Atoms Under the Floorboards: Notes and References

Last updated: 11 March 2018.

Atoms Under the Floorboards cover - large

This is a slightly expanded, handily web-linked version of the notes and references in the book. Since Atoms was always intended to be a popular science book, not a scholarly work, I've not referenced everything. Even so, you'll find sources for most of the factual information here.

123 Notes shown like this refer to the superscripts you'll find in the printed text.

p.123 Since the book went to press in January 2015, I've expanded a few of the notes and added a few new ones. The new ones are listed by page number, inserted in this list at the appropriate place between the existing notes.

p.123 No matter how hard you try, some mistakes always creep into books. My sincere apologies for any errors that you find and any confusion they cause. The ones I'm aware of are listed here and marked in blue. They're indicated by page number, inserted in the appropriate place in the list of notes (as above). Most of these should have been fixed in the 2016 paperback edition.

Where web pages are marked with a fixed publication date, that's what I've quoted; otherwise I've quoted the original date of access. Where possible, I've linked book names directly to Google Books.

Since I first compiled these notes, a few web pages have moved URLs or become unavailable. Wherever possible, I've attempted to source redirected or archived versions (usually on the Wayback Machine), though I've not bothered to change the access date. Where I haven't been able to find alternative URLs, I've deleted the dead links to avoid your seeing a 404 "Not found" error page when you click through. All the links on this page were working as of January 2018.



1. I've gleaned details about Einstein from the eminently readable biography Isaacson, W. 2007. Einstein: His Life and Universe. Simon & Schuster, New York. Dropping out of high school is covered on p.23, his polytechnic entrance failure on p.25 and his job-seeking struggles on pp.58–65.

2. Public opinion statistics quoted here are from both the United States and the UK: NSF: Science and Engineering Indicators 2012: Chapter 7. Science and Technology: Public Attitudes and Understanding, accessed 17 October 2014; Science Dull and Hard, Pupils Say. BBC News, 16 June 2005; Children Losing Interest in Science Through their Education, Report Claims. Daily Telegraph, 7 September 2008; and Ipsos MORI. 2011. Public Attitudes to Science (PAS) 2011, 2 May 2011. Since the book went to press, the Pew Research Center has published A Look at What the Public Knows and Does Not Know About Science (10 September 2015), confirming that scientific knowledge is somewhat patchy. While most of us (86 percent) know that the Earth's core is its hottest feature, far fewer (35 percent) know that amplitude is the property of a sound wave related to loudness. There are also some worrying ethnic differences.

Chapter 1: Firm Foundations

p.11. "Stone pyramids that have stood... for about 5000 years" is a reference to the pyramids of Brazil dating from around 3000BC. See World's oldest pyramids are discovered, The Independent, 19 November 1996.

p.12. Oldest buildings in the world. See Oldest Buildings in the World: By Age, which suggests the earliest still-surviving structures are about 6800 years old—about 75 times older than a 90-year-old person (but I've rounded up).

p.13. Typical adult male = 75kg (165lb). I've assumed this weight quite often through the book, working on the assumption that many of you readers will be younger and lighter than the average. I checked my figures on a chart reproduced by Dr Steven Halls.

p.14. Line 8: "tire" should, of course, be "tyre".

3. The Official Site of the Empire State Building, accessed 29 October 2013. 340,000 tonnes = 340 million kg, which is equivalent to about 4.5 million people each weighing 75kg (165lb). In 2011, Calcutta's population was roughly 4.5 million people, according to Wikipedia.

4. More precisely, it's 412,500 pascals (Pa) or 60 pounds per square inch (psi).

5. It's roughly 4 million pascals (Pa) or 600 pounds per square inch (psi).

6. Hayes, L. et al. The Latest from Paris: An All-Plastic House. Popular Mechanics, August 1956, p. 89.

7. Hay, D. What Does a House Weigh? Some Mental Heavy Lifting. Seattle Times, 19 December 2004.

8. If that makes you wonder why anything moves at all, the answer is because the action and the reaction affect different things. If you fire a gun, the action is what pings the bullet through the sky; the reaction is the recoil that thumps the gun in the opposite direction. The action moves the bullet, the reaction moves the gun.

9. For every common building material, there's a measurement called Young's modulus (E) – roughly an indication of how stiff or elastic it is. Young's modulus is calculated by dividing the stress on a piece of material (how much force you apply per unit area) by the strain this produces (how much the material stretches compared to its original length). I've assumed a fairly high strength concrete that occupies about a tenth the floor space of my office block. I've also assumed, in effect, that everyone inside the building is standing on the roof compressing the entire structure equally throughout its height. In reality, the bottom of the building is compressed more than the top.

10. How Tall Can a Lego Tower Get?, BBC News, 4 December 2012.

11. Kunreuther, H. et al. Overcoming Decision Biases to Reduce Losses from Natural Catastrophes. In Shafir, E (ed). 2013. The Behavioral Foundations of Public Policy. Princeton University Press, Princeton, NJ, p.405.

p.23. Sources for my table of forces:

12. For a good discussion, see Cool Formula for Calculating Skyscraper Sway. Maths Pig Blog, 21 March 2011. Also Building stiffness and flexibility: earthquake engineering. Architect Javed, 16 October 2011.

13. Levy, M. and Salvadori, M. 2002. Why Buildings Fall Down: How Structures Fail. W. W. Norton, New York, p.204. Teachers interested in exploring the science of buildings should visit the The Salvadori Centre, founded by one of the book's authors (Columbia University Professor Mario Salvadori) in 1976.

14. Taipei 101: Wind Damper. Taipei 101, accessed 14 March 2016.

Chapter 2: Upstairs, Downstairs

15. A red-hot slap across the face turns kinetic energy (a moving hand) into light, heat and sound energy; pigeon 'applause' sees stored chemical energy (food) turned rapidly into kinetic energy (flight); fizzing pills convert chemical energy into sound (and possibly heat); winking smoke alarms turn electrical energy from their batteries into light; trapped flies are waiting to be converted into chemical energy (digested food), while taut spider webs are examples of elastic potential (stored) energy.

16. Calculation: Potential energy = mgh = 30kg × 10m/s/s × 400m = 120,000kJ. Bob uses 260kJ and Andy uses 380kJ. This is purely a calculation based on shifting a mass upwards by a certain distance. It doesn't take into account the body's inefficiency or any other forms of energy use on the way, such as scuffing your shoes on the floor (frictional losses), waving to people on the ground (mechanical energy), whistling a happy tune (losing energy as sound) or anything else.

p.28. Two Maryland cookies contain 108 Calories according to Maryland Cookies Double Chocolate 230g, Ocado, accessed 14 March 2016.

17. If Andy weighs 95kg, his climb to the top requires at least 95kg × 10m/s/s × 400m = 380,000kJ, which is equal to 90 Calories. As we'll discover in Chapter 14, things are not quite this simple: any cookie Andy eats will not be converted 100 per cent into potential energy. Dr Anthony Viera from Chapel Hill School of Medicine, University of North Carolina, has proposed food labels that show not just Calories but how many minutes worth of walking it would take you to burn them off. See Fast-food Consumers May Eat Less if Label Describes How Long it Takes to Walk off Calories. UNC Gillings School of Global Public Health, 21 January 2013.

