Page 1


ISSN 1729-830X

VOLUME 2 • NUMBER 3 • 2006 R20 incl. VAT


Cover stories


Save the hornbills Alan and Meg Kemp Global threats – and solutions that work Power to the nation 3


■ Making power go further

Tsholo Matlala

Getting the most out of what we’ve got

■ Protecting transmission

Hein Vosloo and Philip Frost

New ways to track fires under power lines

Contents Volume 2 • Number 3 • 2006


■ Electricity from the Sun

Chris Engelbrecht

Affordable power from sunlight

Regulars 5

■ Making solar power feasible


Vivian Alberts

Pioneering South African science

Science news Mars rocks; Signs of life on Saturn’s moon? The unpredictable curved ball (p.31) • Galaxy clusters old and far; Greenland’s melting glaciers; Responding to the meltdown: US evangelical Christians, Psychology; Take group exercise for health (p.35) • Where are the data on avian flu? (p.42)

Election-night forecasting Jan Greben Mathematical models that predict results


A termite tale of climate change Martin Pickford

Fact file South Africa’s energy needs


Careers Work to supply the country’s power


Viewpoint Is evidence overrated? Jonathan Jansen


Fossil clues to the ways that savanna turned into desert

The S&T tourist Pilgrimages to our origins Go back in time at the Cradle and the new Origins Centre


Your Questions answered Rainy weather – Chris Reason and Warren Tennant

Features 20

Maths and its meanings Doug Wilson The multifaceted world of numbers


What’s up in the night sky? Case Rijsdijk Looking up into the life of stars


Books The Drumcafé’s Traditional Music of South Africa • and other titles


Crossword puzzle


Measuring up


Letters to Quest Facts for Africa; good writing


ASSAf news


Diary of events


Subscription form • Back page science

Quest 2(3) 2006 

Science for South AfricA

ISSN 1729-830X

Volume 2 • Number 3 • 2006 r20 incl. VAt

Measure for measure


AcAdemy of ScIeNce of South AfrIcA

Juvenile male southern yellow-billed hornbill (Tockus leucomelas) with brown tinged first plumage, eye paler yellow and bill smaller than an adult, but ready to start life independent of its parents. Photograph: Alan Kemp SCIENCE FOR SOUTH AFRICA

ISSN 1729-830X

Editor Elisabeth Lickindorf Editorial Board Wieland Gevers (University of Cape Town) (Chair) Graham Baker (South African Journal of Science) Anusuya Chinsamy-Turan (University of Cape Town) George Ellis (University of Cape Town) Jonathan Jansen (University of Pretoria) Colin Johnson (Rhodes University) Correspondence and The Editor enquiries PO Box 1011, Melville 2109 South Africa Tel./fax: (011) 673 3683 e-mail: (For more information visit Business Manager Neville Pritchard Advertising and Neville Pritchard subscription enquiries PO Box 130614 Bryanston 2074 South Africa Tel.: (011) 781 8388 Fax: (011) 673 3683 Cell: 083 408 3286 e-mail: Copyright © 2006 Academy of Science of South Africa

etting ahead has a great deal to do with identifying gaps and bridging them. “If you want to keep winning you’ve got to keep improving” was how the England coach described the way forward after his cricket team won their all-important Ashes test series against Australia last year. The same goes for any sort of progress. But knowing if you’re improving means quantifying the situation, then finding ways to measure developments in a meaningful way. This issue of QUEST is filled with research that moves ahead on the basis of what the numbers say. Determining whether or not a type of bird or animal is endangered and in need of special care, for instance, means knowing how many species there are (and were in the past), where they live, the sizes of the populations, and how successfully they reproduce. The story of the conservation of hornbills in Africa and Asia (p. 12) is a case study for assessing the health of the environment and finding creative ways to save species. Extinction means failure; stable or increased numbers are measures of success. South African city life keeps being disrupted by power failures or ‘outages’. We highlight three attempts to overcome the problem. The first offers ways to use the supply that’s available as efficiently as possible by managing consumer demand better (p. 3). The next involves reducing the levels of disruption. Fires under power lines interrupt the workings of the country’s transmission system. To put fires out quickly we need to know where they are in real time. A new method of reading and analysing satellite data has been designed to track fires and minimize the destruction they cause (p. 6). Looking to the future, the world is in search of renewable energy sources. Pioneering South African work promises to make solar power as affordable as – and far cleaner than – electricity from coal-fired power stations (p. 8). Whether it’s predicting final election results from mathematical modelling of early voting figures (p. 23), determining the Earth’s climate before the polar ice caps formed (p. 28), or tracking the life and death of stars over millions of years (p. 38), measurement is the heartbeat of our understanding. So it helps to know about the many ways in which the underlying mathematics works (p. 20). But gathering and interpreting the evidence, then choosing what to do with it are equally crucial (p. 33). The philosopher Immanuel Kant famously said, “Science is organized knowledge. Wisdom is organized life.” This issue of QUEST celebrates the work of scientists who have been doing their sums.

Published by the Academy of Science of South Africa (ASSAf) PO Box 72135, Lynnwood Ridge 0040, South Africa (011) 673 3683 Permissions Fax: e-mail: (011) 781 8388 Back issues Tel.: Fax: (011) 673 3683 e-mail: Subscription rates (4 issues and postage) (For subscription form, other countries, see p.48.)

South Africa (incl. VAT) • Individuals – R80.00 • Students/Schoolgoers – R40.00

Production Pritchard Productions cc Design and layout Creating Ripples Printing Paradigm All material is strictly copyright and all rights are reserved. Reproduction without permission is forbidden. Every care is taken in compiling the contents of this publication, but we assume no responsibility in effects arising therefrom. The views expressed in this magazine are not necessarily those of the publisher.

2 Quest 2(3) 2006

Elisabeth Lickindorf Editor – QUEST: Science for South Africa Join QUEST’s knowledge sharing activities Write letters for our regular Letters column – e-mail or fax your letter to The Editor and win a prize. (Write QUEST LETTER in the subject line.) ■ Ask science and technology (S&T) questions for specialist members of the Academy of Science to answer in our regular S&T Questions and Answers column – e-mail or fax your questions to The Editor and win a prize. (Write S&T QUESTION in the subject line.) ■ Inform readers in our regular Diary of Events column about S&T events that you may be organizing. (Write QUEST DIARY clearly on your e-mail or fax and provide full and accurate details.) ■ Contribute if you are a specialist with research to report. Ask the Editor for a copy of QUEST’s Call for Contributions (or find it at, and make arrangements to tell us your story. To contact the Editor, send an e-mail to: or fax your communication to (011) 673 3683. Please give your full name and contact details. ■

Tsholo Matlala explains the mechanics of Demand Side Management, a new way to make the most of South Africa’s electricity supply.


upplying electricity is a demanddriven activity. As with water supply, the product must constantly be available at the consumer’s demand. And while it must be there in sufficient quantity, reserves must not be excessive, because this wastes resources and drives up prices. So the supplier has clearly defined ‘supply side’ targets. On the ‘demand side’, the electricity supplier also has many requirements to meet. Millions of developing world citizens are impatiently demanding access to electricity and the improved quality of life that it brings. Commerce and industry is demanding greater efficiency and reliability, controlled price rises, and improved quality of supply – that is, constant supply (with fewer unplanned outages) and constant frequency (without dips or peaks that can affect sensitive equipment). The environmental sector is demanding that utilities reduce their impact on the environment1. The sum of all these pressures means that electricity supply companies have difficulty in deciding who their most important stakeholders are, let alone how to appease them. One innovative solution, however, goes a long way to meeting most objectives – Demand Side Management (DSM). Instead of tying up large amounts of capital in power stations, ways can be found to ‘manage the demand’ – that is, to persuade consumers to use less electricity, or use electricity more efficiently. It’s a new and powerful alternative to adding more generation capacity. While not replacing the building of new capacity, DSM can moderate the rate of demand growth, allowing the supplier to delay or extend the building of new capacity, optimizing the country’s scarce resources, and to assist in achieving the environmental goals.

amount of energy is used, it does not add to the total at peak time when facilities are strained. This lessens the pressure on Eskom to build new power stations. Also, ‘filling up’ the valleys on the graph enables Eskom to sustain a more consistent level of output, thus improving the efficiency of the supply operation. At present, load shifting has been applied mostly by larger power users such as municipalities, mines, and water supply utilities. Energy used at off-peak times is billed at a lower tariff than energy used at peak times. The user needs a special ‘time of use’ meter that records the time of

Top: Jo’burg lights up. Large office blocks use huge amounts of power, but they can cut electricity bills through energy-efficient lighting, heating, and air-conditioning. Above: Factory operators are being encouraged to use energy efficiently and, where they can, to move demand out of peak periods.

day when energy is used; this allows the supplier to bill the user at the appropriate rate for that time. Underground air cooling systems in mines have benefited significantly from this strategy. Success requires a real-time energy management system and an on-site

1. Editor’s note: Eskom’s 2005 annual report gives details of the facility’s research and development in several areas of alternative energy power sources (relating, for example, to wind, solar, and fuel cells and biomass gasification, as well as the feasibility of ocean energy).

Quest 2(3) 2006 3

▲ ▲

How it works There are two basic DSM strategies: load shifting and load reduction. Load shifting makes use of daily periods of lower demand – the ‘valleys’ on the ‘Peak load reduction’ graph (see next page). It encourages consumers to move their load to off-peak times, when there is spare capacity on the system. Although the same

Ma k i ng power go f u r t her

information system, and significant Peak load reduction savings are possible. Applying load shifting at Harmony gold mines in the Free State2, for example, saved hundreds of thousands of rands a year. As small and medium consumers of electricity discover the savings that larger consumers are making, creative proposals are emerging that offer load shifting opportunities. The other main DSM strategy is load reduction. Here, consumers are shown how electrical products and appliances can achieve savings while using less energy (for example, by installing more efficient lighting and air-conditioning systems, by insulating geysers, and by Above: Demand Side Management using household and aims to reduce electricity demand – particularly at peak periods – by kitchen appliances the equivalent of the electricity efficiently)3. supplied by one large power Reducing the station over 20 years. amount of energy Left: Farmers can save by consumed has installing energy-efficient an immediate controls on their irrigation and permanent systems. effect, and Below: Power-wise lighting effectively for Pick ‘n Pay. postpones the construction of a base load power station. In Eskom’s case, long-term, cumulative energy savings allowed a whole power station project to be removed from the building schedule, saving the country billions of rands4. Electricity-saving savvy 1. Save hot water ■ use less hot water by taking a short shower rather than a bath ■ fit low-flow showerheads ■ set geyser thermostats below 60ºC and insulate the geyser and hot-water pipes ■ use cold water when you can and make sure your hot-water taps don’t leak 2. When you use electrical equipment, use ■ economy cycles and full loads 3. Dry your washing in the sun rather than in a tumble dryer 4. When you cook ■ use a pressure cooker for foods that need long cooking times ■ keep oven doors closed till food is cooked ■ bring foods to the boil on ‘high‘ setting, then reduce to ‘simmer’ till cooked ■ choose pots and pans that cover the stove plates; keep stove-plate reflectors clean ■ defrost food in the refrigerator (rather than in the microwave) ■ cook small or medium quantities of food in

4 Quest 2(3) 2006

a microwave oven and larger portions in a conventional oven 5. When you use the refrigerator ■ let food cool down before putting it into the fridge ■ don’t open the door unnecessarily ■ check for tight door seal; defrost regularly 6. To save on lighting ■ install movement-sensor lamps and fluorescent bulbs ■ switch off incandescent lights when you don’t need them 7. To save on heating ■ close windows and doors and make sure they’re sealed to avoid draughts ■ insulate ceilings and cover concrete floors with carpeting ■ don’t overheat rooms; switch heaters off when nobody’s there ■ use electric blankets (but turn them off when you’re in bed); infrared heaters are most efficient; oil heaters are safest.

Successful load-reduction projects include lighting optimization and airconditioning system modification. With its countrywide chain of super- and hypermarkets, Pick ’n Pay realized that it could save on energy bills by converting its lighting, heating, ventilation, and air conditioning (HVAC) systems to energy-efficient ones5. In a project partly funded by Eskom’s DSM department, the company installed energy-efficient lighting in all its new stores and did a retrofit on 97 established outlets. This saves approximately R1.5 million annually in reduced energy bills, and the capital costs of the project will be recouped in just over two years. After that, all savings are a direct improvement to Pick ’n Pay’s bottom line. DSM is a strategy with great potential. So far, Eskom has made significant energy savings and customers have benefited from lower energy bills. But many more organizations could benefit and more sectors could participate. There is scope for a wider application of these principles. All it takes is some creative thinking and willing participants. Eskom will provide the necessary support. ■ Tsholo Matlala qualified as an electrical engineer and joined Eskom in the late 1990s. He is the Eskom Energy Services Manager and is responsible for funding and implementing the Energy Efficiency & Demand Side Management Programme in South Africa. For more information, phone the DSM Programme at (011) 800-2776 or visit 2. Modelling the results of installing a real-time energy management system at the Harmony mines projected a conservative 13.5 MW being shifted out of the evening peak demand period, equivalent to a financial saving of R1.35 million per year. 3. In line with international practices, Eskom is currently investigating and piloting the local Homeflex – a version of residential peak and off-peak tariff structure – and a strategic decision will be based on the outcome of this study. 4. A new coal-fired power station costs about R30 billion. After preparatory phases, including the environmental impact assessment and various approvals, construction takes 5–8 years. 5. The Pick ’n Pay project involved installing the latest technology in electronic lighting control gear, so far less electricity is now used to start up and run the fluorescent tubes in the overhead lighting than before. Nearly 46 000 of these devices are being installed in Pick ’n Pay stores across the country. In addition to saving energy, they develop only minimal heat, which helps to save on air conditioning and promotes longer lamp-life.

Q Fact file Sout h Africa's energy needs Demand and supply

Timeframe for new capacity outlook

More light, less power: CFLs

The latest available figures (for 2004) are published in Eskom’s Energy-saving fluorescent 2005 annual report. lamps use 20% of the ■ Total energy consumption for electricity of normal household South Africa in 2004 was just globes but emit around five over 206 799 gigawatt hours times the light per watt. (GWh) A fluorescent lamp uses ■ Of this, industrial, mining, and electricity to excite mercury commerce accounted for 77% vapour in argon or neon gas, ■ Eskom’s nominal capacity is which is enclosed in a glass 42 001 MW and its net maximum tube, resulting in a plasma that capacity is 39 810 MW (The produces short-wave ultraviolet difference reflects auxiliary power (UV) light. The inside of the glass consumption and reduced tube of the fluorescent lamp Existing system (including Build pumped storage plants Required capacity (15% reserve margin) capacity caused by the age of the has a coating of phosphors decommissioning) plant and/or low coal quality.) Build open cycle gas turbine Forecast peak demand before demand that emits visible light when Total imports (MW) plants side management (DSM) ■ South Africa’s electricity peak excited by UV light. Phosphors Return to service Build base load plants Peak demand after DSM demand has been rising at are transition metal compounds Simunye plants around 4% a year over the past or rare earth compounds of decade various types. Depending on the building new power stations and by returning ■ Residential users account for composition of the phosphors, different colours of three mothballed power stations to service 17–20% of off-peak consumption; this rises to light can be created. (Komati, Grootvlei, and Camden power stations) 30–35% during peak times (that is, 7:00–10:00 A compact fluorescent lamp (CFL) burns at a ■ It is seeking to reduce consumption, and 18:00–20:00), mainly because of energy much lower temperature than an incandescent particularly at peak periods, by implementing consumed for cooking, lighting, and heating household globe and converts 80% of electricity the DSM programme. Its business objective water and space into light and 20% into heat. An incandescent is to save 153 MW a year. (Savings in 2004 ■ The residential peak demand consumption of globe works the other way around, converting exceeded 197 MW – equivalent to the 11 799 MW in 2004 was equivalent to the 80% of electricity into heat and only 20% into consumption during peak periods of a town output of three large power stations. light, thus wasting electricity. such as Roodepoort and a monetary saving Standard fluorescent tubes have been used of R1.2 billion.) It aims to save 4 255 MW Environmental concerns for many years in shops, offices, workshops, over 20 years – this equates to the generating and elsewhere. We know them as the sources Environmentalists are concerned about emissions – capacity of one six-pack power station. of cold, blue-white lights that flicker and often visible ‘particulate’ emissions and invisible ones, such give off an annoying buzz – which makes many as carbon dioxide (CO2) and sulphur dioxide (SO2), How DSM reduces consumption consumers prefer to use incandescent lamps. as well as excessive consumption of water in the First introduced in the 1980s, CFLs are now Many customers – residential ones, for instance process of generating electricity, and liquid effluent. more reliable, cheaper, and available in many – have not so far paid different rates for using ■ Saving 1 kWh means saving 0.9 kg of CO2 shapes and sizes. An electronic version has electricity at peak and off-peak times and, for that would have been released into the replaced the old magnetic ballast, eliminating the most part, it is large consumers that are atmosphere, as well as 1.29 litres of water. the buzz and the flickering; and the quality of increasingly affected by multiple tariff rates. Taking ■ The Demand Side Management (DSM) the phosphor coating has improved, allowing on a commercial or industrial DSM intervention programme achieved a reduction of manufacturers to produce CFLs with different also offers the potential to renegotiate rates. 359 510 tonnes of CO2 emissions and colour temperatures, from a cold blue-white to After completing several DSM interventions, 515 298 kilolitres of water in 2004. a yellower colour reminiscent of incandescent Harmony’s Masimong4, Harmony3, and Bambani lamps. You can even buy coloured CFLs in blue, Gold mines changed to Time-of-Use metering and, Future prospects green, red, or yellow. Yellow light does not attract by using the best operating schedule, they managed These differ for peak and off-peak demand. insects, so yellow CFLs are good for outdoor use. to shift 13.5 MW of load out of peak hours. ■ Peak demand is fast approaching the capacity There are CFLs for virtually all home and of so-called peaking plants to meet the need. At industrial applications. DSM has worked hard to Underground air cooling in mines the current rate of growth, it is projected that reduce the cost of many energy saving devices, the present excess capacity for peak power will South Africa’s gold mines are constantly cooled by including CFL globes, whose price has fallen to run out next year – most likely in the winter, the circulation of cold water, which is chilled on about R15 each. when the peak demand spikes are highest the surface. The water ends up in a dam at the Tips on using CFLs: ■ During the 2004 financial year, Eskom’s net bottom of the mine and is then pumped up again ■ If you don’t switch a good-quality CFL on and maximum generating capacity was to the refrigeration plant for chilling in a continuous off more than 8 times within a 24-hour period, 39 810 MW, while the peak demand on the process that consumes vast amounts of electricity. it can last for 6 000 hours integrated system was 34 195 MW Savings are achieved by ■ CFLs become mildly warm during operation ■ Off-peak demand is met by supply from base ■ Timing the pumping so as to consume more and can be touched without risk of load plants. If demand grows at the current electricity in cheaper off-peak periods burning rate, it is expected that Eskom will run out of ■ Making more efficient use of Eskom tariffs, ■ They’re best for often-used areas, spare capacity by 2010. which reward load-shifting such as kitchens, passages, ■ More efficient control of the pumping bathrooms, and dining What’s being done? operations, which includes installing rooms, and for outside sophisticated equipment and variable-speed and security lighting, Eskom is tackling the increasing demand for pump motors that reduce the surge in power which often burns electricity. consumption when a pump is started. ■ It is increasing electricity supply capacity by throughout the night. ■

Quest 2(3) 2006 5

P r ot ec t i ng t r a n s m i s s ion


ires in southern Africa are as old Hein Vosloo and Philip Frost describe the as the oldest plant material that real time early-warning system developed in burned readily enough to fuel them when lightning struck, and our vegetation South Africa for monitoring the fires that disrupt is well adapted to fire. The arrival of the country’s electricity transmission. humans and their fire-making ability must have increased the number of fires in the area. Most modern fires are caused by people. We know this because most fires in the interior of South Africa occur during winter, when there is no lightning. Rangeland fires that burn under Eskom’s transmission system present a problem, not only to customers but also to the equipment that carries electricity through the country’s 28 000 km of transmission power lines. A fire under a power line causes a short circuit between the live conductor and the earth, which disrupts, or modifies, the sinusoidal shape of the alternating current1 and modifies the flow of Causes of disruption: 2005 have been found to be orbiting satellites. The system involves electricity. Very expensive two types of satellite data: those detected the most reliable method high-voltage equipment, Pollution 2% by the Moderate Resolution Imaging of preventing flashovers, such as transformers and Other Veld fires Spectroradiometer (MODIS) sensor on and they also have circuit breakers, are exposed 21% 20% NASA’s Terra and Aqua satellites passing ecological benefits for to such modification, overhead, and those available from the plant species in which reduces their life geostationary weather satellites2. the RoWs, provided expectancy and duty the timing of these cycle. These short MODIS data were available, because the burns is carefully circuits, or flashovers, Department of Agriculture had purchased controlled. affect Eskom’s a MODIS antenna in early 2003, situated at The facility has customers too, in Hartebeesthoek and used for agricultural also pioneered a particular continuous applications. These data provide for highfast-reaction fireprocess factories, such resolution fire tracking four times a day. Lightning 21% tracking system to as paper mills, nylon They are widely used around the world enable it to put out spinners, and even steel for this purpose, and the resolution is high Birds fires quickly during the manufacturers. enough to pick up active fires that are as 32% Cane fires Managing the problems small as about 0.25 ha in area. fire season. 4% caused by fire is not easy. It Eskom, however, requires fires to be ‘Eye in the sky’ means managing the vegetation reported within 15 minutes, so another surveillance that fuels them; it also means making sure satellite was needed. Philip Frost’s Putting out fires under power lines quickly that, when fires are started, they are put innovative method of analysing the data enough for them not to cause damage out promptly. provided by a weather satellite allowed such means being able to spot them early Eskom has been involved in several an instrument to be used, in addition, for enough to alert the fire-fighters before the initiatives to reduce the amount of fuel under tracking fires. The resolution of data from fires go out of control. power lines – including cutting the grass or weather satellites is much coarser than that During early 2004, Eskom and the trimming the shrubs that grow underneath from MODIS, so a fire needs to be about Satellite Applications Centre of the CSIR at them. More recently, controlled burns in the 5 ha in size before it can be detected, but it Hartebeesthoek launched a project to track Right of Way (RoW) areas (or servitudes) has the advantage of allowing fire reports to active fires under power lines, using earthbe generated every quarter of an hour. 1. Alternating current, as the name indicates, changes from a positive voltage to a negative voltage many times a second, and this Fire hot spots are immediately published alternation (or oscillation) creates an electrical wave with a sinusoidal shape. Appliances that work on alternating current are designed in near real time on a web page (which to operate on the voltage so provided. In the South African electrical system, the voltage changes 50 times per second (that is, at a frequency of 50 Hz). When the sinusoidal shape of the voltage of uninterrupted flow of electricity is distorted or modified, the devices is available free of charge to anyone who that draw power from it work less efficiently. needs the information), and a fire that 2. The orbiting times of polar orbiting satellites are too slow to provide rapid, 15-minute updates of satellite observations, as they pass over breaks out within 2.5 km of a power line any given place on Earth only four times a day. Geostationary satellites, such as weather satellites, also orbit the Earth, but their orbit is synchronized with the rotation of the Earth, so the satellite appears stationary above the Earth. The scanners on board these satellites is automatically reported via cellphone scan the area of interest every 15 minutes, albeit at a coarser resolution than those of the polar orbiting satellites. This is why only large SMS to Eskom’s National Control Centre fires of about 5 ha in size can be detected by the weather satellites, in contrast to the 0.25-ha fires that the polar orbiters can detect. The as well as to the individual field staff fact that both types of observation are used in conjunction with one another increases the value of the weather satellite for fire tracking.

