’t being dishonest; they are just trying to help.5
Sources uncontaminated by the ‘good subject effect’, such as censuses and national travel surveys, produce less encouraging results. The Australian and British cities that have seen the most intensive applications of behaviour change programmes are precisely those in which walking is either minimal or in decline, or both. The European cities where walking is a major travel mode are those where behaviour change programmes have not been used. Nobody in Zurich has ever seen a walking school bus, but only 3 per cent of students across the whole canton travel to school by car. In Canton Graubunden the share is only 4 per cent, one point higher than the figure for Canton Schaffhausen, where 49 per cent of students walk, 17 per cent cycle and 30 per cent use public transport. Neither canton has seen a walking school bus either.6
So why is car use so low, and walking and cycling so high, for education trips in these Swiss cantons? The trends apply in suburbs and rural areas, not just in inner cities, so density can’t be the explanation. Nor can the extensive pedestrian priority measures found in Swiss inner cities explain the high walking rates of suburban students, although they doubtless help, as do other measures like lower speed limits. The answer lies elsewhere.
Transport planners and researchers often forget that most people are not very interested in transport. They have lives, jobs, friends and families to worry about: transport is simply a means of reaching them. So economic analyses that assume every traveller performs a cost–benefit analysis of the competing modes before setting out on each journey may be as far off the mark as behaviour change literature that wants people to think about global warming or even fitness before doing so. We are all creatures of habit.
People in car-dominated environments can lose the habit of walking. In large sections of Australian, Canadian, US and British suburbia many residents would struggle to recall the last time they left home on foot; the activities of leaving the house and getting into the car have merged. Pious exhortations to walk more for the good of one’s own health or that of the globe may make some people guilty enough to change their own, or their children’s, travel mode for a while, but without the reinforcement of habitual walking, change is unlikely to last. Just as the children of smokers are more likely to smoke themselves, so the children of households where all adult travel is made by car are likely to graduate quickly from the school bus to the automobile.
Recent work by Dutch medical researchers – significantly, not transport planners – confirms the critical importance of habit. A study of cycling rates found that habit was more influential than good intentions. The researchers concluded:
Health education efforts to increase physical activity levels have com-monly used persuasive communication strategies in which the target population is provided with information on, for instance, the health benefits of regular activity … [these] approaches may have limited value for strongly habitual behaviours… Any behaviour that can be performed on a regular basis in a stable context is likely to become habitual … [walking] may also become habitual rather than intentional.7
The Dutch researchers did not discuss in detail the alternative approaches that would respond to the importance of habit, but did note that ‘policy changes at the macro level’ are likely to be required.
Here, then, we have the key. Something needs to happen to the substantive reality of transport policy to make walking a natural, habitual activity. Provision of a comprehensive, high-quality public transport network does this, by creating large numbers of walking trips to and from stops and stations. Even in suburban and rural parts of Canton Zurich, it is rare to find a household where nobody uses public transport. Having acquired the habit of walking to the bus stop, people then walk to other local destinations such as schools and shops.
Comprehensive public transport promotes walking even in urban environments that would otherwise discourage it. The ‘middle suburbs’ of Toronto, the outer sections of the City of Toronto, present a hostile environment for pedestrians. Designed by developers in the 1950s and 1960s in accordance with modernist ideas derived from Le Corbusier, these suburbs feature cul-de-sac street systems, large surface car parks and no corner stores. Main roads are lined by the back fences of houses, and drive-in shopping malls.8 Yet huge numbers of pedestrians negotiate this forbidding landscape to reach the Toronto Transit Commission’s high-quality arterial bus routes. By contrast, the equivalent areas of Melbourne were largely subdivided between the 1880s and 1920s, and feature ‘traditional’ designs with straight streets and plenty of corner shops. But hardly anyone walks and most corner shops have closed for lack of custom. This is not an argument for continuing with pedestrian-unfriendly designs, but it does show that high-quality public transport can generate walking trips in the most unlikely places.
This is why high rates of public transport use are generally accompanied by higher rates of walking, as we saw in Chapters 4 and 8. While high-quality public transport may suppress walking by offering attractive competition for short trips, it also promotes walking by helping create a walking habit in households. It appears that the second trend greatly outweighs the first, with the happy result that walking and public transport can reinforce one another, working as allies in the move beyond the automobile age.
Cycling And Public Transport : The Legacy of the White Bike
The situation with cycling is more complex. We do not see the same happy coincidence we saw with walking: higher cycling rates are usually accompanied by lower public transport use, and often reduced walking as well. In Australia, cycling is positively correlated with car use, at least for travel to work: cycling rates are highest where car use is highest. Sydney and Brisbane, which have the lowest rates of car use, also have the lowest cycling rates (see Table 4.1).
