IELTS Academic Reading Practice Test 40

The steam engine was in use for nearly a century before anyone succeeded in putting it on wheels. The first practical steam engines, developed in Britain in the early eighteenth century, were huge stationary machines used to pump water out of mines. They worked at low pressure and were far too large and heavy to move themselves, let alone pull a load. Meanwhile, coal and iron were carried from mines to rivers and ports along wagonways, simple tracks on which horses pulled wagons. It was only when engineers learned to build compact engines using steam at high pressure, around the end of the eighteenth century, that a self-propelled machine became a realistic possibility.

The first locomotive to haul a load along rails was built by the Cornish engineer Richard Trevithick. In February 1804, at the Penydarren ironworks in South Wales, his engine pulled ten tons of iron, five wagons and seventy men a distance of nearly ten miles, taking a little over four hours. The journey had been arranged to settle a bet made by the owner of the ironworks, Samuel Homfray, who had wagered that a steam engine could do the work of horses. Homfray won his bet, but the experiment was not a lasting success. The locomotive was so heavy that it repeatedly broke the brittle cast-iron rails, and it was soon converted into a stationary engine.

Over the following two decades, engineers gradually overcame these problems with stronger rails and better designs. At the Middleton Railway in Leeds, locomotives began working regularly in 1812, using a toothed wheel that engaged with a rack beside the track. In north-east England, George Stephenson built engines to carry coal from collieries, and in 1825 his Locomotion No. 1 hauled the opening train of the Stockton and Darlington Railway. Even so, horses remained essential on this line for some years, and in its early period many of its trains, including those carrying passengers, were drawn by horses rather than engines.

The decisive moment came in 1829. The directors of the new Liverpool and Manchester Railway had still not decided whether their trains should be pulled by locomotives or hauled along the line by cables wound by stationary engines. To settle the matter, they organised a public competition at Rainhill, near Liverpool, offering a prize of £500 for the best locomotive. Several machines entered, but most broke down.

Rocket, built by George Stephenson and his son Robert, was the only competitor to complete the trials, reaching speeds of almost 30 miles per hour. When the railway opened in 1830, it became the first intercity line to rely entirely on steam locomotives for its trains. The opening day was marred by tragedy, however, when the politician William Huskisson was struck by Rocket and died of his injuries.

Rocket's success depended on combining several features that would remain standard for more than a century. Coal burned in a firebox at the rear of the engine, and the hot gases from the fire were led through many narrow tubes running the length of the boiler. Because these tubes greatly increased the area in contact with the water, the boiler produced steam far more quickly than earlier designs.

The steam was fed into cylinders, where it pushed pistons backwards and forwards, and this motion was transmitted to the driving wheels by connecting rods. Finally, the used steam was released through a blastpipe into the chimney. The rush of exhaust created a strong draught that pulled air through the fire, so that the harder the engine worked, the more fiercely its fire burned.

Railways spread rapidly across Britain, continental Europe and North America, and later throughout the world, transforming trade, travel and even the measurement of time, as towns adopted a standard railway time in place of local solar time. Locomotives grew larger and faster. On 3 July 1938 a streamlined British locomotive named Mallard reached 126 miles per hour, a figure that remains the generally accepted speed record for steam. Yet steam locomotives were never efficient machines. Only a small fraction of the energy in their fuel was converted into useful work, and they needed frequent stops for coal and water, as well as large crews to operate and maintain them.

These weaknesses eventually proved fatal. During the twentieth century, diesel and electric locomotives, which were cleaner, more efficient and cheaper to run, replaced steam on most of the world's railways. British Rail withdrew its last steam locomotives from regular main-line service in 1968, although China continued to build them until late in the century. Today, steam engines survive mainly on heritage railways run by enthusiasts, where they attract large numbers of visitors keen to experience the sights, sounds and smells of the machines that created the modern railway.