18. Calculation: Energy = (mass) × (specific heat capacity of water) × (temperature change). For 1 litre (1kg) of water, heated from 10°C to boiling point (100°C), that gives 1kg × 4200J/kg/°C × 90°C = 378,000 joules. Assuming a dynamo is about 5 watts (5 joules per second), that gives 75,600 seconds = 21 hours. Bicycle dynamo figure from various models on (Dynosys models are about 5 watts; Busch & Müller are about 6 watts). However, if you ditch the puny dynamo and harness the cyclist's pedalling power much more directly, using a decent electricity generator, you could theoretically capture as much as 400 watts and boil your water 80 times faster – because that's the maximum sort of power a racing cyclist is capable of delivering at top speed. That figure comes from the excellent Wilson, D. 2004. Bicycling Science. MIT Press, Cambridge MA and is consistent with the 2014 Tour de France power analysis by Training Peaks, which suggests top riders average about 300–400 watts.

p.32. Sources for my table of energy values. Some are simple calculations, the others come from:

  • Lightning bolt: See How do Thunderstorms Work? by Chris Smith, The Naked Scientists, 2 June 2007.
  • AA alkaline battery: If it's rated at 2000mAh, and it's 1.5 volts, we can calculate that joules = watts × time = voltage × current × time = 1.5V x 2.5A x 3600 seconds = 13,500 joules. If it's rated at 1500mAh, we'd get 8100 joules. I've taken an average and plumped for 10,000 joules.
  • A roof area of about 28 square meters is good for a 4kW panel, according to 4kW Solar Panel Systems by Alex Vasili, The Eco Experts, 6 November 2014.

19. You can read James Joule's own account of his energy experiments in Shamos, M. H. (ed.) 1987. Great Experiments in Physics: Firsthand Accounts from Galileo to Einstein. Dover, New York, pp.166–183.

20. Once again, I am thinking only about potential energy and ignoring everything else.

21. This calculation also measures only how much potential energy Andy needs to lift his body up that distance, taking no account of mechanical efficiency (which will multiply the amount accordingly): 95kg × 10m/s/s × 400m =380,000 joules/1800 seconds = about 210 watts.

22. A fascinating green-energy buff called Piet demonstrated this to me about 20 years ago in his self-sufficient 'mobile office'. His solar powered laptop was wired to an inkjet printer that took its power entirely from a hand crank bolted to his desk. To print each sheet of paper, you had to grind the crank around quite a few times – and it felt about as difficult as turning over a car engine the old-fashioned way. I've never forgotten this, the most vivid demonstration of making energy I have ever had.

23. My estimated hamster power is about 0.5 watts and comes from a guesstimate supplied in Fink, D. "Science for Kids: Hamster Power",, 13 December 2005.

p.34. Sources for my table of power values:

  • Space Shuttle at takeoff: See Power of a Space Shuttle by Staverie Boundouris, The Physics Factbook, 2001.
  • Hoover Dam: 2080 megawatts according to Hoover Dam FAQ, US Bureau of Reclamation, February 2009.
  • Nuclear power: 1.5 gigawatts is reasonable for the biggest reactors. See Nuclear power stations and reactors operational around the world: listed and mapped by Simon Rogers, The Guardian, 18 March 2011.
  • Steam engine power: 1.5 megawatts is typical peak power of a large locomotive, according to Locomotive Power and Indicator Diagrams, The 5AT Group, accessed 5 February 2015.
  • Diesel truck engine: I used a Cummins ISX15, rated as 400–600 horsepower, giving a maximum of about 450 kilowatts.
  • Porsche Turbo: Approximately 520 horsepower or 390 kW, according to Porsche 911 Turbo Features and Specs, Porsche, accessed 5 February 2015.
  • Hand crank: I used a K-TOR Pocket Socket Hand Crank, rated as 10 watts.
  • Hamster: Based on "Science for Kids/Hamster Power?" by Dan Fink, All Experts, 13 December 2005. I have no idea whether this is correct, but it seems reasonable that a hamster would generate only a fraction as much power as a human hand.

24. These are simplified guesstimates that don't take into account energy losses.

25. The amount of energy you need to achieve something (climbing a hill, boiling a kettle or drilling a hole in wood) is always the same, but you generally need more than this amount because whatever you're doing is not 100 per cent efficient. Efficiency is the amount of energy it takes to do something in practice compared to the amount it ought to take in theory. If something takes twice as much energy as it should, it's only 50 per cent efficient. If you want to get your body to the top of a mountain, you'll need to use more energy if you go by car than by bike. That's because a car engine is much less efficient than a bike. Going by car will feel easier because the energy is coming not from your body but from the petrol you're burning. But if you had to pay per unit of energy you use, going by car would cost far more. That's partly because a car weighs about 15-20 times more than you do, so you're lifting a load of metal, rubber and glass up the mountain as well. It's also because car engines and transmissions don't convert all the energy locked in the petrol into kinetic energy that moves the vehicle along.

26. To boil water by stirring with a spoon, you'd need to ensure that all the heat you're adding isn't lost just as quickly to the cooler, surrounding air. Theoretically, you could do it if the spoon spun around in some kind of cunningly designed vacuum flask. That would be a modern spin (in more ways than one) on James Joule's original experiment, where he used a falling weight to spin a paddle wheel inside a closed container of water.

27. James Joule, Letter to the editors, Philosophical Magazine 27 (1845): 205, quoted in Shamos, M. H. (ed.) 1987.Great Experiments in Physics: Firsthand Accounts from Galileo to Einstein. Dover, New York, p.169.

28. Gjertsen, D. 'James Prescott Joule' in Wintle, J. 2013. New Makers of Modern Culture: Volume 1. Routledge, London, p.772.

29. Kinetic energy = ½mv2 where m = mass and v = velocity (or speed, in this case).

p.39. Table 3 is on page 34, not page 30.

30. Since a car's energy is equal to the square of its speed, doubling your speed means you have four times as much energy, while quadrupling your speed gives you sixteen times as much. That's why a car crash at twice the speed is much more than twice as dangerous.

31. Your potential energy = mgh = 75kg × 10m/s/s × 10m = 7500J.

32. Assuming you're travelling at 15m/s when you hit the ground, your momentum is 1125kgm/s and the force from an impact lasting between 0.1 seconds is 10 times this number = in the order of thousands of newtons, which is approaching the sort of force you'll get from an alligator bite (according to our table in Chapter 1).

Chapter 3: Superheroics

33. Suppose Archimedes balances Earth on a kind of super-sized golf tee, with the entire weight acting down through a pin-point. Let's say he sticks his tee on the very end of a lever balanced 1m (3ft or so) from the end. If he jumps down on the other side of the lever using his entire body weight (supposing it's about 75kg or 165lb), how long would the lever need to be to balance horizontally like a see-saw? If the weight of the world is 6,000,000,000,000,000,000,000,000kg, it's an easy sum. For a see-saw to balance, the weight times the distance has to be the same on each side of the pivot. So the length of the lever would be six trillion trillion divided by 75 or 80,000,000,000,000,000,000 (80 million trillion) kilometres (50 million trillion miles). Sounds fine in theory, but you see the problem? The distance from Earth to the Sun is a mere 150,000,000 kilometres (150 million kilometres or 93 million miles). Lost in space, what would he use for gravity? And what exactly would he stand on? The more you think about it, the more ridiculous it seems – which is probably which it's such a memorable thought experiment.

34. The energy you use is given by the formula: potential energy = mgh = 200kg × 10m/s/s × 1m = 2000J.

35. The exact temperature depends on many factors. Softwoods catch fire at lower temperatures than dense, heavy hardwoods. The lower figure of 200°C (400°F) is quoted in Cote, A. & Bugbee, P. 1988. Principles of Fire Protection, National Fire Protection Association, New York.

36. Generally this doesn't happen: sporting stars all understand the importance of the 'follow-through', in which you keep your leg, arm or another part of your body moving after contact to increase the amount of energy you pass on, reduce the force on your limbs and lessen the risk of injury.

37. At school, hydraulic mechanisms like this are explained by a scientific law called Pascal's Principle, dating from 1651, which effectively says that the pressure of the water is the same at all points in a pipe. So, if you divide the force of the fluid by the area of the pipe (to give the pressure), you'll get the same number even if the pipe widens or narrows; a small force on a small pipe translates into a big force on a big pipe, in the case of the hydraulic jack. I think it makes more sense to try to understand hydraulic machines in terms of energy. The energy the jet has when it leaves a water pistol can't be any greater than the energy you put in by squeezing the trigger. Since the water leaving the pistol is travelling faster than the trigger, it moves through a bigger distance each second. The energy something has is equal to the force acting on it multiplied by the distance it moves. If the energy remains constant (as it must), the force must get smaller if the speed increases.