6 Quest 2(3) 2006

Partnerships to fight fires

member responsible for the that line. The National Control Centre can then monitor the situation and temporarily switch out any lines that are under threat. Field staff immediately activate fire suppression teams and report the conditions at the site of the fire directly to the control centre. Using the data The system has been in place for two years, and it performs well, even though the data cannot always keep up with the speed with which a fire burns, and even though low-resolution data don’t pick up fires when they are very small. Fires occurring within 2.5 km of Eskom’s power lines Year

No. of fires


3 987


3 905


6 147

With the data gathered from the system, we can now compare the number of fires observed during a specified period in a defined geographical area over the years. All the fires occurring within 2.5 km of Eskom’s transmission lines during the period 1 January–31 December were compared for the three years 2003–2005. Although the

The initial launch of the fire-tracking system was made possible by government’s acquisition of MODIS satellite data, in conjunction with the expertise available at the Council for Scientific and Industrial Research (CSIR) and from institutions abroad, such as the University of Maryland in the USA. Eskom’s need for such a system completed the loop. The government’s Working on Fire (WoF) programme has also been involved. This expanded public works programme, whose goals are poverty relief, job creation, and skills development, trains fire-fighters both in the prevention and in the suppression of fires. To date, it has trained about 1 200, and plans to expand its operations to some 200 fire bases with 5 000 fire-fighters. WoF personnel were used during the fires raging in the Cape Peninsula in December 2005. Close cooperation with WoF is planned for this year, when prescribed burning under power lines will be conducted as well as fire suppression during the fire season. As part of the process to detect fires early, Eskom and CSIR provide access to the satellite system. Fighting fires successfully is best carried out through partnerships. The National Disaster Management Centre and its regional structures, local authorities, and other government departments are all part of the total picture.

number of fires was virtually the same for 2003 and 2004, it rose sharply during 2005. The reason is not altogether clear, but the increase could be due to the combination of plentiful fuel and people behaving carelessly. Being able to compare the occurrence of fires in specified geographic locations for different years now makes it possible for Eskom to assess fire patterns, as well as the effectiveness of its fire-prevention strategies. Plans for the future Upgrades and improvements to the system’s software are in the pipeline for 2006. A new algorithm that will improve the quality of the 15-minute satellite data is being

3. Fires are detected by observing the reflectance of certain wavelengths that exceed a specified value. The same principle guides the use of data from the weather satellite. To improve the quality of such observations, a new algorithm is being developed. It uses change detection, which means that instead of using a threshold value to identify a fire, the values for a particular pixel over time are compared with the most recent value. If it exceeds the average for that pixel by more than a given amount, the software registers the change as a fire or hot spot. This new algorithm is expected to improve the quality of data from the weather satellite observations.

Far left and left (above and below): Prescribed burns. Above (middle): Oily substances in plant material that fuels fires release smoke particles that extend the life of fires and thereby contribute to flashovers, which cause line faults. Above (right): Controlled fire at one of Eskom’s power stations. Below: The MODIS antenna at Hartebeesthoek.

developed3, and a new web site as well. In terms of the National Fire Act (No. 101 of 1998), Fire Protection Associations (FPAs) are being established countrywide and involve every land owner in South Africa. The plan is to extend the SMS service to the fire protection officer for each FPA, to help them to mobilize their fire-fighters as early as possible. South Africa is the first country in which an electricity utility has used the automatic cell-phone SMS system in fire detection, in an approach as near real time as this one. The web site is accessed by interested parties from over the world. ■ Hein Vosloo is the Corporate Servitude Specialist for Eskom’s Transmission Division and is responsible for developing servitude-related policy and research. His master’s degree is from the University of Johannesburg. Philip Frost (also registered with the University of Johannesburg) is a Remote Sensing Specialist at the CSIR Satellite Applications Centre and is responsible for developing satellite-derived products from the MODIS and other mediumresolution satellite sensors. Check the web site yourself by visiting eskom/checkboxes/eskom.htm and consult QUEST, vol. 1(3) 2005, “Managing fires: the science behind the smoke” by Brian van Wilgen.

Quest 2(3) 2006 7

Right and below: Samples of the 15-watt solar panels, manufactured with the new CIGS technology, at the University of Johannesburg during 2005. Photographs: Cyclops

A South African breakthrough opens new possibilities for affordable power from the Sun, says Chris Engelbrecht.

E lec t r i c i t y f r om t he S u n


he hunt for energy sources that offer alternatives to coal and oil is intensifying. War in Iraq and unrest in the Middle East will continue to affect the price of oil, and global climate change has raised the alarm about the adverse effects of greenhouse gas emissions from burning fossil fuels1. When burned, oil and coal produce carbon dioxide (CO2), which raises the temperature of the atmosphere and contributes significantly to the processes associated with climate change. Furthermore, both are limited resources and many experts claim that oil production (measured in barrels per annum) has already peaked, or will do so in the next few years2. Of the many alternative energy sources proposed, solar power has traditionally been seen as one of the most expensive, but also as the most abundantly available

– in fact, the amount of solar radiation reaching the Earth’s surface in one hour equals the total of all the energy consumed by humans in an entire year in forms such as electricity, fire, petrol, and coal (see box below). Were it not for the high cost of solar panels, we would expect everybody to be tapping in to this readily available energy source. Affordable panels The high cost of solar panels comes from the use of high-purity silicon to make them and, during the last 30 years, researchers have hunted for cheaper alternatives. There have been a few promising materials, but none till now has challenged

silicon’s hold on the marketplace. A South African breakthrough in the industrial growth of one of silicon’s contenders, CIGS (or copper-indium-gallium-diselenide), looks likely to knock silicon off its pedestal and make the cost of solar-generated electricity much lower than before. This feat came from Vivian Alberts of the Department of Physics at the University of Johannesburg, whose revolutionary process is currently being commercialized. Initially, solar panels of standard commercial sizes will be made, but the special features of the CIGS cell invite many other niche applications in future.

A solar calculation How many seconds of solar radiation that reaches the Earth’s surface would be needed to satisfy South Africa’s electricity needs? South Africans use approximately 200 billion kilowatt hours (kWh) of electricity a year. The Earth’s surface area is approximately 500 trillion m2. An average of 175 W of solar radiation makes it through the Earth’s atmosphere to every square metre of the actual surface (taking into account daylight hours and seasons of the year). Answer: About 8 seconds of solar radiation received at the Earth’s surface could supply energy equal to our annual electricity consumption. (Note: South Africans comprise about 0.7% of the world population.) Calculation: (175 W per m2 ) x (500 trillion m2) x (8 seconds) = 7 x 1017 joules of energy, which is approximately equal to South Africa’s annual electricity use of 200 billion kWh (one kWh = 3.6 x 106 joules; 200 billion kWh = 2 x 1011 x 3.6 x 106 joules = 7.2 x 1017 joules).

8 Quest 2(3) 2006

1. In early February 2006, for instance, the South African government adopted a biofuels policy to reduce oil imports. Next year, petrol mixed with 10% ethanol is to be available at the fuel pumps. Ethanol (which can be made from sugar-cane, sugar-beet, maize, sorghum, and agricultural waste products) is being produced in Brazil at between US$28 and US$32 a barrel, compared with current oil prices of around US$60. 2. Sophisticated models (such as the ‘Wopac’ model) predict that global oil production will peak at about 81 million barrels per day during 2006 or 2007, followed by unavoidable decline (due to exhaustion of resources) to about 55 million barrels per day by 2020, unless major undiscovered new sources come on stream. More optimistic models exist, but a peak date in the very near future is gaining consensus (see Colin J. Campbell’s The Essence of Oil and Gas Depletion [MultiScience, Essex, 2003] and visit

Right: The ‘engine’ of a solar cell is a so-called ‘p-n junction’, where two semiconductor layers of (usually) different materials are joined (a p-type and an n-type, respectively: the terms indicate whether positively charged or negatively charged carriers are dominant in the material). The junction creates an electric field in the ‘depletion region’ that originates around the junction as a result of the different carrier properties of the p-type and n-type materials, respectively When the junction is illuminated with radiation (such as light) of the right wavelength, a large number of charge carriers (electrons in the p-type semiconductor and ‘holes’ in the n-type semiconductor) are ‘set free’ in the region that contains the electric field. The electric field generates an electromotive force (e.m.f.) that causes both positive and negative charge carriers to move along the field lines. This generates an electric current without the need for a battery.

▲ ▲

The CIGS story The advantage of CIGS is that it’s much more efficient than silicon at converting incident sunlight into an electric current, that is, less than one micrometre of CIGS in the form of a thin film absorbs more than 99% of available incident solar energy, compared to the 350 micrometres of silicon needed to do the same job. This means that a CIGS cell can be deposited on a flexible material, hugely increasing the scope of possible applications. For instance, thin CIGS sheeting could be used as a coating on various objects, big and small. Such coatings could either provide a direct electric current where needed, or be used to charge batteries. The copper-indium-diselenide (CIS) solar cell was produced in the USA in 1974 and CIGS is its successor. It has taken three decades since the first synthesis of CIS for a full-scale commercial plant to be designed to produce CIGS solar panels. This is because the CIGS-based solar cell is so complex. At any of the four main stages in the process described below, small departures from the quality required can ruin the cell’s performance. ■ First, three metals have to be persuaded to combine into a very high-quality alloy: copper (Cu), indium (In), and gallium (Ga). ‘High quality’ here means absence of defects, and consistency in the material’s crystal structure. ■ Next, this Cu(In,Ga) metallic alloy must be transformed into an equally highquality semiconductor, which is done by introducing selenium into the alloy, creating a totally new crystal structure. This forms the CIGS layer of the cell. ■ A buffer layer made of different semiconductors (for example, cadmium sulphide) has to be laid carefully on top of the CIGS layer, followed by two further layers of intrinsic zinc oxide and of doped zinc oxide, respectively3. 3. The ‘doping’ of semiconductors means the artificial addition of trace amounts of impurity of the right kind to the material so as to obtain the desired electronic behaviour. ‘Intrinsic’ materials are those that have not been artificially doped.

Solar cell Light absorbed in depletion region creates electron-hole pairs

Definitions and facts ■ A solar cell is a small device, about 1 cm x 1 cm in size, produced and tested in laboratories to study the properties and structure of photovoltaic (PV) materials. Photovoltaic materials have the ability to produce an immediate direct current (d.c.) when they are illuminated by electromagnetic radiation of sufficient energy, such as from the Sun. ■ A solar panel is a combination of many individual solar cells into a product that is useful in many practical applications where electric current is required. A typical solar panel might contain a few dozen such cells and deliver of the order of 50 W of power at noon on a sunny day. The cost of solar electricity has been calculated in the 2005 projections by the REN21 Renewable Energy Policy Network (Worldwatch Institute, Washington, DC), which indicate that solar electricity could be produced at a cost of about 10 US cents (that is, 61 South African cents) per kWh by 2010. This amount compares favourably to current retail costs for electricity produced by other means. Silicon vs CIGS solar cells Silicon




Typical thickness

350 micrometres

1 micrometre

Conversion efficiency

approx. 20%

approx. 20%

Cost per peak watt

3 euro

1 euro

Date of first cell

A cutaway diagram of a CIGS solar cell

A cutaway diagram of a CIGS solar cell, showing its layers. The positive metallic contact at the base is made of molybdenum (Mo). (There is also a negative metallic contact on top, not shown here.) The active light absorber is the CIGS layer (orange), which is also the ‘p’ side of the p-n junction. The buffer layer (yellow), made of cadmium sulphide (CdS) – although non-cadmium-containing compounds can also be used – in partnership with the thin layer of intrinsic zinc oxide (i-ZnO), preserve the electronic integrity of the junction (that is, they ensure that the electronic functioning of the layer is sound). The aluminium-doped zinc oxide (ZnO:Al) is the ‘n’ side of the junction.

Quest 2(3) 2006 9

Making solar power feasible

Above: Vivian Alberts holding a solar panel. Photograph: Cyclops

Above right: A solar panel in a rural community. Fridges, lights, computers, and hot plates can be driven by a few panels like the one depicted. Photograph: The Ashden Awards for Sustainable Energy (www.

10 Quest 2(3) 2006

Creating device-quality compound semiconductor thin films for applications in opto-electronic devices is a complex procedure, especially when chalcopyrite* alloys are involved, such as copper-indium-gallium-selenium-sulphide. This particular multiphase semiconductor alloy is considered to be the most promising thin film material for producing photovoltaic devices (such as solar panels) with high conversion efficiencies (of solar to electrical energy, for example). The commercial potential of the technologies, however, has been limited in two important ways. ■ First, the complex reaction kinetics of intermetallic phases results in the partial or even complete separation of ternary phases (that is, CuInSe2 and CuGaSe2) during formation of the alloy. Phase segregation during the production process results in a semiconductor with inferior material properties and, therefore, solar modules with poor conversion efficiencies. ■ Second, there has been a lack of equipment able to manufacture high-quality chalcopyrite thin film alloys commercially under controlled experimental conditions on largearea substrates. This means that, till now, it has not been possible to produce these thin films of consistently high quality on a large scale, so yields were too low to be economically viable. The technology that I developed addresses both of these technological restrictions. A novel chemical reaction process was devised to produce homogeneous singlephase quaternary (that is, CuInGaSe2) and pentenary (that is, CuInGaSSe2) chalcopyrite alloys on large glass substrates of up to 100 cm × 30 cm in area. Using this method, the lattice parameter of the alloy through its entire depth of 2.5 micrometres (µm) varies by less than 1%, which indicates a homogeneous, or uniform, distribution of all five elements in three dimensions throughout the crystal lattice. This, in turn, means that the optical band gap of the semiconductor alloy can be controlled very precisely over a wide range between 1 and 2.4 electronvolts (eV), allowing for the best possible match with the solar spectrum and, therefore, for optimal solar conversion efficiency. Equally important, the single-phase alloys are prepared by a combination of commercially proven in-line sputtering processes and diffusion processes**. These complex technological concepts have already been demonstrated on a pilot scale. They are currently being expanded to full commercial manufacture by a German engineering company under an international licence and technology agreement. – Vivian Alberts Vivian Alberts is professor of physics at the University of Johannesburg (see also p. 47). * Chalcopyrites are a family of compounds, such as CuInSe2, comprising elements in the I, III, and VI groups of the periodic table, with the same crystal structure. ** ‘Sputtering’ is a process for coating a film of a metal onto a solid surface; ‘diffusion’ is the process by which different substances mix to form a more uniform concentration as a result of the random motions of their component atoms.

■ A cell is completed by attaching conductive electrical contacts on either side. In the present CIGS panel design, molybdenum is used as the positive contact. Research groups across the world have studied CIGS since 1974, but they encountered serious problems in their search for a reliable, repeatable, and commercially affordable procedure for producing high quality CIGS cells. In the second stage of the process in particular, when selenium is introduced into the alloy, exactly the right kind of new crystal structure needs to be created uniformly within the cell. In the past, this had been achievable only in a laboratory process that was too expensive to be commercially feasible. To create the same highquality result, consistently, reliably, and affordably, Alberts invented an entirely new procedure based on the way in which the atoms behave from the beginning to the end of the process. It made commercialization possible at last. The first full-scale commercial manufacturing plant for CIGS panels, using his procedure, is now being constructed in Germany4. All being well, it will produce its first commercial CIGS panel before the end of this year. Many possibilities lie ahead with marriages of renewable energy technologies. The cheaper CIGS solar panels could, for instance, serve as the electrical source to manufacture hydrogen fuel cells for future generations of motor vehicles. Products involving CIGS could be particularly useful for the many poor and marginalized communities not connected to electrical grids but enjoying high annual insolation (that is, elevated amounts of sunlight or solar radiation during the year). Time will tell which of the many possibilities will succeed. ■ Dr Chris Engelbrecht is in the Department of Physics at the University of Johannesburg. His research interests are in the fields of astrophysics and materials science. 4. German companies, with decades of expertise in the solar panel industry, will bring additional state-of-the-art production technologies to South African plants in a shorter time than would have been possible if these technologies were to have been developed in South Africa from scratch.

Right: Offering golden career opportunities – an Eskom transmission substation. Photograph: Eskom

Q Careers in S&T


he country’s electricity supply is in the hands of qualified technologists and engineers. Reliable and affordable power generation and transmission lies at the heart of the future growth and economic viability of South Africa and its people. Success depends largely on having sufficient well qualified personnel. There have never been so many career opportunities as there are today in the profession of supplying power to the nation. ■ The existing infrastructure of Eskom, the nation’s supplier of electricity, has to meet demands that increase by the day, and there’s an insatiable need for competent persons who can work with the 28 000 km of power lines in the facility’s transmission system, its 28 power stations, and its 4 000 substations. ■ Expansion will have many new jobs for the right people. There are plans to build two new combined cycle gas power stations in the Western Cape, two new power stations in KwaZulu-Natal, and, in the longer term, a new base load power station on the rich coal fields in Limpopo. The challenges range from very basic technologies to the most advanced engineering applications. South Africa has two main core competencies: first, in the generation of electricity using low-grade coal; second, in transmitting ultrahigh voltage at high altitude. These make the country globally unique in the environment of power generation and transmission, so people trained in these fields are sought after all over the world. What’s available? On the technical side

■ Electricians are needed to install, maintain, and repair generators, transformers, switchgear, and electrical motors at power stations and substations. ■ Fitters and turners specialize in fitting and repairing equipment of all sizes by milling, turning, and fitting components. ■ Instrument mechanics calibrate, maintain, and repair industrial instruments that indicate, register, and control temperature, pressure, and liquid flow. Matric subjects include English, mathematics, and physical science. For work in these fields, plan to qualify at a technical college or an institute of technology by obtaining certificates (at N1, N2, and N3 levels) and national diplomas (at N4, N5, and N6 levels). On the engineering side

■ Electrical Engineering: Heavy Current – this work entails installing high-voltage transmission lines and ancillary protection equipment, running maintenance programmes, and calculating relay protection settings. ■ Electrical Engineering: Light Current – this field has two main directions: one is instrumentation and control, the other is

Work to supply the country’s power South Africa’s power supply is the key to national development and prosperity. And it’s offering the best ever career opportunities in technology and engineering. testing and communication. ■ Mechanical Engineering – this is a specialized area focusing on the operational performance of boilers, turbines, and auxiliary equipment. It is particularly concerned with the design, manufacturing, installation, and maintenance of mechanical equipment. ■ Electro-mechanical Engineering – this choice is ideal for generalists who want to become hands-on professionals involved in the operations and maintenance of ‘cross-overs’ from mechanical to electrical engineering and vice versa. Matric subjects include English, mathematics, and physical science. For further study in these fields, enroll at an institute of technology or a university and obtain a three-year Advanced Diploma or a four-year B.Sc. (Engineering) degree. Other openings

There are also limited career opportunities in business support areas such as financial management, commercial management, information management, communication, and human resources. The work and the people The work environment. In the world of power supply, there’s work in an office environment, and, for those who prefer it, there’s work in the natural environment and opportunities for travel. South Africa’s national power network extends from northern Limpopo down to the Western Cape, and the building of a western corridor from South Africa to the northernmost edge of the SADC region will bring international

work opportunities as well. Personal attributes and skills. A checklist for work in the fields of power supply includes an aptitude for mathematics and physical science, being good at design, and being interested in innovation in the context of developing new technologies and methods for supplying electricity. A facility like Eskom looks for people who are systematic, reliable, and good with detail; self-starters; and professionals able to work well with others in a diverse work force and able to be ambassadors for South Africa in an international setting. Learnership support

To meet its goal of 6% economic growth, South Africa has embarked on the Accelerated Shared Growth Initiative for South Africa (ASGISA). State-owned enterprises are on board with learnerships that offer training, employment, and valuable work experience to young people who qualify at all levels, from schoolleavers to those leaving technical colleges, institutes of technology, and universities. Eskom’s goal is to give this kind of opportunity to as many as 4 000 people who’d like a career in power generation and transmission. QUEST thanks Eskom Communication Media for the generous help in providing the information on this page. ■ For more information, contact Eskom’s Student Development Manager, Bongi Ntuli, at tel. (011) 800 5796, fax (011) 3014, or e-mail For details about qualifications, contact technical colleges, institutes of technology, or universities of your choice.