The mosquito Aedes aegypti is one of the most dangerous animals in the world. Also known as the yellow fever mosquito, it is well adapted to life in towns and cities, where it rests indoors and lays eggs in clean, still water. It carries the viruses that cause dengue, Zika, chikungunya and yellow fever, and dengue alone infects millions of people every year across tropical and subtropical regions. There is no specific treatment for the disease, and controlling the mosquito has proved extremely difficult. It breeds in small containers of water around homes, such as flowerpots, buckets and discarded tyres, and it bites during the day. In many places, populations have also developed resistance to commonly used insecticides, reducing the effectiveness of spraying campaigns.

In recent decades, researchers have pursued an unusual alternative: infecting mosquitoes with a bacterium. The bacterium was first identified in 1924 in the reproductive tissues of a mosquito and was later named after the American pathologist Simeon Burt Wolbach. Wolbachia lives inside the cells of insects and is remarkably widespread; studies suggest that it is present in around 40 per cent or more of insect species. It is passed from mothers to their offspring through their eggs. Aedes aegypti, however, does not naturally carry Wolbachia. To introduce it, scientists took a strain of the bacterium from fruit flies and injected it into mosquito eggs, a painstaking process that required many thousands of attempts before a stable infection was established.

Wolbachia affects mosquitoes in two important ways. First, it reduces the ability of viruses such as dengue to multiply inside the insect, so that an infected mosquito is much less likely to transmit disease when it bites. Second, it manipulates reproduction through an effect known as cytoplasmic incompatibility. When a male carrying Wolbachia mates with a female that lacks it, her eggs fail to hatch. Infected females, on the other hand, produce infected young whether or not their partner carries the bacterium. This gives infected females a reproductive advantage, because they can successfully mate with any male, while uninfected females cannot use sperm from infected males. As a result, once enough infected mosquitoes have been released, Wolbachia tends to spread through the wild population and to persist without further releases.

The first field releases took place in 2011 in two communities near Cairns, in northern Australia, led by a team under the biologist Scott O'Neill, whose programme later became the World Mosquito Program. The programme used the wMel strain, originally taken from the fruit fly Drosophila melanogaster, which had been shown to block virus replication without greatly shortening the mosquito's lifespan. Within weeks of the releases ending, the great majority of mosquitoes in both areas carried Wolbachia, and monitoring in later years showed that the infection had remained at high levels. Releases were subsequently extended to other towns in northern Queensland, and the programme expanded to countries in Asia and Latin America.

The strongest evidence came from a trial in the Indonesian city of Yogyakarta, carried out between 2017 and 2020. The city was divided into 24 clusters, and Wolbachia mosquitoes were released in 12 of them, chosen at random, while the other 12 served as a comparison. Researchers followed the health of more than 8,000 residents, testing blood samples when they developed fever. The results, published in the New England Journal of Medicine in 2021, showed that dengue cases were 77 per cent lower in the areas where releases had taken place, and hospitalisations for dengue were 86 per cent lower. The method was effective against all four serotypes, or types, of the dengue virus.

A second strategy uses the same bacterium in a different way. Instead of replacing wild mosquitoes with infected ones, it aims to suppress the population by releasing only males carrying Wolbachia. Male mosquitoes do not bite, and when they mate with wild females, the resulting eggs do not hatch, so mosquito numbers fall over time. Singapore has used this approach in a programme run by its National Environment Agency. The Singapore programme, known as Project Wolbachia, has expanded from a small pilot to cover a substantial share of the city-state's public housing blocks. For it to work, very few females must be released by mistake, since infected females would allow Wolbachia to spread instead. Meanwhile, production is being scaled up elsewhere: a factory in Curitiba, Brazil, which began operating in 2025, is intended to produce around 100 million mosquito eggs a week.

Wolbachia is not a complete solution. Some strains of the bacterium become less abundant inside mosquitoes when temperatures are very high, which may weaken their effects in the hottest regions. Even when infection levels are high, mosquitoes can still breed in containers, and people must continue to empty standing water around their homes. Producing and releasing mosquitoes on a large scale is costly, and because the method involves deliberately releasing insects into neighbourhoods, it depends on the support of the communities concerned. Scientists also continue to monitor whether viruses might eventually adapt. Supporters point out, however, that the technique does not involve altering the mosquito's own genes, which may make it more acceptable to the public than some other approaches.