38. Cross, R. 2011. Physics of Baseball & Softball. Springer, New York. Rod Cross has an excellent personal website covering the physics of ball sports.

Chapter 4: The Beauty of Bikes

39. Lance Armstrong & Oprah Winfrey: Cyclist Sorry for Doping. BBC News, 18 January 2013.

p.67 It should be obvious from this chapter that I know how gears work. Nevertheless, a mistake crept into the text here in the final round of editing. The gear ratio should, of course, be the other way round. If we divide the number of teeth on the back wheel by the number on the pedal wheel, we get a ratio of 1:5. Alternatively, if we divide it the other way, we get 5:1. Either way, in the example I've given, there are five times more teeth on the pedal wheel than on the back wheel.

40. For a fascinating insight into what top cyclists are capable of, see Padilla, S. et al. (2000). Scientific Approach to the 1-h Cycling World Record. Journal of Applied Physiology, 89, pp.1522–1527. For an even more vivid demonstration, watch British TT racer Guy Martin's astonishing (and successful) attempt to break the British land-speed cycling record in the slipstream of a truck! With an unusual arrangement of two chains and three gear wheels, Guy's bike produced a gear ratio of about 7.5:1, propelling him to an unbelievable 180 km/h (112mph). There's a short clip at Britain's Fastest Cyclist and the full programme is episode 1 of Speed with Guy Martin (you might or might not be able to watch this online depending on where in the world you are).

41. Cars typically weigh around 1500kg (3300lb), while bikes weigh about 15–20kg (33–44lb). I'm thinking of a small car like a VW Polo. A Raleigh Flyte mountain bike has a boxed weight of 18kg, while an Apollo Phaze comes in at just 15kg.

42. Alam, F. et al. 2009. Aerodynamics of Bicycle Helmets. In Estivalet, M. and Brisson, P. 2009. The Engineering of Sport. Springer, Paris, p.337.

43. Mack, J. 2007. Don't be a drag. Bicycling, August 2007, p.46.

44. For a variety of plausible suggestions, see Sumner, J. Why Do Cyclists Shave Their Legs?, Bicycling, accessed 9 October 2014. Also worth a look is Ward, T. Can shaving your legs offer any advantage for regular male cyclists?, The Guardian, 17 April 2013. The comment by Mike Burrows—that a little bit of hair can give an aerodynamic advantage—is counterintuitive, but particularly intriguing!

45. Wilson, D. 2004. Bicycling Science. MIT Press, Cambridge, MA. p.188.

Chapter 5: Car Crazy

46. Ward's Auto, the motor industry analysis centre, announced this in World Vehicle Population Tops 1 Billion Units. Ward's Auto, 15 August 2011.

47. The figure of six billion is based on the number of SIM cards, not the number of users or handsets. UN: Six Billion Mobile Phone Subscriptions in the World. BBC News, accessed 12 October 2012.

48. One billion sheep cited by Compassion in World Farming, Factsheet: Sheep [PDF format], accessed 7 December 2013.

49. At 3000 rpm, a 3.8-litre Porsche Turbo uses (3000 × 3.8)/2 = 5,700 litres per minute. If a cyclist breaths about 10 times per minute at 50 per cent of lung capacity (50 per cent of 5 litres), that suggests they need only 25 litres per minute. Similar figures are quoted by Hammond, R. 2008. Car Science. Dorling Kindersley, London, p.22.

50. American Motors Corporation offered a high-altitude package on its cars, which 'brought a special engine/gearbox/final drive combo, factory adjustments to carb fuel mixture, engine idle speed, and ignition timing and some changes to emission controls', according to Cranswick, M. 2011. The Cars of American Motors. McFarland Publishers. Jefferson, NC, p.187.

51. Denny, M. 2011. Gliding for Gold: The Physics of Winter Sports. Johns Hopkins University Press, Baltimore, MD. See 'Note 10: Getting High on Speed'.

52. A Ford Focus weighs in at about 1800–1900kg (4000–4200lb) according to Ford Focus: Specifications. Ford, accessed 10 January 2015.

53. United States Office of Technology Assessment, Congress. 1995. >Advanced Automotive Technology: Visions of a Super-efficient Family Car. Diane Publishing, Darby, PA, p. 203. There are numerous different strengths of steel used in a car and the trend is toward lighter and lighter materials. This article about the Audi A3: design story (see the section "Lightweight body design") gives you an idea of which steels are used where and how lighter materials are replacing heavy steels wherever possible.

p.78. One reader has asked me about an apparent contradiction between the pie chart and the caption above it. How can a car be 55 percent steel (in the chart) if "The steel body accounts for up to a third of the total weight" (in the caption). The answer is that the body isn't the only steel component: it makes up ~33 percent of the weight, but other steel parts (some of the engine, the transmission, gearbox, differential, and so on) make up the rest.

54. Fuel Economy: Where the Energy Goes. US Department of Energy Office of Energy Efficiency & Renewable Energy, accessed 7 December 2013. Surprisingly, very little progress has been made on making cars lighter since the early 20th century. According to Vaclav Smil, we still "get the worst ratio of vehicle-to-passenger weight since a mahout last rode a bull elephant to work", and even a Boeing 787 airplane does better than a small passenger car. See Smil, V. Cars Weigh Too Much. IEEE Spectrum, 19 December 2014.

55. Prof. Ferdinand Porsche Created the First Functional Hybrid Car. Porsche, 20 April 2011. If you like that, check out this great video of an old hybrid Porsche Semper Vivus on YouTube.

56. Electric car maker Tesla quotes about 40km (23 miles) of range per hour of charging, as a typical figure, although its superchargers promise a much more convenient 275km (170 miles) of range for just 30 minutes of charging. See Tesla Charging, Tesla, accessed 10 October 2014.

57. This chart was compiled by comparing data from three different sources: Energy Density. Wikipedia, accessed 8 December 2013; Hammond, R. 2008. Car Science. Dorling Kindersley, London, p.83.; MacKay, D. 2008. Sustainable Energy Without the Hot Air. UIT Cambridge, Cambridge, p.199, which gives figures in kWh that I've converted to MJ.

58. There's no exact figure for how hot a car's brakes get: it obviously depends on all kinds of factors, including the type of material they're made from, the mass of that material, how fast the car is going, how well the brakes are cooled, the ambient temperature, whether it's raining and so on. For a ballpark figure, I consulted Burt, W. 2001. Stock Car Race Shop. Motorbooks International, Osceola, WI, p.199, which says brakes 'begin to glow' at 650°C (1200°F). Stapleton, D. 2008. The MG Midget and Austin Healey Sprite High Performance Manual. Veloce Publishing, Dorchester, p.124, suggests the upper limit for typical car brakes is 700°C (1292°F). For extremes of speed, the official Formula 1® website quotes a figure of 750°C (1382°F). See Brakes, accessed 10 October 2014.

59. United States Office of Technology Assessment, Congress, Advanced Automotive Technology: Visions of a Super-efficient Family Car. Diane Publishing, Darby, PA, p.165 suggests a typical value is 8–10 per cent and 17–18 per cent is the maximum. See also Drivers on Track for Greener Trains. BBC News, 10 October 2009, which says 'Train companies can save around 15% of their energy use' through regenerative braking.

60.Motorbikes are just as impressive: the contact area between tyre and road is a 'postage-stamp sized patch of material, which started its life in a tree'. See Dunlop. 2012. Sportmax Range [PDF format], accessed 14 March 2016.

Trains are equally surprising, as James May explains in a funky little video called Why can't trains go uphill?: apparently the contact area for a train is about the size of two 50p coins. The difference is that trains don't have tyres—they run metal wheels on a metal track— which is why they struggle to climb hills. Maybe James should have asked (and answered) the question "Why don't trains have tyres?"