Quest 2(3) 2006 11

Ornithologists Alan and Meg Kemp explore the world of hornbills. What threatens these unique birds and what can be done to save them? Why should we care?

Save the hornbills

12 Quest 2(3) 2006


ornbills (of the avian order

A first for Africa and for South Africa The latest International Hornbill Conference (the fourth in the series, held for the first time in South Africa and in Africa) took place at Mabula Game Lodge, Limpopo Province, on 6–10 November 2005 and attracted 85 delegates from 20 countries. The theme was “The Active Management of Hornbills and their Habitats for Conservation”, and the conference covered Key Regional Projects, Special Species Studies, Habitat Fragmentation, Population Genetics, Status and Threats, Feeding Biology, Study Techniques, and Culture and Education, with special sessions on The Conservation of Ground Hornbills in Africa and The Need for International Management of In- and Ex-situ Conservation. The main message was the need for innovative conservation management of the many threatened hornbills worldwide, and for sharing in Africa the techniques developed by countries in the lead such as India, Thailand, Indonesia, and the Philippines. The key conservation lessons featured in this article are based on the latest data, presented at the conference. For details, visit Alan and Meg Kemp at the registration desk of the 4th International Hornbill Conference that they organized. Photograph: Lucy Kemp

they occur. Eco-tourists love to see them, and they’re also widely exhibited in zoos outside their home range. Many hornbill species are ecologically important – some as seed-dispersers for the plants whose fruits they eat, others as top ▲ ▲

Bucerotiformes) are active, noisy, attractive, and useful birds. But their worlds – and numbers – are shrinking as forest and savanna are cleared for human development. Learning how to look after them also means understanding better how to save other species that face the same threats. Hornbills are remarkably varied. The largest (the southern ground hornbill, Bucorvus leadbeateri, which weighs about 4 kg) lives in Africa – as does the smallest, the red-billed dwarf hornbill (Tockus camurus), which weighs just 100 g! But in the rainforests of South-East Asia are found the species with the most spectacular voices, colour, and form, as indicated by their names of rhinoceros, wreathed, or great pied hornbills. There are 54 hornbill species in all, extending east across sub-Saharan Africa from Senegal though the Middle and Far East to the Philippines, and south to South Africa, Sri Lanka, and the Solomon Islands. All but one of the 30 Asian species mainly inhabit forests and eat fruit, while two-thirds of the 24 African species predominately inhabit savanna and feed on small animals. Like other animals with a dramatic presence and a fascinating biology, they often form part of the traditions of their human neighbours and have become symbols of the habitats in which How hornbills disperse seeds

Studying the habits of animals is time-consuming, meticulous work. Researchers discover what fruit are eaten by wild hornbills by watching them feed, or by watching what they deliver to their nests and what they’ve dropped below their nests. We can also follow hornbills, to see how far and how often they travel to or from any fruiting trees at which they feed – we can observe them directly, or we can catch them, attach a small radio transmitter, and then track their movements. If we feed captive hornbills with samples of the fruits that previously observed birds in the wild have been eating, and watch to see when large seeds are regurgitated or when small seeds are passed in the droppings (faeces), we can estimate how long it takes from the time they swallow the fruit to the time they drop the seed(s). From these data, we can predict the distances from the parent tree(s) at which seeds are likely to be dropped by a hornbill. Then we can compare our prediction with the pattern and size of seedlings growing around the parent tree(s). Some trees seem to need their fruit to pass through the digestive system of an animal before the seeds will germinate; some fruits show improved germination after passage through a gut where enzymes soften the skin or shell to trigger growth. So if we collect fresh fruit from trees (or uneaten fruit accidentally dropped below nests) and compare their seeds to those passed through guts (either ejected below nests, or fed to captive birds), there may be important differences in the probability of the two sets of seeds germinating and growing successfully.

Photographs by the authors, unless otherwise specified.

Photographs on opposite page (clockwise from top left): Southern yellow-billed hornbill (Tockus leucomelas), adult male in the Kruger National Park. Southern red-billed hornbill (Tockus rufirostris), adult male preening its wing coverts in the Kruger National Park. Great pied hornbill (Buceros bicornis), adult male perched by its nest in the Khao Yai National Park, Thailand. Photograph: Thailand Hornbill Project

Visayan writhe-billed hornbill (Aceros waldeni), captive adult female (left), adult male (centre), and the first offspring ever raised in captivity of this critically endangered Philippine endemic species. Photograph: Pavel Hospodársky ˇ `

Rhinoceros hornbill (Buceros rhinoceros), adult male perched above the nest entrance after feeding the female inside, in the Budo-Sungai Padi National Park, Thailand. Photograph: Thailand Hornbill Project

Rufous-necked Hornbill (Aceros nipalensis), adult male of probably the most beautiful hornbill species in the world, in the Huay Kha Khaeng World Heritage Area, Thailand. Photograph: Thailand Hornbill Project

Meg about to enter an observation hide/blind to watch animals coming to a fruiting fig tree at Halabala Wildlife Sanctuary, Thailand.

Quest 2(3) 2006 13

Save the hornbills

predators of the small animals on which they prey. In tropical forests in particular, where large mammalian frugivores (fruit eaters) such as monkeys, apes, elephants, and rhinos have been decimated, hornbills are now the main dispersers of the seeds of many forest trees and creepers. All this makes some hornbill species ideal flagships for conserving the ecosystems that they share with other threatened animal and plant species.

▲ ▲

Unique nesting habits Hornbills are unique among birds for their nesting habits. The breeding female is incarcerated by plastering herself into a nest cavity, leaving open only a vertical slit through which the male delivers food. During the egglaying process, she moults her major flight feathers and for a while cannot fly. The male – strictly monogamous – takes on the burden of feeding his entire family and, most remarkably, even provides the nutrients that the female needs to replace her major flight feathers. For her part, the female has a special spermstorage capacity. This means that her eggs will be fertilized, even if she lays them up to three weeks after her final copulation before entering the nest. She keeps the nest clean by throwing away food remnants through the slit and squirting droppings out the same way. Later, her offspring do the same. In most species the female, together with her chicks, leaves the nest only after the nesting

cycle has ended. But in some species she leaves when the chicks are only half grown, to help the male to feed their growing brood. In these species, the chicks, regardless of sex, reseal their nest all by themselves – the final amazing feat of hornbill nesting. Only the two species of African ground hornbill, ancestral to all other members of the order, deviate from the norm by neither sealing the nest entrance nor ejecting their waste, although the female remains on the nest throughout incubation and the early nestling periods. From the top: Two southern ground hornbill (Bucorvus leadbeateri) chicks in the Kruger National Park. The elder, from the first-laid egg, is about a week old with its skin already turned black; the younger from the second-laid egg with its skin still pink has just hatched. Both chicks still have their eyes closed. The smaller chick normally starves to death within about a week, apparently from parental neglect. Adult southern ground hornbill female incubating on the nest in the Kruger National Park. This is one of two species in the only genus of hornbills that does not seal the entrance to the nest cavity. White-crowned hornbill (Aceros [Berenicornis] comatus), adult male hanging at the entrance to the nest while feeding the female inside, in the Budo-Sungai Padi National Park, Thailand. Photograph: Thailand Hornbill Project Great pied hornbill (Buceros bicornis) adult male flying away from its nest in the Khao Yai National Park, Thailand. This is probably the most spectacular large forest hornbill. Photograph: Thailand Hornbill Project

New nests Reducing the factors that limit the populations of species we want to protect is another method of conservation, and it has worked particularly well with hornbills. Each breeding pair of hornbills uses a natural cavity as a nest site – in a tree or rock face – so a shortage of cavities limits populations. Fortunately, we know from studies in zoos and in the wild that most hornbills accept a modified natural or artificial cavity. In Thailand’s Khao Yai National Park, for instance, 4 species have used 48 natural tree cavities that were repaired or modified for them. The great pied hornbill (Buceros bicornis) nearly doubled its breeding success as a result, and artificial cavities are also being supplied to this species in the fragmented forest patches of western India. In the Philippines, artificial nest boxes have been used to breed the rare Visayan writhe-billed hornbill (Aceros waldeni), both in captivity and in their deforested natural habitat, to boost their critically reduced numbers. In Senegal and Namibia, artificial nest boxes were used for research into the nesting biology of the West African red-billed hornbill (Tockus kempi)*, Damara red-billed hornbill (T. damarensis), and Monteiro’s hornbill (T. monteiri), respectively. Incidentally, they boosted the local population density in both of these arid savannas. Excavated nest logs have been readily accepted by southern ground hornbills in South Africa, so providing more of them forms an important pillar in plans to reverse the local decline of this species. * A study of the forms of red-billed hornbill across Africa revealed two new ones, one in West Africa that colleagues (Bernard Treca and Christian Erard) named as a subspecies after Alan Kemp in 2000, and one in Tanzania that Alan and Wayne Delport named Tockus erythrorhynchus ruahae, after the Ruaha National Park where the type specimen was collected by Susan Stolberger in 2002. Wayne’s detailed study of two forms in Namibia had supported their recognition as separate species and we proposed that this applied to all five forms recognized in the red-billed hornbill complex.

Artifical nest log for southern ground hornbills. It was cut from a pine tree trunk and erected in a knobthorn tree in the Associated Private Nature Reserves bordering the Kruger National Park by the Mabula Project. It is being researched by the Percy FitzPatrick Institute of African Ornitholgy (University of Cape Town).

Quest 2(3) 2006 15

Save the hornbills

Southern ground hornbills and conservation Some facts. Southern ground hornbills (Bucorvus leadbeateri) are the subject of pioneering research – for Africa, and for hornbill species as a whole – at Mabula. Our 30-year study of B. leadbeateri in the Kruger National Park has yielded interesting information on their natural history: ■ southern ground hornbills live in cooperative groups of 2–11 birds, with only one alpha (or leading) pair of breeding adults per group ■ groups occur at a low density of 100 km2 per group ■ on average, only a single chick every 9 years in each group is raised to fledging (that is, till it is able to leave the nest and fly) ■ adult survival is high (estimated at 98% per annum). Conservation context. These details about the world’s largest and most carnivorous hornbills also exemplify some of the survival issues facing other vulnerable organisms characterized by ■ large size ■ complex social structure ■ low density ■ slow productivity ■ high adult survival ■ high position in the food-chain. All these translate into a small total population, deceptive persistence, conflict with human interests, and slow recovery once their numbers have been depleted. Southern ground hornbills occur throughout African savannas south of the equator, so, as a species, they are not yet endangered. But within the 5% of their range that extends into South Africa, they are vulnerable and their populations are declining. While there are about 200 breeding pairs/groups in the safety of the Kruger National Park (that is, 720 individuals, given a mean group size of 3.6), we estimate that a similar number is all that’s left in the unprotected areas of Limpopo, Mpumalanga, KwaZulu-Natal, and the Eastern Cape, where the species may have lost 50% of its previous range and where numbers are falling. Solutions. Our research revealed that, in 80% of clutches, southern ground hornbills laid two eggs, even though they normally raised only a single chick. We also found that breeding success was limited by the availability of suitable tree hollows for nest sites – but that the birds readily accepted repaired natural cavities or specially constructed artificial cavities as nest sites. We now know that the second-hatched chicks can be removed, raised in captivity, and used for conservation interventions. These include reintroducing birds into parts of their previous range from which they have disappeared; adding new birds to an existing wild group that is dysfunctional because it lacks a breeding male or female or has no young helpers to form a group (known as ‘augmentation’); and captive breeding. These techniques, supported by research and awareness programmes about the risks to southern ground hornbills and their habitats (led by various organizations, including the Mabula Project, the universities of Cape Town and Limpopo, the National Zoological Gardens, and the Endangered Wildlife Trust), are now being applied to reverse the decline of B. leadbeateri populations in South Africa. Southern ground hornbills have also taken on a ‘flagship’ role for other vulnerable savanna species, such as cheetah, wild dog, eagles, and vultures. For more, visit hornbill/ground_hornbill.htm Eugene Marais, from the National Zoological Gardens, Tshwane, inspects a southern ground hornbill nest site in the Kruger National Park. He ascends by ladder so as not to leave a scent trail up the tree that might attract predators to the nest.

Southern ground hornbills (Bucorvus leadbeateri). Top: Adult male, with his all-red bare throat skin, uttering his deep booming call from a perch in the Kruger National Park. Middle: Close-up of the eye and naked facial skin of an adult male, showing the long flattened lashes forming a sun-shield above the eye. Below: Two adult males leading a group while foraging, wings apart and upper wing covert feathers raised to disperse heat from the bare undersides of the forearm.

16 Quest 2(3) 2006

Save the hornbills

Danger ahead No fewer than 19 (that is, 35%) of hornbill species are listed in the IUCN Red Data Book for endangered species. Those in danger are forest-living, and Asian in their distribution: 10 are classified as ‘near threatened’ (of which 7 are large species), 5 as ‘vulnerable’ (3 of these are restricted to small islands), and 4 as ‘endangered’ (2 of these ‘critically’) on the fragmented and deforested islands of the Philippines. African hornbills are on the whole under less pressure than most of their Asian cousins. Warning lights are flashing, though, for certain forest species – such as the yellow-casqued wattled hornbill (Ceratogymna elata) and the brown-cheeked hornbill (C. cylindricus) in West Africa – and for a few savanna species, such as the southern ground hornbill in southern Africa. South Africa leads the continent in applying special genetic studies, artificial nest boxes, captive management, and detailed field studies for purposes of conservation. For example, southern ground hornbills form part of pioneering efforts at Mabula Game Reserve near Bela Bela (Warmbaths) to reintroduce this flagship species of the savanna habitat to the central bushveld. The stock came from the Kruger National Park, and these efforts are part of a wider project to augment the species in Limpopo through reintroduction, by adding birds to dysfunctional groups, and by providing artificial nest boxes. Lessons in conservation Population fragmentation: effects of geography

Range and species sensitivity How can hornbills and people coexist better? To find out, it’s useful to study the biology of hornbill species that are already adapted to life in a small patch of habitat, and compare them with more widespread species. The Indian island of Narcondam, for example, is a single volcanic peak with only 6.82 km2 protruding 750 m above the Andaman Sea. It supports an estimated 320–340 individuals of the unique Narcondam hornbill (Aceros narcondami), the smallest range and natural population size known for any hornbill species. Several species of Aceros have colonized remote islands, including the only hornbill to reach Australasia (on New Guinea and adjacent islands), but A. narcondami is the smallest – its size could even be an evolutionary solution to pack a viable population onto such a limited area! They’ve also discovered high-rise living, with adults roosting below 255 m, among the large fig trees that provide their staple diet and nest sites, and juveniles at the less productive higher elevations. At the other extreme, the African grey hornbill (Tockus nasutus) extends from the Sahel and Arabian Peninsula to South Africa in all but the most densely forested and wooded habitats, across thousands of square kilometres of habitat, and numbering tens of thousands of individuals. It occurs in two distinct forms, each approximately on either side of the equator, and these might even be distinct species. Even so, these small wide-ranging species would still outnumber the island endemics by orders of magnitude. This makes them much less vulnerable to threats at present, whether natural ones such as the cyclones and tsunamis that batter Narcondam, or human-induced ones such as the small police post on Narcondam, and, on a larger scale, desertification or wood gathering in Africa. Female (left) and male Narcondam hornbills (Aceros narcondami). Probably the only two live members of this species to leave their tiny island, they were taken by S.A. Hussain in the early 1970s to be studied in captivity at the headquarters of the Bombay Natural History Society. All previous examples of the species had left the island dead, prepared as specimens for natural history museums. Unusually for birds, juveniles of both sexes of many hornbill species resemble the adult male. Only at the first moult to replace her body feathers do the black head and neck develop in Narcondam hornbill females.

▲ ▲

The conservation status of hornbills in Africa and Asia differs considerably, partly because each region has a different topographical structure. The African continent is a single landmass, so the range of each hornbill population is restricted only by the extent of its preferred habitat – some form of forest or savanna. The Asian region, however, is divided into many islands and peninsulas, most (at least historically) covered in tropical forest, so the range of many populations is already isolated or fragmented by water. At one extreme are species extending all the way across all the savannas of sub-Saharan Africa and the Arabian Peninsula; at the other are species restricted to a single oceanic island. The earliest credible fossil record for ground hornbills is 15 million years ago, from mid-Miocene deposits in Morocco. Later prehistoric tectonic and climatic changes split up the ranges on both the African and Asian landscapes. Now hornbills face alterations that come from human activity – including rapid, devastating changes both to habitat and to the global climate, which fragment their habitats and populations into dangerously small units

Top: At a lookout point over the rainforest canopy at Halabala Wildlife Sanctuary, Thailand, Alan conducts a routine count of hornbills and birds of prey. Photograph: Justin Kemp

Quest 2(3) 2006 17

Save the hornbills

Villagers become conservators Budo-Sungai Padi National Park in southern Thailand supports eight species of hornbills, but the surrounding villagers have traditionally harvested trees and poached hornbills from the forest for their own use. In 1998, the Hornbill Foundation Thailand initiated a scheme whereby urban Thais and foreigners pay an annual fee to adopt a hornbill nest. These fees are shared among the converted hunters and poachers, who now use their forest skills to help with studies of the hornbills and to convert their neighbours into forest and hornbill conservators. The villagers are also trained as eco-guides and, for additional income, they show adopters and other tourists the hornbills and forests, while the village children conduct surveys of indigenous tree saplings and then grow and plant seedlings for forest regeneration. This scheme is ongoing and growing all the time. In 2005, it was shortlisted for a Rolex Award (instituted by the Rolex watch company of Geneva, Switzerland).

and in this way threaten their survival. All that can be done now is to stem the tide of habitat loss, protect tracts of original habitat that are as large as possible, and encourage linkages between remaining patches of habitat by corridors or translocations1. Human population density and demand for land

Below: Indian tribal wearing a central tail feather of a great pied hornbill to embellish his hairstyle.

People live at high densities across much of the Asian region, but over extensive tracts of the African continent they are thinly dispersed. Groups of humans use the resources around

them for their own immediate survival, or they exploit those resources to generate wealth (often for people who live elsewhere). Both types of behaviour transform the natural habitat, but the most extensive and extreme effects come from wealth creation (such as logging, mining, plantations, crops, and human settlements). Hornbill habitats have shrunk across Africa, but most alarmingly in Asia, where high human densities combined with the sensitivity of forest habitats have both reduced the area and increased the fragmentation. One form of hornbill from the Philippines, the Ticao race of the Visayan tarictic hornbill (Penelopides panini ticaensis), whose island forest has been completely cleared, has just had the dubious distinction of being the first hornbill taxon to be declared extinct. (This fact was confirmed for the first time at the Mabula conference last year.) These same pressures now threaten or endanger various others in Asia and, while no African species is yet under such threat, the deforestation of large tracts in Africa is cause for concern, especially in West Africa and the Congo Basin. The only feasible remedies seem to be to create protected areas that are as big as possible

Saving tribal traditions – and birds Sometimes local traditions can conflict with conservation. But tribes in India have discovered creative solutions. The Nyishi tribe inhabits rainforests in the northeast Indian province of Arunachal Pradesh. Senior members of the tribe wear a headdress formed around the upper mandible and casque of the great pied hornbill, a bird of special significance in their culture of forest appreciation and wise use. A forester, who was also a member of the tribe, recognized the threat to their traditions of hunting the declining hornbill population for its bills. Against considerable resistance, he convinced the leaders of his tribe to substitute a fibreglass cast of the bill for the real object. In this way, they could save the wild hornbills as symbols of the forest they revere while continuing to wear their traditional headdress. The manufacture of artificial hornbill bills became a thriving cottage industry and now senior members take even more pride in the quality of their headdress, in maintaining their tribal traditions, and in conserving their forest. Hornbill bills and feathers also embellish traditional dress among various tribes in Asia and Africa, so now many western zoos collect feathers moulted by their captive birds and send them to tribes in India and Malaysia to help reduce hunting pressure on wild populations. Right: A Nyishi elder in northeast India, proudly wearing a fibreglass copy of the upper bill and casque of a great pied hornbill (Buceros bicornis) as part of his traditional ceremonial headdress. Photograph: Divvya Mudappa

Right (below): The head, bill, and casque of a young 3-year-old male great pied hornbill.