The use of Wolbachia is not limited to dengue. Because Aedes aegypti also transmits Zika and chikungunya, reducing its ability to carry viruses should also reduce the spread of those diseases, and early evidence supports this. Programmes are now measuring whether the approach can protect communities against multiple viruses at once. In addition, scientists are exploring whether Wolbachia can be transferred to Anopheles mosquitoes, which transmit malaria, although this work is at a much earlier stage. In the longer term, the technique may become part of routine disease control, but for the moment it remains an experimental tool that requires careful evaluation wherever it is used.

In 2012 the New York Times declared it the year of the MOOC, the massive open online course. Free courses from some of the world's most prestigious universities were suddenly available to anyone with an internet connection, and commentators predicted that traditional higher education would be transformed, or even made obsolete. A decade later, that revolution has plainly not happened, and critics frequently point to one statistic as proof of failure: the tiny proportion of people who finish the courses they sign up for. I want to argue that this statistic is widely misunderstood – but also that, read correctly, it reveals a different and more important problem.

The movement grew out of an experiment at Stanford University in 2011, when the computer scientists Sebastian Thrun and Peter Norvig offered their artificial intelligence course free online and around 160,000 people enrolled. Thrun went on to found the company Udacity, and in 2012 two other Stanford academics, Daphne Koller and Andrew Ng, launched Coursera, a platform offering courses from many universities, while Harvard and MIT created a non-profit equivalent, edX. Within months, millions of people had registered for courses on subjects ranging from philosophy to machine learning.

Completion rates, however, were low from the start, and they did not improve. In 2019 Justin Reich and José Ruipérez-Valiente published an analysis in the journal Science of data from 5.63 million learners on Harvard and MIT courses over six years. They found that completion rates had remained low throughout the period and had even fallen. In the most recent year they studied, only around 3 per cent of all participants completed their courses, and many of those who registered never returned after the first week.

At first sight these figures are damning, but they need to be interpreted with care. Registering for an online course costs nothing and takes seconds. Many people sign up out of curiosity, intending only to watch a few lectures or to learn about one particular topic, much as they might browse a book in a library. Judging such courses by the proportion of people who finish them is rather like judging a library by the number of books that are read from cover to cover. A learner who watches three lectures and gains what she wanted has not failed; she has simply used the course differently.

This argument, though, can be taken too far, because it was not only casual visitors who failed to finish. Reich and Ruipérez-Valiente found that completion rates were low even among learners who said they intended to complete, and they declined among those who had paid for a certificate. The more serious problem concerns who benefits. MOOCs were promoted as a way of opening elite education to people who had been excluded from it, yet surveys repeatedly found that most participants already held a university degree. Free access, it turned out, was not the same as equal access.

The clearest warning came from an early attempt to use online courses for students who needed them most. In 2013 Udacity and San Jose State University in California ran a pilot in which students took introductory mathematics and statistics courses online for credit. Pass rates were lower than in the university's traditional classes, and the experiment was soon scaled back. Later that year, Thrun publicly acknowledged that his company had a lousy product, and Udacity moved its focus towards vocational training for professionals. The lesson seems to be that self-paced courses work best for people who are already confident, disciplined learners, and least well for those who struggle.

That does not mean online learning is doomed to fail such students. Courses that set regular deadlines, group learners into cohorts and provide contact with real tutors tend to retain far more of their participants. Some universities have also used online delivery to make full degrees cheaper. In 2014 the Georgia Institute of Technology launched an online master's degree in computer science at a small fraction of the usual cost, and it has since enrolled many thousands of students. Such programmes are less open than the original MOOCs, but they offer something more valuable: a recognised qualification and the structure needed to earn it.

Online courses, then, have been neither the revolution their champions promised nor the failure their critics describe. Low completion rates are a poor measure of their success, because many learners never intended to finish. The better question is whether courses help their learners achieve their own goals – and whether they reach people who would otherwise have had no opportunity to learn at all. On that second test, the record so far is disappointing, and improving it will require investment in support and teaching, not simply in more free content.