Chapter 6: Sticky Stuff

61. Vollmer, M. & Möllmann, K. 2013. Is There a Maximum Size of Water Drops in Nature?. Physics Teacher, 51, p.400.

62.The tiny, short-range electrostatic forces that happen between nearby molecules are called Van der Walls forces, after a Dutch physicist named Johannes van der Waals (1837–1923). See Johannes Diderik van der Waals: Biographical: The Nobel Prize in Physics 1910,, accessed 12 December 2013.

63.There are 6×1023 molecules in 18g (0.6oz) of water, so there are 3×1021 molecules in 0.1g (0.004oz), which is quite a large drop of water. Glue would have much bigger and heavier molecules than water. Even so, we're in the right ballpark to consider that a single drop of glue would contain many trillions of molecules.

64.One drop can support about 23 newtons (2.3kg or 5lb), according to Nature's Strongest Glue Could be Used as a Medical Adhesive. Science X Network, accessed 12 December 2013. The sticking power of Caulobacter crescentus is described in the same article.

65. Sticky Moments in 21 Years of Superglue. BBC News, 21 October 1998.

66.Silver, S. et al. 1975. US Patent 3,922,464: Removable Pressure-Sensitive Adhesive Sheet Material. 25 November 1975.

67.The particle size is 25–45 microns compared to a conventional adhesive, which is more like 0.1–2.0 microns, according to Karukstis, K. and Van Hecke, G. 2003. Chemistry Connections: The Chemical Basis of Everyday Phenomena. Academic Press, New York. p.214.

68. 'Such forces become negligible at distances equivalent to only about 4 or 5 atomic diameters.' according to Hecht, E. 1998. Physics Algebra/Trig, Second Edition. Brooks/Cole, Pacific Grove, CA, p.114.

69. 'A billion hairs' comes from Ruibal, R. & Ernst, V. 1965. The Structure of the Digital Setae of Lizards. Journal of Morphology. 117: 271–294. For a broader discussion of gecko glues, see Huber, G. et al. 2005. Evidence For Capillarity Contributions to Gecko Adhesion from Single Spatula Nanomechanical Measurements. Proceedings of the. National. Academy of Sciences, Volume 102 Number 45, pp.16293–16296.

70. A gecko weighing about 150g (5oz) can support a weight of 40kg (90lb), according to Geckos,, accessed 12 December 2013, which means it supports 267 times its own body weight. So if (and that's a big if) a human weighing 75kg (165lb) scaled up in exactly the same way, they could carry about 20,000kg or 20 tonnes. For a more accurate calculation, we'd need to consider the relative size of a gecko's feet and a person's hands and feet and take into account the fact that less of a human's body would come into contact with the ceiling.

71. Slippery Slope: Researchers Take Advice from Carnivorous Plant. Harvard School of Engineering and Applied Sciences, 21 September 2011.

72. For an interesting slant on the physics of ice sports such as curling, see Egan, J. Curling: The Friction Sport (Technical). Sports N'Science, University of Utah, accessed 11 October 2014.

73.You can watch Richard Feynman explaining the slipperiness of ice in an excerpt from the wonderful BBC series Fun to Imagine. Note how he very neatly sidesteps responsibility for someone else's theory with the insertion of a strategic 'they say'.

74. This is the theory expounded by Somorjai, G. & van Hove, M. Getting a Grip on Ice. Science, 9 December 1996.

p.98. Chad Orzel gave a very clear explanation of the various theories currently "in play" in The Surprisingly Complicated Physics Of Sliding On Ice, Forbes, 9 December 2016.

Chapter 7: The Inside Story

p.102. How big is an atom? It obviously depends which element you're talking about. The Physics Factbook suggests 0.1–0.5 nanometers is a good range for the diameter, while quotes a third of a nanometer for gold. I've opted toward the lower end of the range since most of the atoms in the world around us are made from simpler elements (such as hydrogen, carbon, oxygen, and so on).

75. According to two answers on the Argonne National Laboratory 'Ask a Scientist' website (Air in Clouds and Weight of Clouds), a typical cloud has a volume of several billion cubic metres. Assuming a liquid water content of about 0.3g (0.01oz) per cubic metre, we can calculate that the volume of the water in the cloud is probably somewhere between a few hundred thousand and a few million litres. Figures for cloud water content come from Linacre, E. & Geerts, B. Cloud Liquid Water Content, Drop Sizes, and Number of Droplets, accessed 11 October 2014.

p.104 John Dalton added "atomic insight" to an earlier theory called the Law of Definite Proportions (also called Proust's Law). In its original version, this theory notes how chemicals combine in certain constant ratios ("definite proportions") by weight. So, for example, water contains about 8 times the weight of oxygen relative to hydrogen (because oxygen has heavier atoms, even though there are twice as many hydrogen atoms in water). Picturing substances built out of atoms, Dalton took the idea much further and realized that the number of atoms of each substance that combine in molecules such as water are also in constant ratios (two atoms of hydrogen to one of water). He oversimplified and got it a little bit wrong (for example, concluding that water contains one atom of hydrogen and one atom of oxygen) but it was, nevertheless, a brilliant insight at a time when no-one had seen atoms and their very existence was still controversial. For a good account of why it took about 100 years for Dalton's ideas to be accepted, see Michalovic, M. 2008. John Dalton and the Scientific Method. Chemical Heritage Foundation, Spring 2008.

76.Cambridge University's Cavendish Laboratory has a clear and simple overview of J.J.Thomson's 1897 experiment. See Cambridge Physics: Discovery of the Electron, accessed 10 January 2014.

77. The anecdote is told by Thomson's grandson David in Davis, E. & Falconer, I. 1997. J.J. Thomson and the Discovery of the Electron. Taylor & Francis, London, p.xi.

78. It was a Nobel sacrifice, as well as a noble one. Marie and her husband Pierre jointly won the 1903 Nobel Prize in Physics for this work, with Henri Becquerel. See The Nobel Prize in Physics 1903,, accessed 10 January 2014.

79.Movie of the Week: Madame Curie, LIFE, 13 December 1943, pp.118–122.

80. Berkeley physics professor Richard Muller estimates that over a period of 50 years, the slightly higher natural radioactivity of Denver, Colorado will 'cause 4800 excess cancer deaths. That's more excess death than is expected from the Chernobyl nuclear accident!'. See Muller, R. 2008. Physics for Future Presidents. W.W. Norton, New York, p.117.

81. Gleason, S. 1955. 'Finding Uranium in the Dark'. Popular Science, July 1955, p.71.

82. Schoolboy, 13, Creates Nuclear Fusion in Penwortham. BBC News, 5 March 2014.

p.107 Splitting the atom: Brian Clegg has picked me up on a point of detail here—and I should clarify what I mean. To me, "splitting the atom" means not literally smashing atoms into bits (or "transmuting" one atom into another) but the broader collective effort of "figuring out what's inside atoms and how those bits are arranged". That's the point I'm making at the top of page 107. Although Rutherford was clearly the headline figure in the story, popular accounts of him "splitting the atom" suggest he was the only person involved in figuring out what atoms contain. Likewise, to say that Rutherford "split the atom" in a single experiment pretty much ignores his own, decades-long effort involved in reaching that point: if Rutherford "split the atom", it took him his entire working life to do it.

In fact, science is always a team effort and a long haul; understanding emerges and evolves collectively, by degrees, over years and decades. By my definition of "splitting the atom", the credit should be shared between a whole group of practical experimenters (Ernest Rutherford, Hans Geiger, Ernest Marsden, Lawrence Bragg, James Chadwick, Enrico Fermi, John Cockcroft, Ernest Walton, not to mention Henri Becquerel, Pierre Curie and Marie Curie, J.J. Thomson, et al) as well as theoreticians like Max Planck, Niels Bohr, Erwin Schrödinger, Paul Dirac, Werner Heisenberg, and Albert Einstein; all these people, working together, split—fathomed out the weird inner secrets of—the atom. By my reckoning, at least sixteen people won a Nobel Prize for their contribution to atomic physics (although Rutherford, somewhat ironically, won a "Stamp Collecting"—Chemistry—prize instead); the links I've added here take you to their Nobel Prize biography pages.