18 Quest 2(3) 2006

1. Such interventions can help any species whose population fragments have become isolated and need help with maintenance of genetic options through influx of new genes (‘blood’) by immigration – natural or translocated. At present, this is most important for several Asian hornbill species, but will become relevant in Africa if trends (such as local extinctions of southern ground hornbills) continue.

Save the hornbills

Left: A vendor with his children and a colourful ‘flock’ of carved hornbill curios hanging at his outdoor stall along the road leading to the main entrance gate of the Kruger National Park near Skukuza. Below: A southern ground hornbill adult female (blue patch under the throat on the bare red facial skin) leads her group of three adult males, only one of which is her biological partner. For the most up-to-date, comprehensive information about the hornbill family and its species, consult A.C. Kemp, The Hornbills: order Bucerotifomes, Oxford University Press, Oxford, 1995 (out of print); and A.C. Kemp, “Family Bucerotidae (Hornbills)”, in J. del Hoyo, A. Elliott, & J. Sargatal (eds.), Handbook of the Birds of the World. Vol. 6. Mousebirds to Hornbills (Lynx Ediciones, Barcelona, 2001), pp. 436–523 (still in print and an excellently illustrated summary of the 1995 book). There is no dedicated hornbill web site, but a Google search will help you to find minor regional or organizational sites.

Ground hornbills and Africa’s drumming The two species of ground hornbill, the members of an ancient genus unique to the savannas of Africa, utter a deep booming four-note call at dawn – Doo Doo DuDu. In many rural areas this call, normally uttered as a duet between males and females, heralds the start of an African day. Ground hornbills feature in the folklores and songs of many African tribes, and they interpret the ‘conversation’ between the male and female duettists in various ways. The 1940 edition of Roberts’ Birds of Southern Africa, for instance, quotes one in which the female calls “I’m going, I’m going, I’m going home to my relations”, to which the male replies, “You can go, you can go, you can go home to your relations!” Traditional African drums produce beats at the same low frequency, and experienced drummers can replicate the call of the ground hornbill so well that the hornbills are sometimes induced to reply. It has even been suggested that the tone and tempo of some African drumming was inspired originally by the calls of ground hornbills – maybe along with those of lions, eagle-owls, large bustards, and ostriches, whose similar low-energy low-frequency calls are also best for transmitting over long distances through the still, temperature-layered air of dawn.

and, outside these areas, to reduce threats such as logging and hunting. Joint human–hornbill initiatives

In many areas of Africa and Asia, people and hornbills can only continue to coexist if both to some extent modify their behaviour. People will have to learn more about what hornbills need, and to alter those activities that – often unintentionally – affect the birds adversely. The possibilities include trying not to cut or disturb trees that hornbills need for food or as nest sites, saving large tracts of forest from logging to conserve sufficient areas of hornbill habitat, preventing accidental poisoning, and looking for ways to enjoy hornbills and to profit from their eco-tourism potential. For their part, hornbills may have to get used to sharing their forest environment with people, accept artificial nest structures, and submit to carefully controlled forms of harvesting. For example, we know that in several larger species

of hornbills only a single chick is raised, even though second or third eggs are laid. So if pairs can be persuaded to lay those eggs in accessible nests and accept the removal of the surplus chicks, specimens might become available for various forms of conservation and use. Ways are being found in Africa and Asia to achieve this balance between people and hornbills. Education and awareness are the cornerstones of reducing conflict with hornbills, but finding ways to benefit from these splendid birds, directly or indirectly, most probably offers the best long-term salvation. ■ Alan and Meg Kemp (who, in their retirement, style themselves ‘naturalists and nomads’) are best known for their work with birds and for their many publications. Alan’s hornbill studies began in the Kruger National Park and continued throughout his 30 years as head of the Department of Birds at the Transvaal Museum and beyond. Their hornbill work has taken Alan and Meg all over the world.

Quest 2(3) 2006 19

Maths and its meanings Mathematics means many different things to different people, says Doug Wilson.


nature (even though these are essentially constructs of the human mind), then mathematics is a science, since it is predominately a study and exploitation of number and logic. ‘Scientist’ mathematicians see mathematics as a structure existing in nature, there to be discovered and not invented. When they uncover one of its properties, they set about performing

sk a mathematician “What is mathematics?” and you’ll receive different, seemingly contradictory replies. The truth is that there are at least half a dozen different but pertinent approaches to mathematics. It can be an art form, a science, a game, or a language; we can use it as a tool or as modelling material. Or we can incorporate all six in our thinking.

a  b  b  a; if a  b and b  c then a  c ). (Axioms are also called postulates.) It’s upon such a basic set of premises that a mathematical structure is built. Definitions are very important in this building process. By means of clearly stated definitions, new concepts are introduced as the mathematical structure is developed. Important truths that arise in the process of building this structure are called theorems.

An art form An art form can be viewed as a structure made from basic building elements. Its beauty is perceived in the structure of the art form and not in the basic building elements themselves. Consider a painting as an art form. Its basic elements are the initial set of different colours of paints on the artist’s palette. The artist arranges these into a two-dimensional spatial structure on canvas. Whether the painting depicts a real-world situation or is purely abstract is not relevant to justify the work as art. What is important is the indefinable beauty experienced by the viewer in the end product. This beauty resides in the spatial structure into which the colours have been arranged, not in the individual colours themselves. Similarly, the basic elements in a musical work are sounds made by instruments or the human voice, which the composer has arranged into a very complex time-based structure. Again, it’s in the structure that the beauty of the music is perceived. If this temporal structure were randomly disarranged, the music would become meaningless. Other art forms such as sculpture, architecture, and poetry can be analysed in much the same way. So too can mathematics! The basic elements of any mathematical structure – for example algebra, geometry, number theory, set theory, and matrix algebra – are the basic premises of that branch of mathematics. These comprise certain fundamental concepts (such as those of a point, a line, space, time, and number), operations (such as , , , ), relations (such as a  b and x  y), and axioms (such as

The more tools one has in one's mathematical toolbox, the greater is one's power to solve practical real-world problems.

20 Quest 2(3) 2006

an experiment (proof) to verify that the property is consistent with what is already known to be true. Mathematicians interested in number theory, for instance, tend to be ‘scientist’ types1.

The beauty of mathematics is perceived in the complex and logically consistent structure that emerges. Such an approach to mathematics as an art form is called the axiomatic or postulation approach. It’s essentially that of the creative artist, who is not so much concerned with the use of his or her art form as with its inherent beauty. For the ‘artist’ mathematician, mathematics is creative – a process of invention rather than discovery. The philosopher and mathematician, Bertrand Russell, once said: “Mathematics possesses not only truth, but supreme beauty – a beauty cold and austere, like that of sculpture”. A science A science can be defined as a systematic study of some aspect of nature. If one accepts that number and logic exist in

A game A game is an activity in which certain elements can behave or be manipulated, according to certain prespecified rules, with clear goals in mind, such as rugby, soccer, or chess. Mathematics too can be viewed as a game played by mathematicians. The elements are mathematical symbols; the rules are the basic axioms of a particular branch of mathematics. The ‘player’ mathematician enjoys playing this game of skill by doing mathematics and exploring all its possibilities with clear goals in mind, and he or she often also decides to create new mathematical games with their own symbols and rules. The German mathematician, David Hilbert, once described mathematics as nothing more than a game played according to certain simple rules with meaningless marks on paper! A language A language, in its broadest sense, is simply a medium through which information can be transferred. The concept is not confined to verbal languages. An interesting property of all art forms is that each can

1. For ‘scientist’ mathematicians, for example, it is an act of great discovery to find and prove, in the realm of real numbers, that between any two rational numbers there exists an infinite number of irrational numbers; that between any two irrational numbers there exists an infinite number of rational numbers; yet the probability of selecting a rational number at random from the set of real numbers is zero. In other words, speaking less precisely, there exist infinitely more irrational numbers than there are rational numbers!

be used as a language. A painting of a countryside scene conveys information about that landscape; music and poetry can convey feelings and deep emotions. Mathematics is, and has always been, an essential part of the language of science for, in it, words alone are insufficient, and the symbolic nature of mathematics has tremendous power for abstraction, generalization, and precision. The ‘linguist’ mathematician sees mathematics as a language and combines verbal, diagrammatic, pictorial, and mathematical representations to convey ideas and information. Logical reasoning is an essential part of the language of mathematics.

whereas basic building materials used for constructing a mathematical model are mathematical symbols. In the same way that one uses the properties of the building materials to build a physical model, one uses the rules of mathematics to build a mathematical model. In addition, mathematical models are dynamic and possess the capacity to make predictions. This means that the model will possess input features (the independent variables) through which certain data can be inserted, and output features (the dependent variables) through which the

resulting predictions are made. It’s by means of these features that a model is tested against the real-world situation being modelled. Perhaps the most famous mathematical modeller is Albert Einstein who, through his cosmological mathematical model, the theory of relativity, was able to make numerous astounding and testable predictions. Theoretical physicists are, in fact, all good examples of mathematical modellers, as are scientists working in less exact fields such as weather forecasting or predicting population growth, the spread of disease, and stock market prices, in all of which variability, probability, and statistics play an important part. Such models are referred to as stochastic mathematical models. ‘Modeller’ mathematicians regard mathematics as inexpensive material with which to build models of real-world situations and, instead of experimenting with the real thing, which may be costly or even impractical, they experiment with this mathematical device (normally with help from a computer). A good model may suggest information and offer predictions that are impossible to obtain by any other means. A tool Mathematics can be looked upon as a set of tools for solving real-world problems. As with the tools of any craftsman, mathematical tools require great skill if they’re to be used effectively. The more tools one has in one’s mathematical toolbox (and this includes the computer) the greater is one’s power to solve practical real-world problems using mathematics. This approach to mathematics is that of the ‘problem-solver’. Engineers would tend to view mathematics in this way. Combined approaches The approaches of the ‘artist’, ‘scientist’, ‘player’, ‘linguist’, ‘modeller’, and ‘problemsolver’ mathematicians are all different. ▲ ▲

Modelling material The word ‘model’ is used extensively in modern scientific disciplines. Its meaning stems from the familiar activity of building a physical model of something that is/ was either already in existence, such as a scale model of a vintage motorcar, or of something planned for the future, such as a scale model of a proposed new housing complex. One also thinks of physical models of abstract entities such as atoms and molecules. A model is never identical with the entity it is modelling but models it only in respect of certain essential features. Models can be of two kinds: static and dynamic. Those referred to in the previous paragraph are essentially static. One cannot extract more information from a static model than has already been put into them. In other words, such models cannot be used for making unforeseen predictions. Dynamic models, however, have predictive properties. A good example of a dynamic model is a planetarium, which is a physical model of the sky above us as seen from any particular point on the surface of the earth on a clear night. The console in the planetarium is a device that accepts certain data (year, month, day, time, latitude, and longitude), which we shall call the input. The sky then moves according to the rules built into the central projector to the position that would be seen from the surface of the earth at the time and place specified. The resulting configuration of

heavenly bodies in the ‘sky’ (the output) is the prediction, which, furthermore, can be tested within certain time limitations. Our confidence in the model will be determined by the extent to which predictions match observed data. Building a mathematical model is exactly analogous to the act of building a physical model. But the basic building materials of a physical model are physical materials,

Quest 2(3) 2006 21

problems. In pure mathematics, the ultimate test must be found within mathematics itself, in the form of a mathematical proof. In applied mathematics, the ultimate test is found in the extent to which the mathematics used describes, models, predicts, or solves real-world situations. If applied mathematicians find that there are no suitable mathematical materials available with which to describe, model, or solve a problem, they must either abandon the quest, or create the necessary mathematics for the purpose. Choosing the latter option means adopting the approach of the pure ‘artist’ mathematician. Isaac Newton, for example, did exactly that when he invented the calculus, which enabled him to address countless phenomena and problems in the physical world he would otherwise have been unable to do.

The rare, well-rounded mathematician is comfortable with all six approaches, and happily and unconsciously works with any one of them as the need arises. Professional mathematicians at universities tend, for many reasons, to lean toward either the ‘pure’ or ‘applied’ arena and to concentrate within it for researching specialized areas of interest and problems. But at school level a balanced approach is best, so that young people can experience both kinds and have the chance to enjoy mathematics in its broadest and fullest sense. ■ Since his retirement, Douglas Wilson has been an Honorary Research Fellow in the School of Computational and Applied Mathematics at the University of the Witwatersrand. His research interests are in the psychology of mathematical problem-solving and the pedagogy of mathematical modelling.

Election-night forecasting Jan Greben describes the technique, specially designed for South Africa, of preparing fast and accurate forecasts of final election results.


ll elections attract a lot of media attention, which comes to a climax on election night. How can people get accurate predictions, long before final results are confirmed? Here’s where the quantitative tools of statisticians and mathematicians can play a vital role. The first results that come in from the polls cause much excitement and anticipation – and sometimes disappointment – for the public and the political commentators. But initial results are often poor predictors of final results, so much of the early hype is misdirected. Then, by the time the final results are in after a day or two, public interest in the elections has waned and the media are focusing on other things. So it’s very useful to speed up the process of making accurate forecasts and to give the media and political analysts the right information while levels of interest are still high.

70 65 60 ANC 55 50 45 40 35 30 25 DA 20 60% = 15 5% = 20% = IFP 100% = ID 21 hrs 10 4 hrs 10% = 8 hrs 5 hrs 48 hrs 5 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Percentage of votes counted

This figure illustrates the biases for the 2004 elections. The results for the African National Congress (ANC) started out low, while the Democratic Alliance (DA) results started out high. The initial Inkatha Freedom Party (IFP) results were very low compared with the final result. The Independent Democrats were the only party whose result did not change much over time. The graph also indicates at what time after the closing of the voting booths the various percentages of vote counts were reached. It is worth noting that it took 48 hours for the final results to come in, whereas most media attention was focused on the first few hours. After four hours, however, only 5% of votes had been counted and the results at that time (ANC 60%, DA 22%) were still very different from the final results (ANC 70%, DA 12%).

districts dominated by certain parties (such as the Democratic Party [DA] and the New National Party [NNP] in 1999 and 2000) come in much earlier than those dominated by other parties (such as the Inkatha Freedom Party [IFP]). Initial results therefore

▲ ▲

Developing the science of prediction Election-night forecasting is the activity whereby every time new results become available after the polls close, these results are used to construct forecasts of the final results. Special scientific challenges are involved when forecasting the outcome of South African elections at a time when only a minimal number of results have come in. The main reason it is so difficult to predict the final results from the actual returns at an early stage is that the results of voting

South African elections 2004

Party Percentages

The first creates mathematics; the second discovers mathematical truths; the third does mathematics for sheer pleasure; the fourth uses mathematics to convey concepts, insights, and facts; the fifth uses mathematics to build models of the real world for predictive purposes; and the sixth uses mathematics as a set of tools with which to solve practical problems. The first three essentially characterize the pure mathematician, who sees mathematics as an end in itself. The remaining three are those of the applied mathematician, who sees mathematics as a means to an end. Mathematical problem-solving underpins both pure and applied mathematics. In pure mathematics, problems and their solutions are suggested and found within the body of mathematics itself, whereas applied mathematics seeks solutions for real-world

Quest 2(3) 2006 23

Since the momentous year of 1994, South Africa has held regular democratic elections. National elections were held in 1994, 1999, and 2004, while municipal elections were held in 1995, 2000, and most recently on 1 March 2006. The Independent Electoral Commission (IEC) has been in charge of organizing elections in South Africa from 1994 onwards. The Council for Scientific and Industrial Research (CSIR) became involved in electionnight forecasting in 1999 at the invitation of the IEC, and has since then been at work during every South African election. The team of researchers is now part of the group focusing on logistics and quantitative methods in the CSIR Built Environment unit, and focuses on multidisciplinary problems under the banner of the scientific field called operations research. In 2004 the South African Broadcasting Corporation (SABC) asked the CSIR to conduct election-night forecasts. As much of the interest in election-night forecasts is media-related, this relationship with the SABC is a natural, ongoing one.

favour these ‘early’ parties, leading some politicians and commentators to premature conclusions about the final outcomes. Although most commentators realize that these initial results are far from final, they have not had the systematic means to correct and replace them with more accurate predictions at an early stage. This was the challenge facing the team of statisticians, mathematicians, and computer specialists at the Council for Scientific and Industrial Research (CSIR) in 19991. With such a multidisciplinary group of scientists at work, discussions were lively about ways to solve the election problem. There were many options, but little guidance from the available literature, as forecasters abroad were not inclined to advertise their tools in the scientific journals. Furthermore, the South African elections had some unique and unanticipated characteristics. In South Africa, elections follow the principle of proportional representation, where every vote is tallied, and seats in the national parliament and provincial legislature go to parties on a proportional basis. Different systems are used all over the world. In the USA, for instance, each state has its own election apparatus, with the local media in charge of forecasting. In the UK, the constituency system of ‘winner takes all’ requires different forecasting approaches. In addition, where the bias in the order of incoming results is relatively small, as it is in most developed countries, simple statistical methods work well. Against this background a new approach had to be designed to address the peculiarities of the South African situation. A factor working in favour of the research team was that the Independent Electoral Commission (IEC) disseminated all election results at a central point. This enabled us to set up a computer system with access to the latest figures within minutes of their being released. The lack of relevant literature had an up-side: it freed us to use our creative powers to develop new methods and a system tailor-made for South Africa.

Working out the maths The use of suitable mathematical tools can help to solve relevant issues such as this one. The main challenge here was to counter the bias arising from the order in which the results were received. If we could group the population into segments that voted in similar ways, we could determine the voting behaviour of each segment from just a few results, and then multiply each of these results by the size of the electorate in the respective segments. This might give us an accurate forecast. But how to break up the electorate into suitable segments? The electorate in 1999 was made up of some 15 000 voting districts, so if we knew how people voted in the 1994 election, we could possibly cluster the voting districts into groups of similar voting behaviours. Unfortunately, because the 1994 elections were unique in several ways and different from those in 1999, we couldn’t use the 1994 results. We knew, however, that voting patterns in South Africa closely follow demographic lines, so another option was to segment the electorate by race and other demographic features. Since the Census results of 1996 were available for each voting district in 1999, we chose this as a basis for our segmentation. The next question was: how to use the available Census data to segment the population? This can be done by means of a statistical technique called ‘clustering’2. Clustering is best described by using the language of mathematics, which is extremely efficient in expressing vague ideas crisply and clearly, avoids the ambiguities inherent in normal human speech, and also often leads to the derivation or discovery of new relationships that normally remain hidden. Clustering aims to combine similar objects in clusters. So how do we define ‘similarity’ in mathematical terms? We can do it by defining a ‘distance’ between items (in this case, between the voting districts). This is not a physical distance, but an abstract one between points in a multi-dimensional mathematical space. The ‘dimensions’ are

1. Work on such a project involves a team of people with diverse backgrounds. Chris Elphinstone, Jenny Holloway, and Peter Schmitz (all from the Logistics and Quantitative Methods group) and Rosalie de Villiers (from the Meraka Institute, CSIR) were involved in discussions of the model and the execution of the project in real time. Hans Ittmann (manager of the Logistics and Quantitative Methods group) took charge of interactions with the media on election night. 2. We used a new technology called fuzzy clustering, whereby objects have a shared membership of each cluster, with the largest membership allocated to the cluster closest to the object. The more commonly used kk-means method assigns discrete memberships of one or zero to objects. Fuzzy clustering has distinct advantages over these traditional approaches, which are most easily explained in mathematical terms. However, even without resorting to mathematics, it is plausible that fuzzy memberships will give more flexibility and more accurate descriptions.

24 Quest 2(3) 2006

Municipal elections, 1 March 2006. Photographs: Cyclops

Example of a cluster This figure illustrates the largest of the 20 clusters that we used in the 2000 (and 2004) model. The cluster is characterized by its average Census characteristics, but it was constructed from the 1999 election results and not from the Census. This cluster is very big as it contains 38% of the electorate. Most (94.8%) of this group voted for the African National Congress in 1999. In the second-largest cluster, most people also voted for the ANC (87%), but some supported the United Democratic Movement (UDM) and the Inkatha Freedom Party (IFP). Subsequent clusters feature major support for other parties, such as the Democratic Alliance (DA).

Indian 0.1% Coloured 1.0%


Age, Education and Income 3.0

White 0.5%


National Elections 1999 PAC UDM 0.9% 1.5%

1.0 0.0 -1.0

Black 98.4%


Education Income

ANC 94.8%

Language 25 20 15 10 5













VOTER MEMBERSHIP: 7 876 380 (38.1%)

Getting the results right In 1999, the results of our predictive method were encouraging, in that the bias inherent in the order of the actual results was largely eliminated by our use of the clustering model. After a few hours, we already had results that deviated very little from the final results. Of course, we did not know this on election night, so we hesitated at first about advertising our predictions too widely! To quote some results: the final African National Congress (ANC) proportion of votes in the 1999 national elections was 66.2%. When results from 20% of the voting districts were in, we predicted 63%, even though the actual results released by the IEC up to that point still indicated a modest 50%. So the error in our predictions at the time was a factor 4 smaller than the deviation between the actual and the final result. Another interesting output of the model was the identification of strange or unexpected results, but none raised a sufficient number of eyebrows to warrant further investigation. ▲ ▲

defined according to different attributes of the voting districts – that is, according to demographic characteristics. The 1996 Census, which we used in 1999, contains, for each voting district, data on race, language, income, education, rural/urban ratio, average age, gender, and geographical position (latitude/longitude). But which of these factors play a big role in determining the voting behaviour? Although we suspected that race was a major player, we used all the factors evenly to define an overall ‘distance’. The next question is: how to convert differences in these attributes into numbers, and combine these numbers to give an overall distance? Consider, for example, education. We assigned an ascending set of numbers to increasing educational levels. If two voting districts have similar educational levels, then their educational coordinates are close and the distance (the difference between the assigned numbers) is small. Quantification of income is pretty obvious, as it’s expressed in terms of rands. Less clear, however, is how to combine these distances in a single overall distance. Putting equal weight on each factor requires normalizing each attribute, so the allowed scale is always from zero to one, or any other fixed range. This leaves us with other quantities to deal with (such as language) called ‘categorical variables’. There’s no way to assign different languages to a linear scale, so we have to use another prescription: each language is treated as an independent variable (dimension), whose value can be one or zero. Once an overall ‘distance’ between voting districts is defined, many ways remain to carry out the clustering process mathematically. Although commercial software packages were available, we needed access to the detailed results of the clustering process, so we developed the software ourselves3. The clusters are constructed before the elections, so that during election night the cluster memberships of each voting district can be accessed by the system. Once partial results in each cluster have been obtained, one can lay out these results over the whole cluster and predict the final outcome of the elections. With the addition of new results, the cluster results become more accurate – and so do the predictions.