This bad news may come as a surprise, given that many commentators regard cycling as the epitome of sustainable transport. It certainly generates more excitement than walking, despite catering for fewer urban trips, even in cities hailed as cycling models (this is often overlooked because surveys understate walking). Unlike walking, cycling is a hobby as well as a mode of transport, and for some people it is more than that.
The humble bicycle played a central role in countercultural movements of the 1960s, beginning with the Provos, a surrealist-style group operating in Amsterdam. The Provos announced a series of ‘white plans’ covering everything from childcare to policing. The most famous was the White Bike plan, which called for the city council to provide thousands of free bikes, painted white and scattered around the city, as an alternative to cars. The Provos staged a publicity stunt to promote the idea, painting ten bikes white and leaving them unlocked on city streets. The council confiscated them and the proposal came to nothing.9
Despite the lack of results, the White Bike concept became a legend among countercultural movements, and it was widely believed that there had been thousands of white bikes in Amsterdam. The legend was reinforced when Ivan Illich anointed the bicycle as the ideal urban transport mode in his 1974 manifesto Energy and Equity. Transport problems could be solved with simple, do-it-yourself measures: like the car in the 1920s (see Chapter 2), the bicycle would avoid the need for messy collective action or planning. University of Toronto philosopher Joseph Heath argues that some forms of cycling activism are ‘an individualized and basically apolitical environmentalism’ that evades the fact that urban transport is a classic instance of a ‘collective action problem’.10
Copenhagen actually installed a White-Bike-style scheme in 1995, and just as in Amsterdam the publicity has been out of all proportion to the results. Copenhagen has about 1500 ‘City Bikes’, and surveys show that they are used ‘primarily by tourists and young men’, for short trips that would otherwise be made on foot. As a result, ‘the City Bike adds nothing to the cycle traffic of Copenhagen.’11 Yet, in a development that could have been scripted by surrealists, over a dozen other cities have been inspired to instal similar schemes.
The main legacy of the surrealist White Bike seems to be a preference for stunts and myth-making over collective action and planning, and a disdain for the unpleasant realities of empirical evidence. When the results of the 2006 Australian census were released, the country’s three main broadsheet papers all led their coverage with the story that cycling rates had increased. And indeed they had: cycling numbers across the country’s seven capital cities had jumped by nearly 13,000 since the 2001 census (to 1.1 per cent of the workforce). But nobody reported the much larger increase of 32,000 in the number who walked to work, or the 72,000 additional public transport travellers – let alone the fact that the numbers travelling by car had increased by 312,000.12
Tim Pharoah, a strong advocate of cycling and walking, points out that more sustainable modes can compete with each other as well as with the car. Just as there is no environmental benefit if increased public transport patronage comes at the expense of walking or cycling, there is no gain if cycling rises at the expense of walking, or because people are cycling for recreation (although there may be health benefits). He offers the example of Munster, Germany, where cycling’s share of trips rose from 25 to 31 per cent between 1982 and 1990, exactly matching a fall in walking from 13 to 7 per cent. Car use remained unchanged at 55 per cent, as did public transport, with only 7 per cent of the travel market.13 The average resident made 73 public transport trips in 1990, a lower rate than in many Canadian, Australian and US cities. The low usage rate reflected relatively poor service. Following the closure of Munster’s trams in the 1950s, bus routes became less frequent and more complicated, even incorporating a separate ‘night network’ operating after 9:00 pm. Since then, public transport has been improved, and the trip rate has risen to 117, higher than in most Australian, US or British cities, but still less than half the rate in Schaffhausen and a seventh the rate in Zurich City.14 Munster has attracted (and deserves) much praise for its excellent cycling facilities, and even proclaims itself a ‘model city’, but it is not yet a model of sustainable urban transport.15
So when one hears that an increase in cycling has been achieved or proposed, the question should be: was this at the expense of the car? The two cycling capitals of Britain are the cities of Oxford and Cambridge. While Oxford often wins the University Boat Race, its traditional rival wins the bike race hands down. Cycling accounts for 28 per cent of work trips in Cambridge, compared with 16 per cent in Oxford. But travel to work by car is almost identical in the two cities, at 45.7 per cent in Oxford and 45.0 in Cambridge – because Oxford has more walking (16 versus 15 per cent) and more than twice as much public transport use (20 versus 9 per cent). And since there is little cycling outside the central urban areas, car use across the whole of Cambridgeshire is actually higher than in Oxfordshire (73 per cent versus 69).16
There are exceptions to this depressing picture. Hass-Klau cites German data suggesting that in the cities of Munich and Freiburg, public transport and cycling have increased in tandem, leading to falls in car trip shares.17 The small city of Troisdorf, near Cologne, apparently increased cycling from 16 to 21 per cent of trips without adversely affecting walking or public transport: instead, car use fell from 56 to 51 per cent.18 So cycling can complement other sustainable modes in the right policy environment. The question is what is the right environment?