That's why the artwork on page 107 is labelled "Splitting the atom, old style": the gold-foil experiment was, to me, a key part of the broader effort that split—helped us understand the workings of—the atom.

For a good summary, please see Kumar, M. 2011. The Man who Went Nuclear: How Ernest Rutherford Ushered in the Atomic Age, The Independent, 3 March 2011, which puts Rutherford centre stage. For a longer version of the same story, see Cathcart, B. 2005. The Fly in the Cathedral. Farrar, Straus and Giroux, New York. That book tells the story of the other key figures and focuses on things that happened during Rutherford's tenure at the ("Old") Cavendish Laboratory in Cambridge—a lovely building I used to walk past in awe each morning as a student—whereas, of course, much of his early work was done in Canada and Manchester.

83. The Particle Adventure, an excellent website produced by the Particle Data Group at Lawrence Berkeley National Laboratory (LBNL), is the place to start if you're new to atomic physics. Accessed 10 January 2014.

84. Image of a 7 TeV proton-proton collision in CMS producing more than 100 charged particles, CERN, accessed 10 January 2014, This artwork is credited to Lucas Taylor, copyright © CERN 2010, and reproduced here under CERN's copyright conditions allowing free 'educational and informational use'.

85. There are 6×1023 atoms in 235g (8oz) or one mole of uranium-235 and each splitting atom will release about 3.2×10-11J according to Sowerby, M. Kaye and Laby: Tables of Physical and Chemical Constants, National Physical Laboratory, accessed 12 October 2014. So 235g of uranium-235 will produce about 20 trillion joules and 1g of uranium will produce about 100 gigajoules (or 100 gigawatts if it happens in one second).

86. In science, this kind of atomic arrangement is called close packing. Iron (a typical metal) has an interior, crystalline form called body-centred cubic (BCC), which means the atoms are arranged in the form of a cube, with eight sitting at the edges and one more squatting in the centre, in between the others. The basic 'unit cell', as this is called, repeats over and over again like a kind of three-dimensional patterned wallpaper.

87. Bacteria digest trash in landfills and there is some evidence that marine bacteria can pull off the same trick. See Zaikab, G. 2011. Marine Microbes Digest Plastic., 28 March 2011.

88. Nylon, the first real synthetic plastic textile, was launched on 27 October 1938. There were earlier plastics, but the arrival of nylon marked the true beginnings of the modern plastic age.

89. Kevlar., accessed January 10, 2014.

90. The figures in this chart are necessarily only a rough indication. They're compiled from various different sources including Household Waste – That's Garbage!, Michigan Waste Stewardship Program, accessed 10 January 2014; Save Our Beach, accessed 10 January 2014; and Surfers Against Sewage. 2010. Motivocean: Marine Litter: Your Guide. Surfers Against Sewage, St Agnes, Cornwall.

Chapter 8: Amazing Glazing

91.Professor Eric Le Bourhis dates the first glass to somewhere between 3500 and 5000 BC. See Le Bourhis, E. 2008. Glass: Mechanics and Technology. Wiley-VCH, Weinheim, p.29.

92. See Fosbroke, T. 1843. Encyclopaedia of Antiquities and Elements of Archaeology, Classical and Mediaeval, Volume 1. M.A. Nattali, London. 'Beckman observes that transparent windows were in the time of Seneca quite novel, Stubbs ascribes the introduction here of stone and glass windows to Wulfrid, Bishop of Worcester, in 736'.

93. See the section 'Pigmenting glass' in Langhamer, A. 2003. The Legend of Bohemian Glass: A Thousand Years of Glassmaking in the Heart of Europe. Tigris, Czech Republic.

94. Szasz, F. 2006. 'J. Robert Oppenheimer and the State of New Mexico' in Kelly, C. (ed.) 2006. Oppenheimer and the Manhattan Project. World Scientific, Singapore.

95.For a very detailed discussion, see Zallen, R. 2008. The Physics of Amorphous Solids. John Wiley & Sons, Weinheim, p.3: 'In amorphous solids, long-range order is absent; the array of equilibrium atomic positions is strongly disordered'.

96. Debennetti, P. & Stanley, H. 2003. Supercooled and Glassy Water. Physics Today, June 2003, p.40: 'Glassy water may be the most common form of water in the Universe. It is observed as a frost on interstellar dust, constitutes the bulk of matter in comets, and is thought to play an important role in the phenomena associated with planetary activity'.

97. Szczepanowska, H. 2013. Conservation of Cultural Heritage: Key Principles and Approaches. Routledge, London, p.239. The section 'Myths about glass' quotes Dr Robert Brill, pointing out that 'the viscosity of glass is probably a billion times higher than metallic lead and we never see lead flowing down from stained glass windows (Brill, 2000)'.

98. In materials science jargon, glass has a low fracture toughness and a low work of fracture. These are two related (but different) measurements of how much energy it takes to make a crack spread through a material. The incoming energy has to go somewhere and if it can't deform a material, the material breaks instead. Balloons have such a low work of fracture that even the tiny point of a pin can make cracks spread through them instantly, making them burst violently.

99. See Coefficients of Cubical Expansion of Solids' in Lange's Handbook of Chemistry. McGraw-Hill, New York, 2004.

100. The density of glass is about 2500 kg/m3, so a cubic metre of glass would weigh 2.5 tonnes.

101. Without going too deeply into the physics, this is essentially a band gap argument, summarised in Smallman, R. & Bishop, R. 1999. Modern Physical Metallurgy and Materials Engineering: Science, Process, Applications. Butterworth-Heinemann, Oxford, p.195. There's also an explanation in Chandrasekhar, B. 1997. Why Things Are the Way They Are. Cambridge University Press, Cambridge, pp.173–177. Chandrasekhar's book is well worth a look for a deeper understanding of how solids really work "inside", though readers who enjoy my book will probably find it much too complex.

p.125.Heat-reflecting windows: I've compressed this explanation in the interests of keeping things simple, but I can see that some people might be confused: "Hang on, glass doesn't let infrared radiation through it... and isn't sunlight more about ultraviolet?" So let me just expand things a bit...

First, sunlight is a mixture of wavelengths. Using the familiar electromagnetic bands, we can say that it's roughly 50–50 visible light and infrared (actually 44–47 percent visible and 51–53 percent infrared, but let's not quibble), with a tiny amount of ultraviolet (just 2–3 percent) thrown in (there are various sources for this online, including the graph linked below, but I've not found a really good definitive source). So it seems fair to say that most of the heat in sunlight is infrared.

Second, the term "infrared" covers a vast band of wavelengths, from the red edge of the visible spectrum (informally called near infrared, roughly 750 nanometers, and including the infrared we use for remote controls, more like 900–950 nanometers) to the edge of the shortest radio waves (1mm long—or 1 million nanometers), which is far infrared. If you need a quick refresher, NASA have a great little guide to Infrared Energy.

Now the crucial point is that glass doesn't treat all infrared radiation the same way. The shorter wavelengths in near infrared are mostly transmitted: glass transmits over 90 percent of near-IR with a wavelength of around 900 nanometers, according to this handy graph of the solar energy spectrum. That makes sense, if you think about it, because that sort of infrared is not so very different from visible light and the distinction we make between visible light and infrared is based, essentially, on what our eyes can detect, not any hard and fast division in nature itself. Once we get into the longer IR wavelengths, transmission falls off dramatically. Judging by this graph, an IR remote control with a wavelength of about 1000 nanometers would be more absorbed than transmitted, which is why you'll struggle to switch your DVD player on and off through closed glass doors. When we get into the really long wavelengths of the far-infrared, you can see that transmission falls right off; these wavelengths are almost entirely absorbed. One reason why normal windows lose heat is because the longer wavelengths of infrared coming from a warm room are absorbed by the glass and then lost to the air by conduction (through the glass) and then convection (through the air outside). The other reason is that they tend to leak warm air or let in cold draughts. Covering glass with a reflective coating means the infrared bounces back into the room, reducing the heat loss.