3. Commercial software often lacks the accessibility of home-grown software, although it has other advantages, such as fancy graphics, which were not very important in this exercise.

Quest 2(3) 2006 25

Dr Jan M. Greben is with the group for Logistics and Quantitative Methods in the CSIR Built Environment unit. He has worked as a theoretical physicist in the Netherlands, the USA, and Canada before joining the CSIR. He is a keen chess player and was the chess champion of Pretoria 10 times between 1984 and 1998.

During the 2004 national election the CSIR research team was sponsored by the SABC. The possibility of conveying our results directly to the media added to the excitement of being involved. To illustrate some of the outcomes of the model in 2004: at 1:17 a.m. on 15 April 2004, when only 2% of the returns had come in, the SABC could report, by means of our model, that the ANC’s national results would be 69.6%, and that the ID would get 1.7%. (The IEC totals at the same time showed 60.5% for the ANC and 2.6% for the ID.) Both of our predictions proved accurate when the final results were known. A graph of the predicted results as a function of time (or votes counted) is shown below.

National votes for ANC in 2004 73 72 71 70 69 68 67 % Votes

Progress to date For the 1 March 2006 municipal elections, we updated our clusters by using the 2004 election results as a basis. Our predictions were of similar accuracy to those in the previous elections. With 10% of the results in (as was the case at about 3:00 a.m. of election night) we predicted that the ANC would end up with 66.5%, the DA with 17.2%, and the Independent Democrats (ID) with 2.4%. These numbers were close to the final results (65.7%, 16.3%, and 2.2%, respectively) at a time that the official IEC results were still far away from these figures (62.6%, 22.6%, and 3.4% respectively). A new element in our analysis in the 2006 elections was the possibility of predicting the trends in real time. Mathematically, trends can be defined by relating old and new election results. These trend analyses provide useful material for political analysts at a time when the media are still focusing on the elections. We intend to improve the accuracy of these trend predictions in future and to encourage their use in political discussions so as to ensure a better interpretation of electoral intentions. Too often, the politicians decide what is good for the electorate – with this tool, it might become more difficult to ignore voters’ wishes. This election-night forecasting project is a good example of how to combine mathematical models and real-time analysis. It’s a challenge to make sure that the software is operating correctly in real time, for instance, when only parts of it can be tested beforehand. The project also illustrates the continuous progress in science: despite the early success of our election tools for South Africa, one keeps looking for further improvements and extensions that might raise the quality of political processes in our country. ■

Accurate predictions


• Actual • Predicted

65 64 63 62 61 60 59 58




15 20

25 30

35 40 45 50 55

60 65

70 75

80 85 90 95 100

% Votes in

The DA’s performance was harder to predict, but the results were also satisfactory.

National votes for DA in 2004 26 25 24 23 22

• Actual • Predicted

21 % Votes

The IEC again tasked us with election-night forecasts in 2000. Since the 1999 election results were available, and the 2000 voting districts were not very different from those of 1999, we could use their results to cluster the 2000 voting districts according to voting behaviour. The election results were already numerical so we had fewer problems with quantifying the distances. We constructed a new set of clusters based on the 1999 election results and, since this procedure is more objective than that based on the Census data, our predictions in 2000 were even more accurate.

20 19 18 17 16 15 14 13 12 0



15 20

25 30

35 40 45 50 55

60 65

70 75

80 85 90 95 100

% Votes in

Quest 2(3) 2006 27

A termite tale of climate change Once upon a time the Namib was wooded savanna. What turned it into desert, when, and how? Fossil termite hives offer clues, writes Martin Pickford.


hroughout sub-Saharan Africa and other parts of the world, termites play a huge ecological and geological role. Evidence of their activity is widespread, but geologists seldom appreciate the work that these social insects do because most of it lies hidden beneath their feet and because it is often difficult to identify the species responsible. The discovery of fossilized harvester termite hives in the arid coastal regions of Nambia, however, is now helping palaeontologists to piece together the story of massive climate change that occurred millions of years ago. Signs of activity Buried stone lines in the soils and palaeosols of the tropics, which are frequently seen in roadcuts, are normally caused by the busy work of termites. For millennia they have been carrying clay, silt, and sand upwards in

Termite activity in ancient sand dunes of the Namib Sand Sea near Awasib, NamibNaukluft Park, Namibia. The lack of a termite mound above the disturbed sediment indicates that this complex may have been made by Hodotermes, although no hive is exposed in the deposits to confirm the identification.

Termite earth-movers A vast biomass of termites inhabits the continent of Africa. Many species spend most of their lives underground or hidden in the depths of trees, so it’s easy to underestimate just how many there are and what they do. They use a wide range of plant material, often to make subterranean fungus gardens, where they grow the small mushrooms that are their food. Taken together, African termites consume more vegetation than all the mammalian herbivores put together. Many termite species build mounds that poke up above ground as low hillocks or tall, chimney-like constructions. Most of us have seen and been impressed by these mounds, but few people realize that, taken together, termites do more geological work than any other living creatures except perhaps for the humble earthworm. Termites in Africa collectively move millions of tons of earth every year, which makes them important geological agents. To find out more, consult Vivienne Uys’s A Guide to the Termite Genera of Southern Africa (see page 29).

28 Quest 2(3) 2006

the soil, then dumping it at the surface as small earth-lined foraging tunnels or as larger mounds and chimneys. As a result, over long periods of time, stones tend to ‘drift’ downwards in the soil profile, accumulating at or near the local water table, which termites cannot penetrate. Under normal circumstances, termites are almost never preserved as fossils – their bodies are too soft and they decay rapidly1. Above ground termite mounds are seldom preserved as fossils because they tend to erode away before being buried. If they are buried rapidly enough, however, the mounds and tunnel systems that these insects make can be fossilized, as happened in East Africa when lava flows and volcanic ashes blanketed the countryside, and in the Namib Desert when sand dunes over-rode and buried these structures. Fossil termite mounds have been recorded in South Africa, Namibia, Tanzania, Kenya, and West Africa. But published records are rare, and the makers of the mounds are seldom, if ever, precisely identified. Termite foraging tunnels (called Termitichnus, originally described from indurated [hardened] sand dunes in Namibia) 1. Although termite fossils are extremely rare, some have been recorded in amber (or, fossilized tree resin) and in shales in various parts of the world. In Kenya, for example, there is a shale deposit at Waril in the Tugen Hills aged 10 million years, which preserves imprints of the wings and bodies of alate termites that fell onto damp silts exposed along an ancient lake shore.

Opposite (left): Harvester termites, Hodotermes mossambicus (Hagen), in a circle of grass that has been harvested for food. Photograph: ARC-Plant Protection Research Institute

Opposite (right):Termitichnus is the scientific name for fossilized foraging tunnels made by termites at Rooilepel, Namibia. Here the tunnels have been filled with sand, and the surrounding dune sediment has been blasted away by the wind, exposing the ancient tunnel system in three dimensions in the position in which it was formed.

Above left: An almost complete hive of the harvester termite Hodotermes, exposed at Grillental, northern Sperrgebiet, Namibia. This specimen has the polar (top) part preserved. The presence of a hard sandstone deposit behind the fossil protected the upper part from destruction by wind-blown sand. Above right: Ball-shaped structure now preserved in sand, originally constructed at Rooilepel, Namibia, by the sand termite Psammotermes some 16 million years ago.

cell walls can become are often preserved, but on impregnated by calcite, and their own they’re difficult to this preserves them as fossils. tell apart from tunnels made Not only can individual by ants or other insects. hives become fossilized but, It’s easier to identify the on occasion, so can entire maker if the termite mound polycalate complexes. is preserved. If the fungus Usually however, only garden is preserved (as the outer parts of the hives happened in East Africa), it get calcified: the centres are is sometimes even possible left unaltered and decay to determine the family of away to leave hollow, balltermites responsible. Detail of the cell structure of a Hodotermes hive from Grillental, shaped fossils. As erosion Probably the most makes them weather out confidently identified termite Namibia, showing the large cells connected to each other and arranged into horizontal shelves. of the ground, they often constructions are the hives form ring-shaped structures, of the harvester termite surface mounds. Instead, they build almost like an inflated inner tube lying Hodotermes, which are not fungus spherical underground hives up to on the ground. This is because the gardens but the nests where young half a metre in diameter from ‘carton’ upper ‘polar’ part of the sphere, being and adult termites congregate. (a mixture of organic and inorganic the first bit exposed, erodes away or Namibia’s Hodotermes material with the consistency of brittle breaks into small pieces, leaving the fossil hives cardboard), consisting of masses lower ‘equatorial’ part of the hive The coastal area of Namibia is of large cells arranged roughly into exceptional because evidence of horizontal layers or ‘shelves’, with each intact. This section resembles a ring. ancient termite activity abounds, and cell having open connections or ramps Many such termite rings have been recognized in the Sperrgebiet and the at least two kinds of termites can be to its neighbours above and below. Namib-Naukluft Park in sediments identified from the traces that they There is often a vertical tube running ranging in age from 20 million years to left. In the fossil sand dunes of the through the hive from top to bottom, Recent times. Namib Sand Sea are the ubiquitous providing a sort of ‘highway’ for rapid access to various parts of the hive. Hodotermes is one of the few ‘polycalate’ termites. This means that one group builds several hives, up to a metre beneath the surface, which are connected by underground tunnels. The hives of these termites normally decay when the insects abandon the complex but, in the right conditions, such as those in the Namib, the

Habitat for harvesters As their name implies, harvester termites eat grass, so they occur most often in places where grass is plentiful, such as steppe, savanna, and wooded savanna. They are rare or absent in forested areas where the rainfall is greater than 750 mm per year, and they cannot survive where there is less than 125 mm annual rainfall. In ▲ ▲

(and often well preserved) hives, foraging tunnels, and networks of the sand termite Psammotermes. But perhaps the most unexpected fossils are the calcified hives of the harvester termite Hodotermes, which occur in the Early Miocene (20 million years old) deposits of the Sperrgebiet and elsewhere in the Namib. Harvester termites do not construct

Quest 2(3) 2006 29

Winter and summer rainfall belts of southern Africa, and the distribution of the harvester termite, Hodotermes. It occurs mainly in the summerrainfall belt, but encroaches into the margins of the winter-rainfall belt in some areas (usually where intermittent rainfall sometimes occurs in summer). But fossil hives of Hodotermes occur deep inside what is today the winter-rainfall belt, attesting to climatic change in the past.

Changes in global climate brought about by growth of the polar ice caps, first the Antarctic one (17–16 million years ago) and, second, the Arctic one (about 8–7 million years ago). The ecoclimatic belts were pushed northwards by growth of the Antarctic ice cap and then squeezed back equatorwards by growth of the Arctic ice cap. This explains why the southern Namib enjoyed summer rainfall 20 million years ago but, since 16 million years ago, has lain mainly in the winter-rainfall belt.

experience winter rainfall? How could addition, they are found naturally only there are fossils of mammals such as the climate have been so different in at the edges of winter-rainfall areas, pigs, water chevrotains, dassies, and the past, with the coast of southern because the reproductive season, proboscideans, whose teeth indicate Namibia located in the summerwhen alates leave the nest, is timed that they consumed leaves and fruit, rainfall belt? to occur during the summer rains so there must have been substantial Before the onset of the Namib when the climate is warm. When a stands of trees in the region 20 million Desert 17–16 million years ago, the fossilized hive of Hodotermes is found, years ago. All the indications together Antarctic ice cap was considerably therefore, it indicates that, at the time yield a picture of a wooded savanna smaller than it is today. Indeed, it was constructed, the region most with about 750 mm annual rainfall. ice was initially confined to likely enjoyed summer rainfall mountain tops. But about of between 125 and 750 mm 17 million years ago, towards the per year. end of the Early Miocene epoch, Abundant fossilized global atmospheric cooling Hodotermes hives have been brought a cover of ice to much discovered in Early Miocene larger areas until, by about deposits in many parts of 16 million years ago, the Namibia’s coastal region, continent was entirely buried ranging from the Orange beneath a vast ice cap. River Valley at Arrisdrift The growth of this continental and Gypsum-Plate Pan near Harvester termites at the entrance to the nest, to which they take ice cap forced the warm Oranjemund in the south, to food that they have collected. Photograph: ARC-Plant Protection Research Institute ecoclimatic belts of the world Grillental nearer Lüderitz in northwards. Southern Namibia, the north, and even further which lay in the summer-rainfall belt This palaeoenvironmental north in the Namib-Naukluft Park at prior to Antarctic ice cap expansion, reconstruction is confirmed by a Awasib, and elsewhere. The finds are now came to lie within the winterstudy of the palaeosols (fossil soils) a sign that this zone once enjoyed the rainfall belt as the summer-rainfall belt that yield the fossils and in which summer rainfall in which Hodotermes was pushed towards the equator. It also Hodotermes built its hives. These soils flourishes – in strong contrast with experienced far less rainfall and became are rich in calcrete nodules and even today’s winter-rainfall pattern of the calcrete sheets. Today, similar calcretes hyper-arid, leading to the onset of desert Sperrgebiet, with far less rain (less conditions by 16 million years ago, and form in areas such as Grootfontein than 100 mm a year) than fell there resulting in the Namib Desert. and Etosha in northern Namibia, during the Early Miocene. At that time, there was no land where the rainfall is up to 750 mm per In the Sperrgebiet, the fossil hives near the north pole on which ice year, but where evaporation rates are of Hodotermes are associated with could accumulate, so there was no high (over 2 000 mm per year). other animal remains, notably the counterbalancing ice-cap in the north. bones and teeth of various mammals, What made the climate change Only about 8–7 million years ago did birds, reptiles, and amphibians. Many How can one account for the presence the global climate cool sufficiently for of these confirm the region’s summer of fossil constructions made by a northern ice cap to form, first on rainfall palaeoclimate2. summer-rainfall termites such as land masses such as Greenland and In the same deposits, for instance, Hodotermes in regions which today then on the sea itself, culminating 2. Bones and skulls of clawed toads (Pipidae, Xenopus) are common at Langental and Grillental. Today, these toads occur widely 5–4 million years ago in the Arctic in sub-equatorial Africa where there is permanent fresh water in open countryside, but they generally avoid winter-rainfall ice cap. The growth of the Arctic ice areas and forest. The diminutive extinct ostrich Struthio coppensi, the earliest known in the world, occurs at several sites, notably at Elisabethfeld, near Lüderitz. It is likely that this species included grass in its diet, just as extant ostriches do. Several cap forced the ecoclimatic belts of the of the small mammals, notably some of the rodents and elephant shrews, have hypsodont (high crowned) cheek teeth, world back towards the south, thus indicating that their diets consisted mainly of grass. The early antelope Namibiomeryx probably also included grass in its diet, squeezing them towards the equator. as did the rhinos, and other ruminants such as Sperrgebietomeryx, a peculiar creature distantly related to giraffes and antelopes.

30 Quest 2(3) 2006

Q News

More on termites Termites, writes Vivienne Uys (in her excellent, easy-to-read, illustrated Guide to the termites in our region) are secretive, discreet, elusive; dull in colour; and not easy to identify. Feared by farmers and builders as pests that destroy crops and structural timber, they benefit soil quality in tropical ecosystems through processes such as the decomposition of organic material, and they’re a valuable food source for predators (including humans). In southern Africa, there are on record 54 genera, in five families. This Guide summarizes current knowledge, and is used by people in fields such as termite systematics, natural resource management, ecology, and pest control. Vivienne Uys, A Guide to the Termite Genera of Southern Africa. Plant Protection Research Institute Handbook No. 15. (Pretoria: Agricultural Research Council, 2002). To order, phone Eunice Mudzusi on (012) 808 8222 or e-mail

The combined geological and palaeontological evidence underscores the hints provided by the fossil termite hives – that, 20 million years ago, the Sperrgebiet was not a desert but was covered in wooded savanna with permanent bodies of fresh water. Not only have termites played a major role in Africa for the last 20 million years, but their fossilized underground constructions can be studied by palaeontologists to throw light on past climates. So don’t fly into a rage the next time termites eat your books or the wooden floors in your house. Instead, ponder on the fact that they have been around a lot longer than humans have, and consider the possibility that they may well still be here long after we’ve disappeared. ■ Dr Martin Pickford is Maître de Conférences in the Chaire de Paléoanthropologie et de Préhistoire, Collège de France, and is a member of the Département Histoire de la Terre, UMR 5143 du CNRS, Paris. He has conducted palaeontological research in many African countries and is particularly interested in the palaeoenvironments and palaeoclimates in which evolutionary processes occurred. For general information, see A. Emerson, “The Biogeography of Termites” in the Bulletin of the American Museum of Natural History, vol. 99(1952), pp. 217–225. For more on Hodotermes, read W. Coaton’s “Fossilized nests of Hodotermitidae (Isoptera) from the Clanwilliam District, Cape Province” in the Journal of the Entomological Society of South Africa, vol. 44(1981), pp. 79–81. Also read W.G.H. Coaton and J.L. Sheasby, “National survey of the Isoptera of southern Africa. 10. The genus Hodotermes Hagan (Hodotermitidae)” in the Namibian Entomological Journal Cimbebasia, Ser. A, vol. 3(1975), pp. 105–138. For details about ancient termite nests, consult J.M. Moore and M.D. Picker, “Heuweltjies (earth mounds) in the Clanwilliam district, Cape Province, South Africa: 4,000 year old termite nests” in Oecologia, vol. 86 (1991), pp. 424–432.

Mars rocks For the past few weeks, the Spirit rover has been examining layered rocks exposed on the surface of Mars on the edges of the area known as ‘Home Plate’. Images of the most complex layering patterns seen so far (right) offer a record of the geological history of the formation of the rocks. In the centre of the image, one group of layers slopes downwards to the right, whereas the layers above and below the group are more nearly horizontal. Where layers of different orientations intersect, other ones are truncated, indicating that complex patterns of alternating erosion and deposition were occurring when these layers were being deposited. Scientists suggest that the rocks were formed in the wake of a volcanic explosion or impact event, and it’s possible that wind may also have played a part in redistributing the materials after this kind of event. Above right: A false-colour rendering of rock layers on Mars, combining separate images through different filters taken by the Pancam (panoramic camera) on 8 March 2006, and enhanced to emphasize colour differences among the rocks and soils. Image: NASA/JPL-Caltech/Cornell

Right: A composite picture of the Syrtis Major face of Mars (from daily images of the planet taken previously), showing how Mars was looking around mid-March 2006. It travels 360° around the Sun in one martian year. Ls (solar longitude) is the measure of the time of year on Mars. The year begins at Ls 0° and lasts 686.98 days. This picture represents the season of Northern Spring (Ls 25°) and Southern Autumn. Image: NASA/JPL/Malin Space Science Systems

Signs of life on Saturn’s moon? Saturn’s tiny, icy moon, Enceladus, is a billion and a half kilometres from the Sun with a diameter of only 500 km. New findings from NASA’s Cassini spacecraft show geysers of water vapour and ice particles spewing from the region of the south pole. A possible source is liquid water boiling off from reservoirs that are perhaps just a few tens of metres below the surface, but nobody knows yet what energy source is keeping the water liquid.

There are traces of acetylene and propane in the plumes, which suggest the presence of organic compounds as well. According to Carolyn Porco, imaging team leader of the Space Science Institute in Boulder, Colorado: “If we had done nothing else, these findings alone would have made the Cassini mission worthwhile”. Cassini will pass Enceladus again in 2008 and offer the chance to learn more. Reported from Science in New Scientist (18 March 2006).

The unpredictable curved ball Why does putting spin on a ball make it hard to hit (if you’re batting) or to catch (if you’re fielding) or to block (in a soccer game)? Psychologist Cathy Craig, at Queen’s University Belfast in the UK, says it’s because human vision is not equipped to follow the course of a fast-spinning curved ball. She tested the ability of experienced players to anticipate the trajectories of balls with side spin. She asked them to watch a virtual reality display that simulated shots with a spin of 600 revolutions a minute and then predict whether or not they would

end up in the goal. Even professionals could not say how the spin would affect the trajectory. The spin side on the ball produces a Magnus force, explains Craig, which accelerates the ball in a direction that human eyesight seems unable to process. Evolution made it necessary for us to anticipate gravity’s effect on moving objects, but spinning balls don’t occur in nature. “Why would nature bother having a visual system that’s adapted to them?” Craig concludes. Reported from Naturwissenschaften in New Scientist (4 March 2006).