In a refreshingly frank discussion of the tensions that occur between pedestrians, cyclists and public transport vehicles, Pharoah points out that some kinds of cyclist mix well with pedestrians and public transport passengers, but others do not. Young children and other ‘slow’ cyclists can coexist with pedestrians, but commuting and racing cyclists generally do not. ‘A great deal depends on cultural context of walking and cycling’, he notes: in cities like Amsterdam, Freiburg and Copenhagen, most cyclists ride ‘sedate’ machines, while British cyclists tend to ride racing or mountain bikes.19 Australian cyclists have an even stronger preference for expensive racingstyle bicycles, as well as Olympic-style lycra uniforms. Helmets, which are rarely seen in Europe, are now compulsory throughout Australia.
These differences reflect the composition of cyclists in high- and low-cycling countries. Where cycling is widespread, it is popular with both sexes and a range of ages and classes; as cycling declines in importance, it becomes increasingly restricted to young-ish middle-class males. In Copenhagen, the majority of cyclists are women, but 75 per cent of workers cycling in Melbourne in 2006 were male, with the Sydney figure even higher at 83 per cent. The UK National Travel Survey for 2005 found that cyclists came from households with higher than average incomes, in contrast with walking, which is most important in low-income households; male cyclists outnumbered females by between two and five to one, depending on the age group.20 This narrow profile contributes to, and is exacerbated by, the poor relationships with both motorists and pedestrians that come from the marginal status of cycling and the lack of safe bicycle paths.
Politicians have also learned from the White Bike that declaring support for cycling can create the illusion of a sustainable transport policy. This is handy when governments wish to deflect public concern over automobile-dominated substantive priorities. In Melbourne, every new freeway has its accompanying bike path, even though freeways, which bypass places people want to reach, provide unsuitable cycle routes. The state’s automobile association sponsors ‘Ride to Work Day’, while the road agency sponsors the ‘Great Victorian Bike Ride’.
Because it does not form part of a coordinated effort to reduce car use, tokenistic cycling policy is usually poorly thought through, with a focus on stunts and satisfying the demands of the small, but vocal, group of existing cyclists – whose male, middle-class demographic profile mirrors that of transport planners. So measures that appeal to ‘racing’ cyclists, such as showers at work and high-security storage for expensive bikes, predominate at the expense of practical issues like safe cycle paths and reduced speed limits. Where bicycle routes are provided, this is often at the expense of pedestrians, rather than motorists.
An example of this confused thinking is the curious notion of shared bicycle–bus lanes, a concept most fully developed in Auckland. The city’s ‘cycle network’ consists mainly of bike–bus lanes, with some shared footpaths and a very small number of separate bicycle roadways.21 It is easy to see why automobile-oriented transport planners might lump the ‘sustainable modes’ together on the same infrastructure, but a moment’s reflection reveals problems and dangers. As Pharoah argues and illustrates with an excellent photograph, ‘buses and cycles hinder one another when they share the same space’.22 Moving buses travel faster than bicycles but then stop to pick up and set down passengers, so bus drivers and cyclists race one another between stops. This slows buses down, increases driver stress and endangers cyclists: after all, buses are much bigger than cars, and take longer to stop.
Across the Tasman Sea in Melbourne, similar problems were created in 2008 when the Melbourne City Council introduced bike lanes on the roads feeding into Swanston Street, the city’s main tram artery. Swanston Street was converted to a transit mall in the 1990s, and was not designed to accommodate cyclists. Within months, a cyclist struck and seriously injured a passenger alighting from a tram, and another cyclist died after skidding on the tram tracks and falling into the path of a bus. The fact that walking, cycling and public transport are all more sustainable than car travel does not invalidate the laws of physics.
Another example of confused objectives is the widespread support for the carriage of bicycles on urban buses, trams and trains. Recreational cycling in rural areas is an environmentally benign form of tourism that is encouraged by sensible public transport providers: bicycles can be carried on RhB trains and Postbuses throughout rural Graubunden. But bikes are not carried on urban buses, or on peak-period RhB trains, because they take up too much space.
Public transport, it should be recalled, only offers environmental benefits relative to the car if high occupancy rates can be maintained. This is not possible if significant numbers of passengers bring bicycles with them, because a bike takes up as much space as three or four people, and slows boarding and alighting into the bargain. While the occasional cyclist can be accommodated, especially in off-peak periods, if all passengers took bikes with them, the environmental outcome would be no better than if they travelled by car.