For a longer discussion of keeping heat in and out of buildings, please see "Chapter 9. Controlling the Radiation of Heat" in Allen, E. 2005. How Buildings Work: The Natural Order of Architecture. Oxford University Press, Oxford/New York.

102. McCollough, F. 2008. Complete Guide to High Dynamic Range Digital Photography. Lark Books, New York, p.13. This suggests that the average luminance in candela per square metre for a sunny sky is about 100,000, compared to just 50 for indoors.

103. See Boyd, R. 1957. Design of glass for daylighting in Windows and Glass in the Exterior of Buildings: A Research Correlation Conference Conducted by the Building Research Institute. National Academies Press, Washington, 1957, p.8.

104. Ferrell McCollough (see note 102 above) quotes indoor light as 50 candela per square metre.

105. For a detailed explanation of electrochromic windows, see Arntz, F. et al. 1992. US Patent 5,171,413: Methods for Manufacturing Solid State Ionic Devices, 15 December 1992. The essential similarity between the two technologies is obvious from the very first sentence, describing a 'device usable as an electrochromic window and/or as a rechargeable battery'.

106. Armistead, W. & Stookey, S. 1962. US Patent 3,208,860: Phototropic Material and Article Made Therefrom. 28 September 1965.

Chapter 9: Saggy Sofas, Squeaky Floors

p.132. Razor blades versus hairs: You'll find this fascinating topic covered in "Chapter 1. Indomitable" of Miodownik, M. 2014. Stuff Matters. Penguin, London. Razor blades are covered in a whistle-stop history of steel and the explanation for why puny hairs blunt tough blades is on pp.20–21 of the ebook (sorry, I don't have the printed copy).

107.. 'In 1928, Walter Diemer.... created bubblegum. He used a rubber tree latex'. See Chapter 3. The History of the Chewing Gum Industry in the Americas in Mathews, J. 2009. Chicle: The Chewing Gum of the Americas, From the Ancient Maya to William Wrigley. University of Arizona Press, Tuczon, AZ.

108. 'Chewing gum contains a base that is made from natural rubber, styrene butadiene, or polyvinyl acetate'. Askeland, D. et al. 2010. Essentials of Materials Science and Engineering. Cengage Learning, Stamford, CT, p.527.

109. If you've got a pair of polarising sunglasses (non-polarising ones won't work) and a laptop or tablet computer, you can experiment with photoelasticity for yourself. Open a word-processing program with a blank document and maximize it so the whole screen is completely white. Now put on your sunglasses and hold some transparent plastic objects between your eyes and the computer screen. You should see some amazing psychedelic spectral colours. What happens when you stress the objects or rotate your head? Here's my own little experiment on birefringence and photoelasticity, following the very simple method outlined in a great tutorial by PhotoExtremist Evan Sharboneau.

110. Professor James Gordon argued that the invention of the pneumatic tyre was as significant as the internal combustion engine: '....if an effective pneumatic tyre had been available around 1830, we might have gone direct to mechanical road transport without passing through the intervening stage of railways at all'. Gordon, J. 1978. Structures. Penguin, London, p.314.

111. The modulus of elasticity of steel (also called Young's modulus) is about 200,000MPa, compared to rubber, which is about 1MPa. Glaser, R. 2001. Biophysics. Springer, Berlin, p.213.

112. Sun, J. et al. 2012. Highly stretchable and tough hydrogels. Nature 489, p.133–136.

113. Gordon, J. 1978. Structures. Penguin, London, p.54 quotes a figure of 0.2MPa for the modulus of elasticity of a pregnant locust compared to 7MPa for rubber.

114. See Section 5.5 The Reversibility in Chandrasekaran, V. 2010. Rubber as a Construction Material for Corrosion Protection. John Wiley & Sons, Hoboken.

115. For a general discussion of the mechanical properties of facial skin, see Piérard, G. et al. 2010. Chapter 27. Facial Skin Rheology. In Farage, M. et al (eds). Textbook of Aging Skin. Springer, Heidelberg.

116. The modulus of elasticity of glass is about 70,000MPa, which suggests glass is at least twice as elastic as steel.

117. This figure is loosely inspired by the more abstract drawing of how cracks concentrate stress in Gordon, J. 1978. Structures. Penguin, London, p.66.

118. For a brief overview, see Chapter 8. When Metals Tire. In Levy, M. & Salvadori, M. 1992. Why Buildings Fall Down. W.W. Norton, New York. There is a much more detailed discussion in Withey, P. 1966. Fatigue Failure of the De Havilland Comet I. In Jones, D (ed). 2001. Failure Analysis Case Studies II. Elsevier Science, Oxford.

119. Goldsmith-Carter, G. 1969. Sailing Ships & Sailing Craft. Hamlyn, London, p.140

120. For a description of how the naturally flexible wooden hull of the St Roch enabled it to break free of the Arctic ice, see Delgado, J. 1985. Across the Top of the World: The Quest for the Northwest Passage. Douglas & McIntyre, Vancouver, p.185

121. The Polar Ship Fram, Fram Museum, accessed 21 January 2014.

122. See Chapter 4. Leather Preservation. In Smith, C. 2003. Archaeological Conservation Using Polymers: Practical Applications for Organic Artifact Stabilization. Texas A&M University Press, College Station, p.61.

123. For a more detailed explanation, see my article Thermochromic color-changing materials, Explain that Stuff, accessed 21 January 2014.

124. Clout, L. Splendour of new Wembley fading already. The Telegraph, 10 May 2007.

p.145. "...over 100 components in creosote have been identified" according to Creosote and its Use as a Wood Preservative, US Environmental Protection Agency, November 2008 (archived page via the Wayback Machine). Bark works like a natural creosote according to Tree Bark by Dan Puplett, Trees for Life, accessed 7 February 2015.

125. Pohanish, R. 2011. Sittig's Handbook of Toxic and Hazardous Substances. Elsevier, Oxford, p.736.

126. Johnson, J. 2011. Old-Time Country Wisdom & Lore: 1000s of Traditional Skills for Simple Living. Voyageur Press, Minneapolis, p.51.

127. A figure of 5km/s (3 miles per second) is quoted by Mia Siochi in NASA's educational video Real World: Self Healing Materials, YouTube, accessed 21 January 2014, uploaded June 12, 2009.

Chapter 10: Light Delights

128. Different sources say 1703 and 1704, but the earliest facsimiles of the cover I've found clearly show a date of MDCCIV (1704). Newton had first published his thoughts about light some 30 years before (coincidentally, at the age of 30), in a paper in the Philosophical Transactions of 1672.

129. Thanks to Cambridge University, you can flick the pages of Newton's notebooks from the comfort of your computer. See Isaac Newton: Laboratory Notebook, Cambridge University Digital Library, accessed 23 October 2014.

130. Wave-particle duality (the concept that light behaves as both particles and waves) is not a new idea. For a good account of how the pioneers of optics saw light in the 18th century, see Shamos, M. 1959. Great Experiments in Physics: Firsthand Accounts from Galileo to Einstein. Dover, New York, p.93.

131. In 1900, Lord Kelvin (William Thompson) daringly told the British Association for the Advancement of Science that 'There is nothing new to be discovered in physics now'. Although this widely recycled quote has been disputed, it seems to be an accurate summary of what he believed. In a paper published the following year, he pointed out that the 'beauty and clearness' of physics was obscured only by two clouds – precisely the areas that would soon be explored by relativity and quantum theory. Kelvin, Lord. 1901. Nineteenth Century Clouds over the Dynamical Theory of Heat and Light. Philosophical Magazine and Journal of Science, S. 6. Vol. 2. No. 7., p.1. Another distinguished physicist, Albert Michelson, apparently said something similar at a lecture in Chicago in 1894 (some sources attribute it later, to 1903 or 1904): "The most important fundamental laws and facts of physical science have all been discovered, and these are now so firmly established that the possibility of their ever being supplemented in consequence of new discoveries is exceedingly remote". However, Michelson did acknowledge the possibility of "future discoveries... in the sixth place of decimals" (minor tweaks only, in other words). I found that quote in Barrow, J. 1998. Impossibility: The Limits of Science and the Science of Limits. Vintage, London, p.54.