Quest 2(3) 2006 31

Q Viewpoint

Is evidence overrated? Jonathan Jansen considers the admissibility of evidence.


The question of political context There is a difference between ‘hard’ and ‘soft’ advocates of evidence, not simply with respect to style and approach, but also in their approaches to truth, culture, and the nature of knowledge.

Interrogating ‘evidence’ Here are some serious questions that arise about the relationship between research and policy, and about the theoretical, methodological, and political problems of evidence. ■ Is ‘scientific evidence’ (such as that derived from careful methods such as meta-analysis) not simply one of many sources of authority in making social or medical decisions? ■ What is the role of judgment in decision-making? ■ What should be the status of traditional beliefs, values, and understandings when faced by scientific claims and conventions? ■ Does evidence-based policy not in fact privilege causal modelling and the material or physical world? ■ Given the instability of evidence as new knowledge replaces old, how should strident claims about evidence be evaluated? ■ How ‘culture-dependent’ is evidence? How well do claims about evidence travel across contexts and cultures? ■ How does evidence-driven policy or practice deal with uncertainty? ■ As claims as to ‘the best evidence’ vary across disciplines – for example, in law compared to medicine – what are the implications for universal or sciencebased claims about evidence? ■ How valid are non-scientific (nonrational?) sources of evidence? ■ How neutral is evidence-based policy and practice? And how far do power, politics, and political interests (funding agencies, pharmaceutical companies, Western governments, and others) in fact determine the choice of questions, the design, and even the outcomes of scientific inquiry? ■ What ethical and moral problems arise with evidence-based scientific methodologies (such as randomization) when they’re applied to humans?

The soft advocacy of evidence can demonstrate sensitivity to, and accommodation of, matters of ethics, autonomy, complexity, judgment, and transparency in randomized controlled trials in the search to establish ‘the best evidence’ for a particular medical intervention, for example, as well as humility and concern for social justice not always apparent among the hard advocates of evidence. But it can, nevertheless, underestimate the political context of evidence and evidence-based pursuits3. Evidence is never neutral; it operates within a political context especially when it advises people in power. The very questions posed, and the designs followed, predispose research towards particular kinds of evidence4. On the local scene, nothing demonstrates more powerfully the politics of evidence than the response of authorities to research questions and results, on topics such as the following: ■ how many teachers are HIV positive in South Africa? ■ how much mathematics do South African teachers actually know? ■ how does South Africa rank in primary school science compared to other African countries? (TIMSS-R) ■ how many new teachers does South Africa really need? Of course, it needn’t be an education question that draws political fire. Try this one on the powerful: ■ does South Africa have more or less crime today than in 1994? Few would deny that the results of commissioned research, whether from within or outside government, are always subject to political intervention. This might not mean direct censorship or incarceration, for new democracies (such as South Africa) have a subtler response to disconcerting information. It takes the form of delay, editing, claims of uncertainty, ▲ ▲

t first blush, the arguments of evidence-based advocates seem deceptively simple: public policy and professional action should be based on, or informed by, the best available evidence preferably derived from systematic reviews (also called research synthesis or meta-analysis). To argue against such a position seems irrational. Who would deny that decisions based on ‘the best evidence’ should inform policy deliberation and professional action across every sphere of decision-making including education, medicine, crime prevention, economics and urban planning? The idea of ‘evidence’ has taken on new significance in universities, governments, science academies, and professional associations, and among practitioners. With its origins in the field of health care1, an evidence-based or, as some prefer, an evidence-informed movement has gathered pace in major parts of the world. In the USA, for instance, federal funding for new programmes has become conditional on the use of ‘randomized control trials’ in areas ranging from health care to education. At the Institute of Education, University of London, an evidence-based unit has been set up in the social sciences: the Evidence for Policy and Practice Information and Coordinating (EPPI) Centre. And there is an entity hosted by the Medical Research Council (MRC) in Cape Town that builds capacity in the science of research synthesis, prepares systematic reviews of the effects of health care interventions, and promotes evidence-based health care practice and policy in sub-Saharan Africa. However, the stridency and tone of certainty of some of the advocates of evidence-based decisions have raised probing questions and even reservations among ‘hard’ and ‘soft’ critics of this movement. What follows are personal reflections on, and criticisms of, some of the powerful positions in the ‘evidence’ debate2.

1. The evidence-based approach has been spearheaded by powerful international groups such as the Cochrane Collaboration (named after the British epidemiologist, Archie Cochrane), to which the EPPI Centre in London is closely linked. The entity hosted at the MRC in Cape Town that focuses on evidence-based activities is the only accredited Cochrane Collaboration entity in Africa. 2. The Academy of Science of South Africa (ASSAf) held a double symposium on “Evidence-based Practice: Problems, Possibilities and Politics”, in Pretoria on 3 March 2006. The first symposium was entitled “Is evidence overrated?” This article is a set of personal reflections by the author on the papers and positions presented there. 3. The brilliant keynote presentation at the ASSAf symposium by the University of Sydney epidemiologist, Les Irwig, was an example of the soft advocacy of evidence, from a medical point of view. 4. That the questions posed and the designs followed predispose research towards particular kinds of evidence is a point long since established by eminent scientists such as Evelyn Fox Keller and Sandra Harding in their celebrated work on gender and science.

Quest 2(3) 2006 33

Viewpoint Q

the questioning of methodology (if not the methodologist), and the citation of rival studies. One of the commonest words in the post-1994 South African lexicon is “moratorium”. That a modern, sciencecommitted state like South Africa could even question, from within the corridors of power, causal agency in the HIV-to-AIDS debate, continues to puzzle the science community and the lay public alike. The claim, therefore, that ‘the evidence speaks for itself’ is without foundation. The issue, rather, is: Who speaks for the evidence?

counts? Our lenses can be criticized for focusing on the seen, material world, and for failing to value evidence in relation to the people that the evidence claims to serve – and also for overvaluing institutions from which evidence comes. What kinds of institution are valued, and deemed fit to pronounce on evidence? By whom are they given such status? In this context, the question of evidence is inescapably a question of power. Such a variety of perspectives mean that disciplinary contexts matter a great deal in speaking about evidence. A common example of how complicated it can get is the body of classic studies on the relationship between class size and student achievement. Probably no subject has received so much attention in the field of education through the application of metaanalytic methods. Yet, depending on which body of research you study, the results are mixed. Why? What are the problems?

The question of disciplinary context How is the subject of evidence treated in different fields – such as law, education, and anthropology5? Lawyers6 think about evidence very differently from researchers in the health sciences in general and in the evidencebased movement in particular. The standards of evidence are different, for legal minds argue by precedent and ‘on the balance of the evidence’. In courtrooms, evidence presented and decided on is Objectivity deliberately selective. Judgment needs to be rescued draws on, but is not confined to, from its laboratory pretence of expert opinion. And, crucially, being universal and timeless, and evidence is led in an adversarial context in which ‘the legal hired redefined as ‘intersubjective hand’ is a much greater factor than agreement’. objective rules of balanced reporting. The legal use of ‘evidence’ leaves one with the intriguing notion of eminence-based decision-making in the field of law. In the field of education7, policymakers One – the complexity problem. The remain wary of research, partly because sheer number and complexity of variables of frequent lack of rigour in social (teacher experience, national culture, sciences fields. But the problem also lies subject matter, teacher qualifications, much deeper in the antagonism between pedagogical strategy, and many more) scientific and practitioner modes of inquiry. involved in seeking to establish the Educational inquiry is charged with holding relationship between the number of poorly codified practices, compared to the students in a class and the level of natural sciences, which therefore offer little academic achievement attainable, are potential for generalizing from the results formidable. of a single study. Yet there appears to be Two – the compositional problem. It growing consensus about the importance depends on who’s in the class in the first of evidence (as well as its limits) in place. Is it middle-class children with high professional activities such as evaluation. levels of cultural capital, or poor children An anthropological perspective8 with illiterate parents? Random assignment uncovers powerful, and often unspoken, of groups doesn’t begin to deal with this connections between evidence, rationality, challenge, especially when the study is and the world of research through riveting conducted in developing contexts whereas questions such as: Who makes the claim the standard research claims are being to have ‘the evidence’? Whose evidence made elsewhere. 5. Three non-medical science scholars at the ASSAf symposium responded to the address by Les Irwig with perspectives on the subject of evidence from their respective fields – law, education, and anthropology. 6. The points highlighted in this paragraph come from the paper by Thea Illsley (University of Pretoria) at the ASSAf symposium on the way in which lawyers think about evidence. 7. The points in this paragraph were raised by Johan Muller (University of Cape Town), who presented arguments and a case study from the field of education at the ASSAf symposium. 8. Anthropologist Mpilo Pearl Sithole (Association for Rural Development), who raised the issues given in this paragraph at the ASSAf symposium, offered three main charges against evidence: the absolutism of evidence, the abstraction of jargon, and the privileging of institution.

34 Quest 2(3) 2006

Three – the curriculum problem. By considering nothing but class size in the research design, and ‘holding constant’ what’s being taught, there is more than methodological finesse at play. Students could be achieving well on an outdated or offensive curriculum, for instance, yet the significant variable of curriculum content is thereby neatly set aside in the pursuit of simple causality. Four – the ethical problem. The research findings come in, then what? What do they mean for poor schools, operating in contexts where the results are simply irrelevant because class size is overwhelmingly determined by the national education budget rather than by definitive research results. Five – the political problem. To whom does class size matter? Sure, the test scores might determine that class size within broad limits (say, no more than 50 students per class) does not matter. But I’ve yet to meet a teacher experiencing the daily toil of classroom life who would find any meaning in such a result. This kind of systematic review of class size effectively has to sacrifice the power of experience and the emotions of teaching for a sanitized account of what, or rather who, counts. I raise the example of class size and student achievement to suggest that where human actors are involved in an endeavour as exhausting as teaching, such a research context is infinitely more complex than that of a simple inquiry to establish if drug X is better than drug Y in dealing with tuberculosis. More questions than answers, and that’s a good thing The health sciences give such a powerful context for the evidence debate because the consequences of intervention (or non-intervention) are immediate: the subjects could, quite literally, die. The ‘harm’ versus ‘good’ question in evidence debates therefore turns the heat onto those who dare question the activity of ‘getting to the truth.’ In such pursuits, however, it’s possible to overlook two significant questions about knowledge, knowing, and power that lie at the heart of the dispute. First, is rigour possible only within systematic review? To answer ‘yes’ constitutes striking arrogance. It means ignoring the methods of inquiry and perspectives on knowledge that have greatly enlightened scholarship across the disciplines in the past 60 years. It accepts the tyranny of positivism to the exclusion of other ways of knowing and of knowing in

Q News different contexts. It refutes the notion that evidence, like the truth, has many faces. Second, what social questions are ignored by the way in which evidencefocused inquiry is pursued? What is ‘held constant’ in class size research can often be much more important than the variables we choose to play with. And so, on to the future… I propose, as have others, the need for humility in both advocates and critics of evidence. Clearly we need evidence; and there is no question that in certain fields, like education and the social sciences more broadly, the degree of rigour and credibility in research remains a major problem. But we should also recognize that what counts as ‘best evidence’ is itself a matter adjudicated by human subjects, within the privileged status of particular kinds of institution that we deem fit to make such pronouncements. In this context objectivity needs (again) to be rescued from its laboratory pretence of being universal and timeless, and redefined simply as ‘intersubjective agreement’. One thing that the evidence evangelists cannot refute is this: the truth, historically, is unstable, and what counts as fervent knowledge claims in one generation of medical practice is typically scorned among the next generation of physicians. Yet, frozen in time, each generation would proclaim its truth with a frightening certitude. We are in danger of taking the social, natural, and medical sciences back into the epistemological dark ages if zealotry rather than humility defines the terms of the evidence debates. ■ Professor Jansen is an outspoken leader in the field of education in South Africa. A former high school teacher, he has a doctorate from Stanford and an honorary doctorate from the University of Edinburgh. He is Dean of Education at the University of Pretoria and Vice-President of the Academy of Science of South Africa. There is more to read on the evidence debate. For the point of view of a hard advocate of evidence, consult Iain Chalmers’s “If evidence-informed policy works in practice, does it matter if it doesn’t work in theory?” in Evidence and Policy, vol. 1, no. 2 (2005), pp.227–242. For an example of ‘hard’ criticism, read Martyn Hammersley’s “Is the evidence-based practice movement doing more good than harm?” in Evidence and Policy, vol. 1, no. 1 (2005), pp.85–100; for ‘soft’ criticism, read Miles Little’s “ ‘Better than numbers….’ A gentle critique of evidence-based medicine” in ANZ Journal of Surgery, vol. 73, no. 4 (2003), pp.177–182. The Proceedings of the Academy of Science of South Africa’s symposium on “Evidence-based Practice: Problems, Possibilities and Politics” (3 March 2006) will be available in due course.

Galaxy clusters old and far

Responding to the meltdown

The Universe is thought to be about 13.7 billion years old. Using NASA’s Spitzer Space Telescope, astronomers have now discovered some galaxy clusters at a time when the Universe was just 4.6 billion years old. Galaxy clusters are groups of galaxies that can contain hundreds of galaxies and trillions of stars. (Our Milky Way belongs to the relatively small Local Group whose other large member is the Andromeda Galaxy.) The false-colour composite in the panel of pictures (below) shows some of the Universe’s oldest galaxy clusters. The galaxy cluster 9.1 billion light years away (lower right panel) is currently the most distant one ever detected.

US evangelical Christians

Below: Galaxy clusters between 8.15 and 9.09 billion light years away. The individual galaxies in the distant clusters show as red dots (captured by Spitzer’s infrared array camera). The green blobs are Milky Way stars along the line of sight and the blue specks are faint galaxies at various distances along the line of sight. (The blue and green data were captured by the visible-light ground-based Mosaic I camera at the Kitt Peak National Observatory in Tucson, Arizona, USA.)

In February, 86 prominent American evangelical Christians signed the “Evangelical Climate Initiative” calling for mandatory limits on greenhousegas emissions. Environmentalists and religious communities in the US often disagree about science. Now climate researchers are optimistic that, given the scientific consensus about the damage that climate change could cause, the move might help to build public support for constructive action. That, in turn, could sway conservative legislators and even the Bush administration, which numbers evangelicals among its key supporters. For many evangelicals, combating climate change became a moral imperative when they recognized that its effects could be worst in the poorest countries – through heat waves, floods, and tropical diseases, as well as rising sea levels that would drown lowlying regions and reduced agricultural productivity in areas such as sub-Saharan Africa. Calvin DeWitt, a professor of environmental studies at the University of Wisconsin, Madison, quotes from scripture, “The time has come … for destroying those that destroy the Earth.”

Image: NASA/JPL-Caltech/UCDavis/Lawrence Livermore National Laboratory

Reported in Nature (9 March 2006).


8.15 billion light years

8.59 billion light years

8.98 billion light years

9.09 billion light years

Games organized by Jochem Marotzke of the Max Planck Institute for Meteorology in Hamburg, Germany and his colleagues show that appreciation helps to persuade people to show they care about climate change. They gave €12 each to 154 undergraduates and asked them to donate none, some, or all the money to a climate change cause. They could keep any leftover money, researchers told them, and the donations would become part of a real advertisement about climate change. Students gave more when their identities were made public and others could reward them for contributing. The results suggest that people might be more willing to act in ways that publicly demonstrate their good intentions (by installing solar panels on the roof, for instance) than through anonymous taxes to fight climate change. Reported from Proceedings of the National Academy of Sciences of the USA in New Scientist (4 March 2006).

Greenland’s melting glaciers

Take group exercise for health

The ice sheet of Greenland – the second largest in the world – is losing a cubic kilometre of water every 40 hours (the same amount that Los Angeles consumes in a year). The most detailed study yet of the sheet’s ‘ice balance’ puts the total annual loss at an astounding 220 km3 – more than twice the amount lost ten years ago. Study leader, Eric Rignot, of the California Institute of Technology in Pasadena, says that “Greenland is probably going to contribute more and faster to sea-level rise than predicted by current models.” Regional warming of 3°C has caused nearly all the glaciers south of the Arctic Circle to accelerate their discharges into the ocean, and the ice sheet may be doomed. Moreover, meltwater is penetrating crevasses, and the rivers of water help the ice to slide into the ocean, speeding up the melting process even more than present models indicate.

Psychological research conducted by Elizabeth Gould and colleagues at Princeton University in the USA shows animals coping better with the stress hormones released by physical activity if they take exercise in company rather than alone. Exercise speeds up cell growth, but also raises the level of the stress hormone corticosterone, which has the opposite effect. Does social contact affect neuron development in rats after they exercise? Three groups were tested: one had no exercise; the second took exercise in isolation; the third exercised in the company of other rats. After exercise, the corticosterone levels in both exercise groups increased by the same amount. Cell growth in the brains of the sedentary group was slowest; faster in those who exercised alone; and fastest of all in those rats that had exercised in company. “Social support helps to lessen the negative consequences of stress,” Gould concludes.

Reported from Science in New Scientist (25 February 2006).

Reported from Nature Neuroscience in New Scientist (18 March 2006).

Quest 2(3) 2006 35

What’s up in the night sky? Case Rijsdijk looks at and beyond the southern skies.


ooking at the sky with the naked eye on a clear, dark night is one way of appreciating the vastness of the Universe. Immediately, questions spring to mind: How old is that star? How far away is it? Is it bigger/brighter than our Sun? Is it a planet perhaps? How can we tell the difference? Do any of the stars have planets around them? Are we alone? How did it all start? Some have been answered by astronomers, others are partially answered, and still others pose more new questions. The northern sky on a summer evening is dominated by the constellation of Orion. To us in the southern hemisphere it looks different, in fact upside down, with the ‘sword’ or ‘dagger’ pointing upwards1. The Great Nebula in Orion is a vast stellar nursery, about 1 600 light years away. Hubble images of this region show massive young stars with nearby propylids (that is, pre-Solar System clouds of gas and dust). Even an More about Orion Betelgeuse is a red supergiant: a massive star nearing the end of its life. It has a surface temperature of about 3 000 K.2 It’s about 650 light years away and, if it were in the Sun’s position, then the orbit of Mars would be inside it! Rigel is a young blue supergiant star: it has a surface temperature of about 10 000 K and is nearly 1 000 light years away. Because it is so massive, it will have a relatively short life of about 100 million years. The centre ‘star’ of the ‘sword’ or three wild dogs is, in fact, not a star at all but an enormous cloud of gas and dust glowing from the reflected light of newly formed stars.

38 Quest 2(3) 2006

ordinary pair of binoculars will reveal the fuzzy nebulosity. isiLimela, or the digging stars, were so called because when they appeared in the early morning sky it was time to start preparing the soil for planting crops. The ancient Greeks called this small cluster the Pleiades, also known as the Seven Sisters: the daughters of Atlas and Pleione. The cluster has been recognized by people all over the world. To the Japanese it is ‘Subaru’ (meaning a conglomerate or collection) while some Chinese legends refer to it as a swarm of bees. Maori and some Pacific Islands people call the Pleiades ‘Matariki’, the star cluster that heralds the start of the Aotearoa Pacific New Year. The isiLimela cluster is in fact a group of many hundreds of bright young stars about 425 light years away in the direction of the northern constellation of Taurus. Their light is reflected and scattered off the remaining clouds of gas and dust from which they formed about 100 million years ago. It appears blue because these tiny motes of interstellar dust scatter blue light more efficiently than the longer wavelengths of red light, and it is streaky because of the distribution of dust particles in space. All these stars are very much brighter and bigger than our own Sun and, as a result, will have a short life: about 1/50th that of our Sun.


Great nebula

Orion Hyades


Aldebaran isiLimela

Top left: A stunning image of the Great Nebula in Orion, M42. Its number means that it was the 42nd object in the Messier Catalogue (of 1781). Charles Joseph Messier (1730–1817) was a French astronomer who searched for comets. To save time, he made a catalogue of 103 ‘fuzzy’ celestial objects that were certainly not comets but might be mistaken for them. Picture: Anglo-Australian Observatory / David Malin Images Top right: isiLimela (or, the Pleiades). Picture: Anglo-Australian Observatory / David Malin Images

Above: A view of the constellation Orion when you look northwards in the evening sky.

For more African starlore stories and a regular feature on What’s Up in the night sky (accompanied by a map), visit the South African Astronomical Observatory’s web site Also consult the excellent, newly revised Dictionary of Astronomy edited by Ian Ridpath (Oxford University Press, 2003).

1. Perhaps South Africans should switch to indigenous lore and remember local descriptions of the night sky, such as the Tswana description of Orion’s belt and sword. The stars of Orion’s sword were dintsa le Dikolobe, three dogs chasing the three pigs of Orion’s belt. Warthogs have their litters while Orion is prominent in the sky – frequently litters of three. For this and other South African traditional views of the night sky, see Dave Laney’s article, “African starlore” in QUEST, vol. 2(2), p.34.