The bike–rail combination is usually proposed as a remedy for problems that don’t exist in genuine public transport networks. For example, Professor Frank Fisher, the long-serving director of the Graduate School of Environmental Science at Melbourne’s Monash University, avoided the rail–bus shambles described in Chapter 1 for decades by bringing his bicycle on the train and cycling from Huntingdale station to the University. As a follower of Ivan Illich, Fisher was practising what he preached, but he goes further, arguing that bike–rail ‘makes for the fastest, healthiest, environmentally sanest, community-oriented transport system in existence’.23 The message is that instead of demanding that governments fix public transport, individuals should help themselves.
There are two problems with this argument. The first is that a train-load of passengers with bikes would not be environmentally sane because, as we have seen, another three or four trains would be needed just to carry the bikes. The second problem with Fisher’s argument is that it places the burden of dealing with poor public transport on individuals, letting governments off the hook. The bike–rail option would not be needed if Monash University was served by buses that connected with trains: at York University in Toronto, where this is the case, even people too lazy to cycle can reach campus easily without using cars.
There is no evidence that on-board carriage of bikes actually attracts significant numbers of former motorists, especially in public transport networks that already offer convenient connections to a full range of destinations. There has been little independent evaluation of bikes-onboard programmes, but the Toronto Transit Commission did report in 2006 on a trial programme to carry bikes on 110 buses serving five TTC routes. The routes were selected in consultation with the city’s cycling committee, and served the outer parts of the City of Toronto (i.e. the ‘middle suburbs’), where it was expected that demand would be greatest owing to longer trip lengths and walking distances to stops. Among those selected was the TTC’s busiest bus line, route 29 along Dufferin Avenue, which carries 44,000 passengers per day. The trial covered two summers and one winter, and cost around Can$200,000. Usage rates were minimal: between 9 and 15 bikes per day on route 29, and 19–32 across the five routes, many of whom were existing TTC patrons. The cost per new passenger was at least $10,000. TTC staff recommended that the project be discontinued in view of these disappointing results, but the elected councillors on the TTC board voted to extend the scheme to all TTC bus lines and investigate fitting bike racks to trams.24 Surrealist cycling policy did not stop with the Provos.
From Surrealism to Practical Policies
The real area where cycling and public transport can work together is access to stops and stations. Successful bike-and-ride schemes, mainly serving rail stations, can be found in Dutch, German, Swiss and Danish cities, among others. Some larger stations offer a range of services, including bike hire and repairs. By cooperating instead of competing, public transport and cycling can deliver environmental and health benefits.25
Cycling offers an environmentally friendly alternative to park-and-ride, costing less to build, using much less land and creating far fewer traffic problems. Cyclists are more likely to ride to their local stop or station, rather than travelling to a cityedge parking station, as happens with less successful car park-and-ride systems. Bicycle parking can be provided at a wider range of locations, including smaller tram and bus stops where lack of space would prevent the provision of car parking. The role of bicycle access to public transport is likely to be greatest in exurban and rural areas where even the best feeder bus networks may not satisfy everyone. Bike parking will be simpler and cheaper to provide in regions where cycling is a significant mode of transport, because these are the places where cyclists are least likely to ride very expensive machines that must be stored in secure lockers.
Public transport and cycling do not show the simple and powerful positive feedback found with walking, but they can support one another if the overall policy environment is designed to promote mode shift from the car to more sustainable modes. The question of the policy environment brings us full circle, back to politics.
Notes
1 DfT (2007/2008).
2 Hass-Klau (2003, 2007).
3 Ott (2002, pp78–79). Interestingly, he was speaking at a seminar on ‘alternatives to congestion charging’.
4 Haq et al (2008, p563).
5 Morton and Mees (2005).
6 Swiss Federal Statistical Office, 2000 Census ‘pendler’ tables A2E, Kantons Zurich (1), Graubunden (18) and Schaffhausen (14).
7 de Bruijn et al (2009, p193).
8 Mees (2000, ch. 7).
9 ‘The White bike comes full circle’, Independent, 16 July 2000.
10 Heath and Potter (2005, p294).
11 Krag (2002, p235).
12 Mees et al (2007, p6, table 1.8).
13 Apel and Pharoah (1995, p258, table 37). The low walking figures suggest that the surveys in question may have understated walking by omitting shorter trips.
14 Munster (2008, p190, table 6.5).
15 www.muenster.de/stadt/exwost/portrait_e.html (accessed 30 August 2009).
16 ONS (2003, table KS-15 – excluding ‘working from home’).
17 Hass-Klau (2003, p191).
18 Bohle (2002, pp212–213).
19 Pharoah (2003, p365).
20 Mees et al (2007, p13, table 2); DfT (2007/2008).
21 City of Auckland (2007, p3, fig. 1).
22 Pharoah (2003, p364).
23 Fisher (2002, p312).
24 TTC meeting 1874, 20 September 2006: ‘Report – Bicycle racks on buses: results of pilot project’; minutes, item 25.
25 McClintock (2002, chs 12 and 13); Martens (2004).