132. Hecht, E. 1998. Physics Algebra/Trig. Brooks Cole, Pacific Grove, p.806.

133. No-one seems to be able to put an exact figure on this, but there are credible estimates varying from 25–60 per cent in many books and across the Web. For a quick overview of how the brain determines what we see, see Grady, D. The Vision Thing: Mainly in the Brain, Discover June 1993. The same theme is explored in much more detail in the classic (and very readable) book: Gregory, R. 1997. Eye and Brain: The Psychology of Seeing. Princeton University Press, Princeton.

134. Interestingly, Newton had a 'premonition' of Einstein's equation three hundred years earlier. In Opticks, question 30, he writes: 'The changing of Bodies into Light, and Light into Bodies, is very conformable to the Course of Nature.' – which sounds very much to me like E=mc2.

135. In terms of length alone, the difference between the astronomical and the 'atomical' is vast. If we divide the estimated diameter of the observable Universe (100 billion light years) by the diameter of an atom (0.25 nanometres), we get about 4 trillion trillion trillion (4×1036) or, in old money, 4,000,000,000,000,000,000, 000,000,000,000,000,000.

136. For a useful summary of the electromagnetic spectrum, see What Wavelength Goes With a Color?, NASA, accessed 29 January 2014.

137. This is the average sort of speed reached by surfers, according to one of the classic books on ocean waves, Bascom, W. 1980. Waves and Beaches. Anchor Press, New York, p.241. Open ocean waves travel much faster, especially in tsunamis.

138. Interestingly, because light is so much bigger than atoms, we have no hope of seeing either atoms or molecules with ordinary optical (light-based) microscopes. That's why electron microscopes were invented. Since (in a crude and simplistic sense) electrons are much 'smaller' than photons, they can make images of much smaller things. The word 'smaller' becomes problematic when we talk about the probabilistic subatomic, but let's not worry about that here.

139. Helium-neon red laser photons each have about 3 billionths of a joule (3×10-19J) of energy according to Hecht, E. 1998. Physics Algebra/Trig. Brooks Cole, Pacific Grove, p.807.

140. A typical flashlight bulb is about 2.2 volts and 0.25 amps, which makes 0.55 watts. Dividing 0.55 watts by the energy of a photon quoted above (3×10-19J) gives us about 2×1018 photons per second.

141. Cathcart, B. 2005. The Fly in the Cathedral. Farrar, Straus and Giroux, New York.

142. Although energy zaps from the Sun to Earth in just a few minutes, it takes thousands of years for it to get from the nuclear core of the Sun to the outer layers so it can make its escape. See Plait, P. 1997. The Long Climb from the Sun's Core, Bad Astronomy, accessed 15 October 2014.

143. Molten volcanic lava comes in at about 1200°C (2200°F) according to Schminke, H. 2004. Volcanism. Springer, Berlin, p.27.

144. For an entertaining account of the extent to which the idealized myth of Thomas Edison departs from what we can establish of historical reality, see Stross, R. 2007. The Wizard of Menlo Park. Crown, New York. The infamous guard bear is described in Jehl, F. 1924. Edison the Man: An Old Friend's Recollections of the Great Inventor. Popular Science, February 1924, p.31.

145. Silicon, one of the commonest chemical elements on Earth, is often confused with silicone, a type of rubbery plastic (polymer) used in breast implants. Although silicone contains silicon, that's where the resemblance begins and ends. The silicon used in 'silicon chips' (computer microchips) is much closer to the raw stuff you find in sand than the stuff lurking in a brassiere.

p.160. In my explanation of n-type and p-type silicon, I'm referring only to the silicon atoms and not the material as a whole. Doping gives the silicon atoms in n-type extra free electrons (compared to the number they have normally), so they are therefore, in a collective sense, negatively charged compared to their undoped state. Even so, the material as a whole is electrically neutral—and there's no contradiction. Each one of those extra electrons comes from an atom of an added impurity, which is locked into the silicon lattice as a positive ion (a positively charged atom). For each extra negatively charged electron, there's a matching positive ion, making no charge overall. If you find that confusing, another way of looking at it is to remember that we start off with neutral silicon and add a neutral impurity and we can't create electric charge out of thin air. The same applies to p-type silicon (the moving, positively charged holes are balanced by the static, negatively charged boron atoms), which is also electrically neutral overall.

146. Sean Palmer suggests a firefly produces light of about 1/50 and 1/400 candlepower. See What distance is a firefly visible from?, Sean B. Palmer's Shared Objects, accessed 15 March 2016.

147. For a short introduction to the Hubble's troubles, see Hubble's painful birth, BBC News, 10 February 2000; there's also a great little NASA picture showing the Hubble mirror polishing as it happened. For a longer and better account, see Zimmerman, R. 2010. 'Chapter 4: Building it' and 'Chapter 5: Saving it' in The Universe in a Mirror: The Saga of the Hubble Telescope and the Visionaries who Built it. Princeton University Press, Princeton, NJ.

p.162. Walking "boats" are, of course, walking boots.

Chapter 11: Radio Gaga

148. The 'luminiferous aether' (sometimes spelled ether) finally packed its bags after the famous 1887 Michelson-Morley experiment, in which Albert Michelson and Edward Morley devised a cunning test to detect its existence and measure its speed. Finding a speed of zero, they effectively proved the aether didn't exist, suggested the speed of light was always the same – and paved the way for Einstein's shiny new theory of relativity about 20 years later.

p.166. "Mars... 20 minutes at most" is a reference to the changing distance between Earth and Mars. See Mars in our night sky, NASA Mars Exploration, accessed 8 February 2015.

149. A Chat with the Man Behind Mobiles. BBC News, 21 April, 2003. Martin Cooper's main patent for the mobile phone was filed in 1973 and granted two years later. See Cooper, M. 1975. US Patent 3,906,166: Radio Telephone System, 16 September 1975.

150. In an entertaining history of how the UK and North America were linked up electrically, Gillian Cookson suggests it took about 12 days to send a message across the Atlantic by a combination of steam ship and electric telegraph during the 1850s. See Cookson, G. 2012. The Cable: The Wire That Changed the World. The History Press, Stroud. Once Cyrus Field's pioneering submarine cable between Ireland and Newfoundland was completed, in 1858, it was soon sending about 150 messages a day (the total traffic in both directions). A PBS article about Field's cable suggests it was sending only 50 messages a day in 1866, mostly due to the high cost of transmission ($10 per word). See The Great Transatlantic Cable, PBS Online, accessed 3 February 2014.

151. It takes light travelling at 300,000km/s (186,000 miles/s) about 0.02 seconds to travel the straight-line distance of approximately 5500km (3400 miles) between London and New York City.

152. See for example the classic paper Nyquist, H. 1924. Certain Factors Affecting Telegraph Speed. Bell System Technical Journal, 3, pp.324–346.

153. For a surprisingly complete and entertaining account, see Elisha Gray and Alexander Bell Telephone Controversy, Wikipedia, accessed 5 February 2014. Few people had heard of Meucci until his contribution was finally recognized in a resolution passed by the United States House of Representatives on June 11, 2002. See Who is Credited as Inventing the Telephone?, US Library of Congress, accessed 5 February 2014.