The life and death of stars


rion is the ideal constellation for showing how stars form and then die. Different stars in Orion are at different stages of their life cycle. Rigel, for instance, is a young, hot blue giant star; Betelgeuse is a red supergiant star towards the end of its life; and the Great Nebula is a stellar nursery. Stars with a mass similar to that of our Sun Stars form from large clouds of gas (usually mainly hydrogen) and dust. Stars with a mass about the same as that of our Sun (those on the lefthand side of the diagram) form fairly slowly. The cloud condenses, gets more and more squashed by gravity, at the same time getting hotter and hotter in the middle. Eventually the temperature becomes so high in the centre or core (about 15 million K)2, that the hydrogen in the cloud starts fusing (burning) into helium, releasing large amounts of energy. The star reaches a steady state, then continues burning and radiating energy (with its surface at a temperature of about 6 000 K), for about 10 billion years. When the hydrogen in the core is used up, the helium starts burning, the temperature rises, and the star swells up. As heavier elements, carbon and oxygen, start burning in the core, the temperature rises further and the star swells more. Because the outer layers are now further from the hot centre, they cool a little. Now the star becomes a red giant, with a core temperature of nearly 100 million K and a surface temperature of around 3 000 K. When the star becomes so big that the outer layers of the red giant are blown away, it becomes a planetary nebula. The core of the star that’s left continues to collapse and becomes a white dwarf. White dwarfs are about the size of the Earth and extremely dense (typically half a solar mass) – a bit like a teaspoonful that weighs many tonnes! They have a high surface temperature, about 50 000 K, and slowly cool to become black dwarf stars, no longer visible. The ejected material drifts off into space and mixes with the remains of other stars to form new clouds of gas and dust from which new stars will form. 2. The Kelvin temperature scale is one in which the zero point – also known as ‘absolute zero’ – is defined as equal to –273.15°C.

Stars with a mass greater The Life Cycle of Stars than our Sun Stars with a mass five or more 10 million years times that of our Sun (those on 100 million years Nebula consisting Blue giant the right-hand side of diagram) of gas and dust star Star like form from a more massive part of our Sun 10–100 million 10 billion the cloud that condenses to form years years a blue giant star. The process is similar to that of a star the size of our Sun, but Red supergiant Red giant takes less time. Also, because the star is more massive, its Ejected materials About core temperature is higher A few form new clouds 5 000 days or (20–40 million K), with a surface of gas and dust, less years which will form temperature of 10 000–40 000 K. new stars Since the temperatures are so Planetary Nebula much higher, the hydrogen fuel is White burnt much more quickly, giving Supernova dwarf star the star a far shorter life, from about 10 million to 100 million years. A few 10 000 years minutes Also, because the temperatures White dwarf are higher, successively heavier Black elements burn (such as neon, Neutron star hole Black dwarf (Pulsar) silicon, potassium, and calcium), the process only stopping when iron is formed in the core at a temperature of about 10 billion The ejected shell of the supernova K. By then, the star has swelled to eventually mixes with the remnants of become a red supergiant with a other supernovae to form new stars. surface temperature of about Out with a bang, not a whimper 3 000 K. Once iron has formed, The actual process of how stars are the core cools and the outer layers born, live, and die is in fact very collapse, causing the star to explode complex and depends greatly on the in an extremely violent explosion mass of the star. On the whole, it is known as a supernova. The energy true to say that stars with low masses released in a few minutes is more tend to live longer and end fairly than the Sun has emitted in its whole life of 10 billion years: for a short time quietly, whereas very massive stars have shorter lives and end them in the supernova can be as bright as the truly spectacular fashion. ■ entire galaxy in which it is located. Case Rijsdijk is a consultant at the South Part of the core that is left can form African Astronomical Observatory on an incredibly dense object known astronomy education and he is also a as a neutron star or pulsar. The researcher in particle physics. He brings gravitational forces are now so strong astronomy to the public through his articles that electrons are squashed into the in the popular press and his weekly SABC protons, forming neutrons. The mass radio broadcast. of about two Suns is compressed into Relative values: mass, temperatures, sizes, and a region only 20 km life-spans of stars or less in diameter. If Mass Surface Core Size Lifetime there is enough mass (Sun = 1) temperature temperature (Sun = 1) (millions after the supernova (K) (million K) of years) explosion – that is, 0.8 5 000 11 0.7 20 000 more than about 3 times the mass of our 1 5 750 14 1.0 10 000 Sun – the gravitational 1.5 7 250 19 1.4 1 800 forces in the core 2.0 9 250 21 1.6 800 are so immense that 5.0 17 000 27 2.6 80 it is squashed into 9.0 24 000 31 3.6 25 something known as 15.0 31 000 34 4.8 11 a singularity, and a black hole is formed. 30.0 40 000 37 7.1 6

Quest 2(3) 2006 39

Pilgrimages to our origins Go back in time to find out where you came from. Enjoy the new developments at the Cradle of Humankind and the wonderful Origins Centre just opened in Johannesburg. The Cradle of Humankind In December 1999, in Marrakech in Morocco, the UNESCO World Heritage Committee granted world heritage status to South Africa’s Cradle of Humankind1. Such status formally recognizes that an area has exceptional value for humanity, and it gets special protection from the global community and from its own country for the benefit of people all over the world. The Cradle comprised a neighbourhood of caves with a focus on Sterkfontein, where, in April 1947, Robert Broom discovered the celebrated “Mrs Ples”, a member of the species Australopithecus africanus and a distant relative of humanity, more

Displays at the Sterkfontein Caves of replicas of fossil specimens representing our distant human relatives.

View from Maropeng (a Setswana word meaning ‘returning to the place of origin’), the visitor centre in the Cradle of Humankind. Photographs (this page): Maropeng a’Afrika

than 2 million years old. Fossil remains of Australopithecus africanus have also been discovered at the Taung Skull Fossil Site in the North West province and the Makapans Valley in Limpopo. In July 2005, when the UNESCO Committee met in Durban, these two areas were added to the world heritage site. The Cradle, with its newest extensions, represents hominid sites of the greatest importance, and visits to it are a kind of pilgrimage, giving one the opportunity to appreciate Africa as the continent from which humanity evolved. On the basis of comparative anatomy of living primates, Charles Darwin had boldly concluded that our human ancestors must have originated on the African continent. Fossil evidence from South Africa has confirmed this view. Discovering our distant relatives In 1924, the ‘Taung Child’, a juvenile specimen of Australopithecus africanus, was blasted out of limestone sediments by miners at Taung (or, the Place of the Lion). This notable fossil was described by anatomist Raymond Dart, based at the University of the Witwatersrand. He claimed that it represented a distant human relative, with some ape-like characteristics (including a relatively small brain), but with the human-like capability of walking upright. Further discoveries of the same kind of fossil were made at Sterkfontein in 1936, when Broom described a fragmented skull. Additional fossils from this site included a partial skeleton that probably belongs to “Mrs Ples” and a large male skull of the same species (catalogued as Stw 505) discovered

by Alun Hughes under the direction of Phillip Tobias. These discoveries demonstrated to a once sceptical world that the South African fossils represented distant forebears – if not direct ancestors – of the single species, Homo sapiens, to which we all belong. “Little Foot” is an extraordinary, almost complete hominid skeleton, discovered in 1998 at Sterkfontein by Ron Clarke, Stephen Motsumi, and Nkwane Molefe, and believed to be some 3 million years old. Just one kilometre west of Sterkfontein is Swartkrans, a cave where C.K. (Bob) Brain discovered about 250 bones of animals that had been burnt at high temperatures. This is the earliest evidence for the controlled use of fire, about 1 million years ago. Kromdraai is the locality of two caves (A and B) where important fossils have been found, between 1.5 and 2 million years old. They include the first known fossil of Paranthropus robustus, a kind of ‘ape-man’, discovered by a schoolboy, Gert Terblanche, in 1938. It is now understood that this species was an evolutionary dead-end in that it died out about 1 milllion years ago, perhaps because it could not adequately compete with its contemporary, Homo ergaster. Kromdraai, Swartkrans, and Sterkfontein are three among more than 12 caves in the Cradle of Humankind, where many other fossils have been discovered, including those of sabre-tooth cats, antelope, baboons, and a diversity of mice. Such fossils are significant because they provide information about past climates and habitats and can also help with dating

1. South Africa’s other world heritage sites are Robben Island, the Greater St Lucia Wetland Park, uKhahlamba Drakensberg Park, Mapungubwe Cultural Landscape, the Cape Floral Region, and the Vredefort Dome.

40 Quest 2(3) 2006

Q The S&T tourist

Visitor information The Cradle of Humankind Maropeng, the visitor centre in the Cradle of Humankind, was opened on 7 December 2005 by President Thabo Mbeki. Its exhibits are experienced as a ‘voyage of discovery’ taking the visitor back into early geological time, through the evolution of life and humanity, up to the present. Open every day 09:00–17:00. Ticket prices: adults R65; children (up to age 14) R35; pensioners R50. For details and group bookings, phone (014) 577 9000*. The Sterkfontein Caves, 8 km from Maropeng, remain an active palaeontological site. Visitors can view the excavation area, go underground into the caves where so many important fossils have been found, and find out more about what it all means in the newly opened Museum exhibition. Open every day 09:00–16:00. Tours of the caves start every half-hour and last about 90 minutes. Ticket prices: adults R40; children R25. For details, phone (011) 668 3200. For group bookings, phone (011) 668 3200 or (011) 668 3216*. * For directions and more on Maropeng and the Sterkfontein Caves, visit The Origins Centre Situated in the University of the Witwatersrand (adjacent to the Yale Road south entrance to the campus), the Centre showcases the origins of humankind and houses an extensive collection of rock art from the University of the Witwatersrand’s Rock Art Research Institute. Open Tues–Fri 09:00–18:00; Sat–Sun 09:00–17:00. Ticket prices: adults R45; children R25; pensioners R35. Tours and educational workshops are available. For information and for group or educational bookings, phone (011) 717 4700 or e-mail For details, visit

the cave deposits. For example, fossilized mice from Way Point 160 on Bolt’s Farm, studied by Frank Sénégas, indicate that the cave deposits are 4–5 million years old. Visitors can pay homage to our distant relatives by going to the Sterkfontein caves and the Maropeng information centre nearby. There they’ll learn that humans certainly did not evolve from chimpanzees or baboons. Instead, there is excellent evidence for the existence of creatures that represent ancestors and distant relatives, which lived in the southern savanna of the continent of Africa and from which humanity evolved. We can be proud of our heritage. – Francis Thackeray Dr J.F. Thackeray is Acting Director of the Transvaal Museum in Pretoria. He has directed excavations at the Kromdraai Caves in the Cradle of Humankind for more than 12 years.

The Origins Centre The displays and activities of the beautiful new Origins Centre at the University of the Witwatersrand reveal to the public the ancient heritage of southern Africa through its magnificent rock art and the cultures of the people who produced it. The project began with an initiative from President Thabo Mbeki and was given life in 2000 by the then South African Minister of Tourism, Valli Moosa. On opening the Centre on 7 March 2006, President Mbeki described the Centre as “a tangible and creative endeavour to illuminate our minds and take us on a journey through the origins of time … to the eloquent voices of the San people.” The rock art exhibits, he

observed, represent “the world’s oldest unbroken art tradition – the San rock painting and engraving” that had been “painted on Africa’s rocks for 27 000 years” up to the early 20th century. This Centre has multiple functions. As a national museum and archive of rock art, it has grown out of the unrivalled collections of the Rock Art Research Institute (RARI) at the University. It educates at all levels – from its workshops and programmes designed for schoolgoers to the academic training offered to undergraduates and postgraduates in rock art and rock art site management. It has a crucial conservation role – given that there are an estimated 250 000 rock art sites south of the Zambezi River – and its displays, meeting and lecture facilities, shop, and café will turn it into a cultural feature of Johannesburg. Visitors to the museum can choose the language through which they interact with the rich and complex cultures they are exploring, and touchscreens and audio-visual effects offer a memorable, personal experience. Says Geoffrey Blundell, the curator of the Origins Centre, “We have tried to stimulate the senses of our visitors through the use of light and sound in the exhibits and by harnessing the power of technology [to give] a sense of returning to the roots of all humankind.” The final exhibit shows how genetic testing helps people to understand their ancestry. Visitors can buy an ‘ancestry test’, discover their genetic lineage, and have the information included in the expanding global database. The words of pioneering rock art scholar David Lewis-Williams capture the heart of this Centre:

Top: “Axis mundi”, sculpture (by Russell Scott) of a termite mound with a tree from which hang carved rock art images, celebrating the transformation process associated with the San trance dance. Above: Entrance to the Origins Centre (with members of the Moving Into Dance Company). Its motto is “We are who we are because of who we were”. Photographs: Cyclops

“By focusing on neglected histories, including rock art made by San hunter-gatherers, Khoekhoe herders and farming Bantu-speaking people, the museum will promote an inclusive view of southern African history, thus contributing to nation building and reconciliation…. Culture offers hope by bringing the wisdom of our past to bear on our future.” President Mbeki did not exaggerate when he congratulated all those archaeologists, representatives of indigenous communities, architects, landscape gardeners, filmmakers, designers, conservators, educators, artists, museum specialists, and staff who brought to fruition what he called “this magnificent gift to the nation and the world”. Visitors to the Centre will not be disappointed. ■

Quest 2(3) 2006 41

Your Q uest ions answered Q

Rainy weather QUESTION The weather in Gauteng this rainy season seems to have been highly variable and highly localized. Is this unusual? What is responsible? What makes it difficult to forecast rainfall in a particular area? ANSWER Summer rainfall over Gauteng and other inland areas of the country typically occurs via thunderstorms, either as random air-mass storms or organized within some larger weather system, such as a tropical or extratropical cloud band. In the case of random air-mass thunderstorms, rainfall varies from one location to the next across a given region, as the name implies. We don’t expect these variations to follow any particular pattern if the region is relatively homogeneous in its land cover and topography (as are some areas in the Northern Cape). While perhaps not as varied as in the Eastern Cape or KwaZulu-Natal, the topography, vegetation, soil moisture, and other land surface aspects across Gauteng certainly differ sufficiently to cause substantial variations in smallscale winds and surface evaporation. These variations would then, under most conditions, lead to changes in uplift and therefore also in thunderstorm activity in different parts of the province. A good example of variation is the Magaliesberg high ground, which would typically have an upslope wind during periods of sunshine and a downslope wind at night. Similarly, in the Hartebeespoort Dam region, a lake breeze often occurs during the day and a land breeze at night. Other land surfaces (such as mine dumps, or large high-density built-up areas) each have their own local wind systems. All such local winds could bring variation in surface evaporation, low-level moisture convergence, and thunderstorm activity – and hence differing rainfall across Gauteng, especially in hilly areas such as eastern Pretoria. Local variations in land cover can also bring rainfall variation when the thunderstorms occur as part of a larger convective weather system, such as a cloud band or a mesoscale convective complex (a very large circular area of organized thunderstorm activity). In these cases, however, one would expect the influences of local land surfaces to be weaker than those described above, because they are to some extent obscured by the larger scale winds of the convective weather system. Nevertheless, depending on the exact orientation


and track of these big weather systems in relation to local features, such as the Magaliesberg, it is certainly possible to experience substantial rainfall variation across the Pretoria or Johannesburg area during an event, and hence during the summer as a whole. Professor Chris Reason, Oceanography Department, University of Cape Town

ANSWER Perhaps it’s worth adding that, while these various effects of topography and vegetation certainly play a role, the final result of all these interactions is complex and in the end – in scientific terms – the rainfall is almost random or chaotic. The main relationship in Gauteng tends to be between rainfall and altitude, with Johannesburg getting more rain on average than Pretoria over most months, except January. January is different because, at that time of year, the tropical air-masses dominate the region. In a tropical air-mass there is not much wind shear in the vertical (that is, little change in wind speed and direction with height) nor are there strong steering winds, so the thunderstorms that do develop don’t move far and we get ‘patchy’ rainfall spread. In these conditions, valley winds and other local influences become more dominant as a forcing feature. January and February 2006 was a particularly ‘tropical’ season, which may give the impression of patchy rainfall. In fact, the wetter the season the more variable the rain becomes. Dr Warren Tennant, South African Weather Service Send your questions to The Editor (write S&T QUESTION in the subject line) by e-mail to OR by fax to (011) 673 3683. Please keep them as short as possible, and include your name and contact details. (We reserve the right to edit for length and clarity.) We will send you R80 for every question that is published with answers from our experts.

News Q Where are the data on avian flu? Inadequate information could undermine the battle against the spreading birdflu virus, because open sharing of data often stops when it could threaten trade or other interests. There’s no comprehensive database of outbreaks of infectious diseases, says a report in Nature (16 March 2006): “We have better data on galaxies 10 billion light years away than on human cases of avian flu in China or Vietnam. Yet the world is imperilled by outbreaks, wherever they happen.” Problems abound at every stage. Gathering data is difficult, as the only notifiable diseases that countries must report to the World Health Organization (WHO) at present are plague, cholera, and yellow fever1. Available data are skimpy, with scanty clinical data or information on familial case clusters. There are too few biological samples and genetic data. Results are often restricted by governments or kept private – in contrast to the Human Genome Project, where all the data were made public 24 hours after sequencing. Access is also

42 Quest 2(3) 2006

a problem: data are strewn across hundreds of individual WHO web sites, and the WHO's Global Health Atlas (which gives rough aggregate data for many diseases) has no H5N1 category under ‘influenza’. Hope comes, however, from the actions of Ilaria Capua of the Tri-Veneto Region Experimental Animal Health Care Institute in Italy who, in February, uploaded her most recent data on Nigeria and Italy to GenBank instead of saving them on private networks, and called on colleagues worldwide to do the same. Only by sharing freely can the global pattern of the evolution of bird flu be established and tracked. “The world badly needs a database for outbreaks of avian flu”, concludes the commentary in Nature. Will the world’s international agencies get together to oblige? 1. The WHO’s powers will broaden in 2007, when new internationally agreed health regulations come into force.

Q Books

Heritage of music-making

The History of Time: A Very Short Introduction. By Leofranc Holford-Strevens. (Oxford University Press, 2005). ISBN 0 19 280499 5

The Drumcafé’s Traditional Music of South Africa.

This informative, beautifully written little book tells the story of the ways in which human beings have defined time. Chapter by readable chapter, it explains the history and the workings of the day, months, and years, prehistory and the emergence of the modern calendar, calculating Easter, dividing the year into weeks and seasons, and marking eras in different cultures. There’s astronomy and mathematics, as well as history, mythology, language, and culture. This overview of time doubles up as a story of human civilization. It’s a book to dip into, to use as a reference, and, best of all, to read for pleasure.

By Laurie Levine (Johannesburg: Jacana, 2005). ISBN 1 77009 046 0


or the first time, a single volume offers the lay reader much of what’s known about the practices and traditions of authentic South African music among the country’s ten largest black language groups and cultures. A wealth of literature exists on the subject, but there was no easily accessible overview. This book fills the gap. It’s a beautifully illustrated and readable introduction to the field, brought to life by the attached 54-track CD, which, Laurie Levine hopes, will help people to “discover sounds that they never knew belonged to South Africa” and inspire our musicians “to revive the playing of South Africa’s rare traditional instruments”. The two-year project came out of the Drumcafé’s efforts to preserve and protect the country’s heritage. As Levine explains, the company’s early work involved using “interactive drumming as a teambuilding tool in the corporate environment” and it has now “developed one of the first-ever traditional South African orchestras, using only traditional instruments”. Deciding on the scope was not easy. The author “had to tease out the concept of ‘traditional’, [which] in this book encompasses the music that was heard when foreigners first

New books

Getting real about climate change The politics within the climate change debate often obscures the scientific facts, theories, predictions, and realities. Three new books tell us all about it.

arrived in South Africa, and overlaps slightly with the music that was developed [with] the migrant labour system.” The content covers “the traditional instruments of South Africa, some of the most distinctive dances, as well as some of the most distinctive musical features of the cultures”, with a strong focus on the rituals and ceremonies that incorporate music and dance. A musician herself, Levine found the research incredibly stimulating, particularly in “coming to understand the importance of music in building and sustaining a community.” It was also demanding, since, as she says, “given the importance of the subject matter, I couldn’t afford to make mistakes.” With Andrew Tracey (of the International Library of African Music at Rhodes University) as editor and guide, however, the challenges were met. The musical instruments and the people who play them are present in wonderfully evocative archival and modern photographs. Traditionally hand-made with natural materials, each instrument is unique and sizes are not standard. Drums, for instance – still popular, and often regarded as the most representative African instrument – used to be made most often from wood or clay with an animal-hide membrane stretched across the top, but those now played and enjoyed are also made from tin containers. This volume, with all its facts, stories, and celebrations of a host of vibrant practices, is a pleasure to read, to listen to, and to look at. ■

The Science and Politics of Global Climate Change: A Guide to the Debate. By Andrew Dessler and Edward Parson (Cambridge University Press, 2006). ISBN 0 52153941 2 This book examines the evidence for global climate change, as well as reasons why the world’s governments are having such difficulty in agreeing on a plan of action. In the words of Adrian Barnett, reviewing the book in New Scientist (25 February 2006), “Free copies should be shipped to anyone who doubts the reality of climate change, starting with presidents in denial.” The Winds of Change: Climate, Weather, and the Destruction of Civilizations. By Eugene Linden (Simon & Schuster, 2006). ISBN 0 684 86352 9 The journalist author considers climate change as a factor in the history of the fall of civilizations, and the palaeoclimate record as a guide to the speed at which profound changes can occur – measured in decades, in some cases, rather than millennia. Unprecedented human-induced greenhouse-gas emissions add a new dimension. Are we better prepared for rapid change than our ancestors were? The answer Linden offers here is “Maybe”. The Weather Makers: The History and Future Impact of Climate Change By Tim Flannery (Allen Lane, 2006). ISBN 07 13999 217 As a scientist, writer, and explorer, Flannery writes particularly well about the impacts of climate change on biodiversity. The problem, as he sees it, is huge; the need for action is urgent; the options for mitigation that should be considered include renewable energy, greater energy efficiency, and – controversially, perhaps – nuclear power. CORRECTION The first volume of the autobiography of Phillip Tobias, featured in QUEST 2(1), was not, in fact, published under the title Reminiscences of a Bone Picker. The details are: Phillip Tobias, Into the Past: A Memoir (Picador Africa and Wits University Press, 2005). ISBN 0 52153941 2

Quest 2(3) 2006 43

Q Measuring up

Q Q uest crossword You’ll find most of the answers in our pages, so it helps to read the magazine before doing the puzzle. 6 8











Divisions that unite us points out the ancient origins of the following basic units: Binary – This Hindu unit divides things into halves, quarters, and eighths. Modern computer programs are based on binary code. Decimal – The Chinese and the Egyptians were the first to use decimals, which are tenths. Duodecimal – The Romans used units of 12. Sexagesimal – The Babylonians used units of 60 (the basis of our seconds and minutes).

How did it pan out? 14 14


15 15




18 20







In September 2005, South Africa hosted and won the World Gold Panning Championships at Pilgrim’s Rest, Mpumalanga. The Netherlands came third. What do the Dutch know about gold? Well, they do know about managing water. A miner’s inch measures not length but the rate of water flow in a miner’s sluice, used for washing metallic ores from gravel. The rate is important as you don’t want to wash everything away. A miner’s inch is a flow of 1.5 cubic feet per minute or about 64 000 litres in 24 hours. The championships in 2006 will be held in Finland – who will Finnish first?

Sacrificed for science 26




30 Jemima

Across 1

A substance that is a mixture wholly or mainly of metals (5) 3 Watery fluid from animal tissue (5) 6 To search for underground water or minerals by holding a Y-shaped stick or rod (5) 8 See 16 Down 10 The sensor on NASA's Terra and Aqua satellites (5) 11 The study of the composition, structure, and history of the earth, and the processes occurring within it (7) 12 See 25 Down 14 Generic name of the world’s largest hornbill, --leadbeateri (8) 17 Member of the family Bucerotidae (8) 18 Geometric feature found at Vredefort (4) 22/15 The smallest hornbill (6,7) 23 The colour of most calcium minerals (5) 24 In computer terms, units of information expressed as a choice between two possibilities (4) 26 A small constellation, and the first sign of the zodiac (5) 27 Material that converts incident sunlight into an electric current more efficiently than silicon (4) 28 Respiratory organ of a fish (4) 29 The discovery site of a juvenile apeman (5) 30 Hot spring and electrical apparatus for heating water (6)

Down 1

A process or set of rules for calculation, especially by computer (9) 2 A test of metal or ore to determine its ingredients and quality (5) 4 A radioactive gaseous element formed by the disintegration of radium; it is the heaviest of the noble gases (5) 5 A famous australopithecine skull (3,4) 7 Betelgeuse and Rigel belong in here (5) 9 The branch of mathematics that deals with formulae and equations in which symbols stand for unknown numbers or other entities (7) 13 One who assembles and adjusts machinery (6) 14/20 A star that has cooled and become invisible (5,5) 15 See 22 Across 16/8 The expanded product of helium burning (3,5) 19 Centre for national archive of rock art (7) 20 See 14 21 Current flowing in one direction (6) 25/12 It turns sunlight into electricity (5,5)

How do you like the crossword puzzle? Was this one too difficult? Too easy? Just right? Would you like a more difficult puzzle as well (with a prize)? Fax The Editor at (011) 673 3683 or e-mail your comments to (Mark your message CROSSWORD COMMENT.)

In the UK, legislation governing the way doctors and scientists can use animals has been growing stricter since it was introduced in 1822. The BBC says the latest figures available show that 2.8 million procedures using animals were recorded in 2003, which is about half of the number carried out in the early 1970s. Most animals (85%) used in tests were rodents bred for that purpose. Birds, fish, amphibians, and reptiles accounted for 11%. Property damage by animal-rights activists in the UK amounted to an estimated £150 million in the past 25 years.

Smashing news Tennis player Andy Roddick holds the record for the fastest serve, at 246 km/h. The fastest serve by a woman player is Venus Williams’s 205 km/h. A new ball has been approved to slow tennis down and make it more watchable. It’s 6% larger in diameter and is slowed by drag force as it travels. A University of Florida study finds that the risk of injury to players is no greater as a result.

Seal in a Speedo The research journal Nature reports that British swimmer Lewis Pugh has broken two world records: “for the most southerly swim ever undertaken in the ocean, and the longest-duration polar swim ever completed. In completing the two feats, he battled temperatures that would kill you or me in minutes … On 14 December, Pugh swam a kilometre in the seas off the Antarctic Peninsula at a latitude of 65º South.… Two days later, he swam a mile off the nearby Deception Island, spending 30 minutes and 30 seconds in the water.… Physiologist Tim Noakes of the University of Cape Town, who accompanied Pugh on the trip, said: "I did not believe it possible to swim for 30 and a half minutes in 2–3 ºC water in just Speedo trunks." Pugh trained his body to tolerate the deadly cold by taking lots of cold showers and chilly swims. Eventually, just the sight of icy water sent his temperature up above normal, giving him a head start. Pugh also holds the record for the fastest swim around Robben Island, which he did in 2004 in three hours and 42 minutes.

Your first exam ever Did you know that the Apgar test of reflexes in newborn babies, invented by anaesthetist Virginia Apgar in 1952, also stands for Activity, Pulse, Grimace, Appearance, and Respiration? In each category, the baby can score 0, 1, or 2 points. Full marks for waving your limbs about, having a pulse over 100/minute, sneezing, being a good ruddy colour, and yelling.

Let this sink in ■ A darcy is a unit of permeability of rock to various substances. It’s named after a French hydraulic engineer, but tell that to anyone who has read Jane Austen’s Pride and Prejudice and encountered the stony-hearted hero who is won over in the end by the love of a good woman. ■ For Scrabble players, a useful word is yrneh, a unit of reciprocal inductance. (The spelling is the reverse of the henry [H], the unit of inductance named after the American physicist Joseph Henry!)

People numbers According to the US Census Bureau, Earth has crossed another population milestone ■ 6.5 billion is the world’s population, as of 25 February 2006 ■ 131 million people will be born in 2006 and 57 million people will die. (Reported in Nature, 2 March 2006)

Quest 2(3) 2006 45

Letters Q

Letters to Facts for Africa


uestions are often posed by foreign tourists concerning South Africa’s elephant population management programme. As a tour guide, I could never answer satisfactorily, as the information available in the popular media is sketchy and often emotionally motivated. Therefore it was a great enlightenment to read your excellent article “Elephantine dilemmas” (Quest vol.1, no.4) dealing with this highly controversial matter. The presentation of facts in an understandable way by scientists in Quest have enabled me, as a lay person, to answer these and other pressing questions about contemporary South Africa in a more comprehensive and less vague manner than before. Information provided by Quest in “Crackdown on Invasive Aliens” in the launch issue and in “Managing fires: the science behind the smoke” (vol.1, no.3), for instance, contributed further to my ability to perform the task of providing concrete and accurate information to our visitors. Since I discovered your magazine, I never miss the opportunity of passing my folder of Quest articles around to tourists young and old. It is a relief to be able to

replace the anecdotal information in many of our newspapers with valuable facts that we can trust. Karoly Pinter, Westdene, Johannesburg


was most delighted to read the SALT issue of your magazine. The government of South Africa must certainly be congratulated, especially the scientists involved in the telescope project and, most particularly, the astronomers. Certainly it was news to me that both Alpha Centauri and Proxima Centauri were discovered in South Africa (Quest, vol.2, no.2)! Astronomy is my hobby and I delight in it. The magazine is designed, as far as I could discern, for a wider audience and to be reader friendly – and complicated mathematical formulae are avoided. What guidelines do you have to simplify science to the marvellous level that is found in Quest? May God bless you through science. Bernard Zingano (Ph.D.), Lilongwe, Malawi.

Our “Call for Contributions and Guidelines for Authors” appear on the web site, – Editor

Notes on good writing “Good writing is writing that is perfectly controlled, the writer says just what he [or she] means, says it with complete clarity and simplicity, using the smallest possible number of words. I do not mean skimping paper …. But, granting that two sentences are at times easier to understand than one sentence containing the double meaning, the author tries to communicate with the reader with the greatest possible dispatch, save where for any one of forty reasons he [or she] does not wish to do so.” Ezra Pound, “The Serious Artist” (1913) “Write with nouns and verbs, not with adjectives and adverbs. The adjective hasn’t been built that can pull a weak or inaccurate noun out of a tight place.” “It is seldom advisable to tell all.” William Strunk and E.B. White, The Elements of Style. The best letter from a reader published in the next issue will win a Shaeffer pen. Address your letters to The Editor and fax them to (011) 673 3683 or e-mail them to (Please keep letters as short as possible. We reserve the right to edit for length and clarity.)

Q ASSAf News Awards and recognition ASSAf/TWAS Young Scientist Award 2005 Professor Vivian Alberts of the University of Johannesburg received this Academy of Science of South Africa (ASSAf) and Academy of Science of the Developing World (TWAS) award from the Minister of Science and Technology, Mr Mosibudi Mangena, in recognition of research findings in the field of novel thin-film photovoltaic (PV) cells (see p. 10). This is work of enormous potential economic and socio-economic value in the area of renewable energy. Alberts’s seminal conceptual breakthrough in the technology for manufacturing thin film CIGS (copper-indium-gallium-diselenide) PV cells makes it possible to manufacture high-quality, durable, and effective solar cells and panels at 20% of the current price of imported panels in South Africa. This is because his process now makes it possible to make commercially affordable solar panels on an industrial scale. Projected production costs are low enough to undercut the cost of coal-fired electricity generation. The invention is based on Alberts’s deep and subtle understanding of the physics of materials, his gift for innovation, and his exceptional laboratory skills. His work promises to generate a leading global industry with its intellectual centre in South Africa. The low cost of producing solar panels worldwide using his process gives the country an exceptional opportunity to supply electricity to the approximately two million national households that are not connected to the national grid, thereby

boosting development in poor communities, specifically in terms of education and health services and industrial development.

The first Sydney Brenner Fellowship Offered jointly by the ASSAf and the US National Academies, this award was initiated by Sidney Brenner’s donation of part of his 2002 Nobel Prize to help to establish a high-level two-year research fellowship to be held in a South African institution (see Quest, vol. 1, no. 4, p. 47). The ASSAf wishes to expand the Sydney Brenner Fellowship scheme through fundraising in South Africa and in the countries where Brenner has worked so productively. Once funds are available, the intention is to award three fellowships each year and to develop them into South Africa’s premier research awards in the molecular life, agricultural, and health sciences. The first Fellow is virologist and bioinformaticist Dr Darren Martin of the Institute of Infectious Diseases and Molecular Medicine at the University of Cape Town. He will be mentored by Brenner from his bases in Japan, Singapore, and the US. Bioinformatics is the computer analysis of the complete genetic sequences of living organisms to deduce how the genes coded within them work, individually or together. This kind of research is used to support life and to explain disease. Martin is particularly interested in the effects of genetic recombination (that is, the crossing over between different chromosomes during the reproduction of organisms). In this process, the genetic material of two individual viruses mixes to produce so-called ‘recombinant offspring’. When disease-causing

virus reproduction involves recombination, such viruses can evolve to greater virulence, multiple-drug resistance, and vaccine escape. Martin will examine a large number of virus genome sequences that have been deposited in public sequence databases and identify ‘recombination hot- and cold-spots’, that is, parts of the sequences where, respectively, many or few recombination events can be detected. Such research is important in attempting to understand HIV, for instance, which undergoes many mutations after infecting a person’s body, a substantial number of which result from recombination events (see also Martin’s article in Quest, vol. 1, no. 4, p. 11). Martin’s work attempts to improve our understanding of the rules of recombination.

Giving and taking advice How is advice requiring the weighing of evidence – that is, science-based advice – best generated and implemented in a developing country such as South Africa? This question was debated at the ASSAf’s ‘double symposium’ on 3 March (see p. 33). The Minister of Science and Technology commented on governments’ need to draw on the knowledge and skills of the science community, and the Deputy Director-General, Dr Adi Paterson, offered advice on how advice should be offered! Examples were presented from the USA and the UK of studies on obesity and nanotechnology, respectively, and South African leaders outlined advisory processes at work in our country. The Proceedings will be available in due course

Q Diary of events National Science Week (13–20 May)

Astronomy events

Research funding

The Department of Science and Technology invites South Africans to celebrate National Science Week – to become more aware of the importance of science in daily life, to encourage young people to study mathematics and science, and to plan careers in science, engineering, and technology. The 2006 theme is “Tomorrow’s science and technology are in our youth’s hands”. Watch for events in your area! Visit

■ ScopeX (16 April) – International Astronomy Day. Celebrate by visiting the ScopeX Telescope and Astronomy Expo (09:00–21:00) at the South African National Museum of Military History (behind the Johannesburg Zoo), hosted by the Johannesburg Centre of the Astronomical Society of Southern Africa (ASSA). View telescope and camera exhibits; find out about space science; attend presentations in the auditorium; watch science shows and demonstrations; enter competitions; and enjoy the Star Party in the evening. Visit or contact Lerika Cross at (contact Lerika to book for Mike Melvill’s keynote lecture, “One Man’s Journey into Space”).

Apply by 16 May 2006 for research funding from the European Union’s Sixth Framework Programme (FP6). For more on the list of projects and the European South African Science and Technology Advancement Programme (ESASTAP) visit

International Museums Day (18 May) The Transvaal Museum (Paul Kruger Street [opposite the city hall], Pretoria) will exhibit the very first fossil hominid – a fragmented skull and cast of the brain – that was discovered at Sterkfontein in 1936. Robert Broom called it ‘Plesianthropus’ (meaning “almost human”) but it was later described as Australopithecus africanus. For more phone (012) 322 7632, fax (012) 322 7939, e-mail, and, for general information visit (under TM, Transvaal Museum).

Name the Satellite Challenge Learners in Grades 7–12 are invited to name South Africa’s micro-satellite. Send your entries by 26 April 2006. You could win (1) a computer, (2) a visit to see the construction of the satellite. Details at www/saasta/ac/za/namethesatellite/ index.shtml

■ Iziko Planetarium – In Cape Town, book for Tony Fairall’s Starfinder Astronomy course (3–24 May, Wednesday nights 20:00–22:00) and visit the planetarium show, “Basic Guide to Stargazing”, starting 1 April. For bookings and details phone (021) 481 3900. ■ Views from the Cradle-Ribeiro Observatory – visit

Arts and crafts ■ Go to the Origins Centre at the University of the Witwatersrand to view the GoetheInstitut’s “Memory and Magic” exhibition of artworks created by contemporary !Xun and Khwe artists. Closes 23 April. For more, e-mail holl@johannesburg. or visit ■ Check out the Zasekhaya Market from 10:00–16:00 on the last Saturday of every month at the Bus Factory craft and design centre at 3 President Street, Newtown, Johannesburg for hand-crafted gifts, artwork, activities, and live entertainment. Secure parking. For details contact Kate or Macky at (011) 834 8760 or e-mail

Plan for ■ Antarctica month – June ■ Marine biosciences month – August ■ African Origins month – September ■ Astronomy month – October

Quest 2(3) 2006 47

Back page science Q So much for useful machines ■ “Want to make your computer go really fast? Throw it out of a window.” Anon. ■ “If the automobile had followed the same development cycle as the computer, a RollsRoyce would today cost $100, get one million miles to the gallon, and explode once a year, killing everyone inside.” Writer and broadcaster Robert X Cringely. ■ “In my childhood we were always assured that the brain was a telephone switchboard. (What else could it be?) Sherrington, the great British neuroscientist, thought that it worked like a telegraph system. Freud often compared it to hydraulic and electromagnetic systems. Leibniz compared it to a mill, and now, obviously, the metaphor is the digital computer.” John R. Searle, philosophy professor (1932– ).

Ideas about ideas ■ “The best way to have a good idea is to have lots of ideas.” Nobel laureate Linus Pauling (1901– ). ■ “One's mind, once stretched by a new idea, never regains its original dimensions.” US author and physician Oliver Wendell Holmes (1809–1894). ■ “Don't worry about people stealing an idea. If it's original, you will have to ram it down their throats.” Computer engineer Howard Aiken (1900–1973). ■ “Discovery is seeing what everyone else has seen and thinking what no one else has thought.” Nobel laureate Albert SzentGyorgyi (1893–1986).

Glamour at last The winner of the UK’s 2005 FameLab competition to find an Idols-type ‘face of science’ is acoustics expert Dr Mark Lewney, whose presentation was on the physics of music. His day job is in the Patents Office, dealing with new inventions

in telecommunications. A runner-up was Dr Maggie Aderin-Pocock, who custom-makes instrumentation, including landmine detectors and satellite subsystems that help predict weather. … but not for all “Pain, tedium, danger, disgust, humiliation – it's all just part of the average workday for the (often proud, more often smelly) members of our third annual honour roll of the Worst Jobs in Science,” says Popular Science. Among the 2005 winners are collectors of orang-utan urine, volcanologists, researchers of microbes that live in arsenicsaturated mud, and human guinea pigs testing safe levels of pesticides.

A job for Clark? The gravity on Superman’s home planet, Krypton, is at least 10 times as strong as Earth’s. That’s why you have to be extra-strong to get anything done on Krypton, and when you visit Earth you find it quite easy to lift up a car, say. For more scientific explanations of the special powers of comic-book characters, visit science/hottopics/superheroes/index.shtml

Grand Prix in the sky Imagine a fleet of rocket planes alternately screaming and gliding 1 500 m above ground, each generating a tail of flame. This is the Rocket Racing League, brainchild of Dr Peter Diamandis, chairman of the X Prize Foundation, a non-profit education organization promoting the formation of a space-tourism industry. He envisages the finals of the league taking place in New Mexico and hopes it “will inspire people of all ages to look once again up into the sky to find inspiration and excitement.” X-Racer planes will be built for the series and it’s hoped that the event will encourage developments in design.

‘bits of hyperbolic space’, for example. These look like the scrunchies that girls use to tie back their hair. One such artist is San Franciscan Eleanor Kent, who knits fractal designs and mathematical formulae. And English mathematician Claire Irving has made the real projective plane and the Klein Bottle (a one-sided closed surface with one hole and one handle) from wool. Some of us end up with this weird stuff when we’re actually trying to make a jersey… … and hear To compose music based on sequences of numbers, visit for the number-crunching and notation and listen to the result. You can start with an existing algorithm such as a Fibonacci sequence, or put in your own choice of numbers. Head-banging stuff!

The science of yuck What disgusts us and why? The London School of Hygiene and Tropical Medicine suggests that disgust evolved to protect us from disease. The more something looks like it could make us sick, the more revolted we are. You can take part in an online experiment, rating your response to various pictures. How do you feel about touching blue slime? What about yellow slime? Would you rather touch a wasp or a worm? Try it on surveys/disgust/ Answers to Crossword (page 45) ACROSS: 1 Alloy, 3 Serum, 6 Dowse, 10 MODIS, 11 Geology, 14 Bucorvus, 17 Hornbill, 18 Dome, 22/15 Tockus camurus, 23 White, 24 Bits, 26 Aries, 27 CIGS, 28 Gill, 29 Taung, 30 Geyser. DOWN: 1 Algorithm, 2 Assay, 4 Radon, 5 Mrs Ples, 7 Orion, 9 Algebra, 13 Fitter, 14/20 Black dwarf, 16/8 Red giant, 19 Origins, 21 Direct, 25/12 Solar panel.

Maths you can see Call it woolly thinking if you like, but some people are devoted to mathematical fibre art – knitting

Subscription form I would like to subscribe to 4 issues of Quest: Science for South Africa. My details are (please print): Title:



Surname: Student number (to qualify for student rates):

Postal address: Work tel: (


Fax: (



Home tel: (

Code: )


Subscription rates (4 issues incl. postage) Please tick appropriate rate. South Africa (incl. VAT) ❑ Individuals/institutions – R 80.00 Neighbouring countries (Southern Africa) ❑ Individuals/institutions – R160.00 Foreign ❑ Individuals/institutions – R180.00

Payment options

❑ ❑

❑ ❑ ❑

Students/Schoolgoers – R 40.00 Students/Schoolgoers – R120.00 Students/Schoolgoers – R140.00

Please select and tick your payment option below.

CHEQUE: Enclose your cheque in South African rands made payable to QUEST MAGAZINE (together with this completed Subscription Form) DIRECT DEPOSIT: Use reference QUEST SUB on your deposit slip and deposit your subscription as follows:

Bankers: Standard Bank Northcliff Account Number: 200 5222 05 Branch code: 006305 Account name: Quest magazine POST the completed Subscription Form together with your cheque or a legible copy of your cheque/deposit slip to: Quest Magazine, PO Box 130614, Bryanston 2074, South Africa OR FAX the Subscription Form together with deposit slip to: QUEST SUBSCRIPTIONS (011) 673 3683. Subscription enquiries: (011) 781 8388/083 408 3286 or e-mail:

48 Quest 2(3) 2006

Quest 2(3)  
Read more
Read more
Similar to
Popular now
Just for you