154.The photophone is described in detail in Bell, A. 1880. US Patent 235,199: Apparatus for Signalling and Communicating, called 'Photophone', 7 December 1880. There's also some discussion of the photophone's historical importance in Bertolotti, M. 2005. The History of the Laser. IOP Publishing, Bristol, p.269.

p.170. James Clerk Maxwell: "playful" is a reference to the way Maxwell visualized essentially abstract, mathematical theories of matter, electromagnetism, and, perhaps most playfully of all, entropy (famously imagining a little demon who could defeat the second law of thermodynamics by sorting out molecules into hot and cold). There's a great introduction to this side of Maxwell in Newman, J. 1955. James Clerk Maxwell. Scientific American, June, p.58. For a friendly introduction to Maxwell, try Will Self's lugubrious road-trip Self Drives: Maxwell's Equations (five short MP3 podcasts).

p.170. James Clerk Maxwell: The date "1873" refers to the publication of Maxwell, J. 1873. A Treatise on Electricity and Magnetism. Clarendon Press, Oxford. Wikipedia's short article about A Treatise on Electricity and Magnetism includes three stunning quotes from Albert Einstein, Richard Feynman, and Carl Sagan that underline the extraordinary importance of Maxwell's work.

155. TIME magazine went some way to setting the record straight by naming Farnsworth one of its 100 most influential figures of the 20th century. See Postman, N. 1999. Electrical Engineer Philo Farnsworth, TIME, 29 March 1999.

156. Suppose you're listening to FM radio on a frequency of about 90MHz. The wavelength is the speed of light divided by the frequency or about 3.3m (11ft), so a suitable FM antenna would be round about 1.5m (5ft) long.

157. The maximum distance you can see a lighthouse is 3.57km (2.2 miles) times the square root of its height in metres. So, if a lighthouse is about 30m (100ft) tall, you can see it about 20km (12 miles) away. I'm using a formula by Young, A. Distance to the Horizon, accessed 5 February 2014.

158. Marconi won the 1909 Nobel Prize in Physics. The lecture he delivered is a fascinating, first-hand account of his experiments and the scientific understanding of the time; the transcript is well worth reading. Marconi, G. 1909. Nobel Lecture: Wireless Telegraphic Communication,, accessed 3 February 2014. The quotations from Edison and Bell come from my own mini-biography: Woodford, C. 2008. "Guglielmo Marconi". In Poole, H. (ed.) Inventions and Inventors: Volume 4, Marshall Cavendish, Tarrytown, p.1011.

159. It was simultaneously developed by Arthur Kennelly and finally confirmed by Edward Appleton in 1924. Appleton won the 1947 Nobel Prize in Physics for this work – an honour that Kennelly and Heaviside, both then deceased, were unable to share. See Edward V. Appleton – Biographical,, accessed 3 February 2014.

160. These and other colourful anecdotes are quoted by Nahin, P. 2002. Oliver Heaviside: The Life, Work, and Times of an Electrical Genius of the Victorian Age. JHU Press, Baltimore, MD, p.xx.

161. Searle, G. 1950. Oliver Heaviside: A Personal Sketch. In The Heaviside Century Volume, IEE, London.

162. Clarke, A. 1945. Extra-Terrestrial Relays: Can Rocket Stations Give World-Wide Radio Coverage?.Wireless World, October 1945.

p.176. For a great photograph of Echo, see Echo: A Passive Communications Satellite, Great Images in NASA, accessed 8 February 2015. Click on the large version of the photo (links at the bottom of the page) and note the scale of the figures and car underneath.

163. The Straight Dope website quantifies it as 1000 watts for a microwave oven compared to a few hundred milliwatts for a cellphone. See How are the Microwaves in Ovens Different from those in Cell Phones?, The Straight Dope, 28 August 2003.

164. This is a wild guesstimate simply based on the Straight Dope's comparison of the power, assuming a mobile phone could beam microwave energy into food in exactly the same way as a microwave oven. In reality, a cellphone wouldn't generate anything like enough temperature to cook food: even if it could eventually beam enough energy into your dinner, that's not the same as cooking it. You can leave your dinner outside in the Sun for as long as you like, but it still won't cook unless it reaches a certain, critical temperature.

Chapter 12: Living by Numbers

165. ENIAC: Celebrating Penn Engineering History. Penn Engineering, accessed 9 February 2014.

166. See Woodford, C. 2001. Vannevar Bush. and Van der Spiegel, J. 2001 ENIAC. In Rojas, R. (ed.) 2001. Encyclopedia of Computers and Computer History. Fitzroy Dearborn, Chicago.

167. See for example the article Digital Versus Film Photography, Wikipedia, accessed 15 October 2014.

168. The only feasible way to listen in on digital mobile calls is to install listening equipment in the handset or tap into the equipment in the phone provider's switching office. That's why hacking into mobile phone messages became such a popular pastime for journalists in the period from the mid-1990s to the mid-2000s. An old telecoms industry friend of mine, who knows about such things, once told me the mobile phone companies have sinister, high-security rooms fitted with secret snooping equipment to which only the security services have access – an idea I instantly brushed aside as conspiracy-theory paranoia. In 2014, the mobile phone operator Vodafone finally confirmed that it is indeed true. See Garside, J. 2014. Vodafone Reveals Existence of Secret Wires that Allow State Surveillance. The Guardian, 6 June 2014.

169. The Library of Congress has an ever-expanding collection: 36 million books is the figure cited at the time of writing the book (2014), though that's since increased to 38 million. See Fascinating Facts, US Library of Congress, accessed 16 March 2016.

170. iTunes began dropping DRM from its music files in 2009. See Apple to End Music Restrictions, BBC News, 7 January 2009.

171. In the United States, at least, under the Digital Millennium Copyright Act (DMCA), signed into law by President Bill Clinton on 28 October 1998.

172. This method of cracking copyright protection is sometimes referred to as the Analog(ue) Hole – and the media multinationals have their sights set on trying to close it. See Analog Hole, Electronic Frontier Foundation, accessed 16 October 2014.

173. In theory, if you sample fast enough, you could capture all the important information in your original sound. We know that from something called the Nyquist-Shannon sampling theorem. In practice, sampling too fast makes MP3 files too big and most people (those who listen to bad MP3s through cheap headphones on noisy commuter trains) are happy to tolerate poorer-quality, highly compressed files if it means they can store far more music on their iPod or mobile phone.

174. CDs are much more impressive than I've made them sound – and seemed even more so when they were first introduced. If you're old enough, you might remember TV demonstrations from the early 1980s where hapless science presenters dropped CDs on the floor or smeared them with jam to prove that they were still playable after rough treatment. That kind of thing is made possible by a handy bit of mathematical magic built into their design, called the cross-interleave Reed-Solomon error-correction code (CIRC), which allows them to whistle past random errors caused by things like scratches and fingermarks. It's so good that it can (theoretically – and very impressively) compensate for a scratch 4000 bits or 2.5mm (0.1in) long. See Baert, L. 1995. Digital Audio and Compact Disc Technology. Focal Press, Boston.

p.194 A couple of people have misunderstood the sentence "For simplicity, let's assume the CD is encoded so each pit is equivalent to one bit". Data storage on CDs is extremely complex and most people don't need (or want) to know exactly how their music is encoded; for precisely that reason, in a number of my children's books, and in my detailed article about CDs and DVDs, I have intentionally chosen to simplify the explanation by suggesting that pits and lands on a CD encode zeros and ones respectively. As I say, this is a very deliberate simplification. The reality is more complex: the transition from a pit to a land encodes a one and an unchanging sequence of either pits or lands encodes a zero. You can read a bit more about it at the end of my article or in a more detailed text like The Compact Disc Handbook by Ken C. Pohlmann, from page 74 onward.

p.195 A fragile permanence: In February 2015, just after this book went to press, Vint Cerf, one of the Internet's founding fathers, made big headlines when he expressed similar fears about the impermanence of online information, proposing something he calls "digital vellum" as a solution (see Google's Vint Cerf warns of 'digital Dark Age' by Pallab Ghosh, BBC News, 13 February 2015). Others have been flagging up this problem for nearly two decades. According to Wikipedia, the term "digital Dark Age" was first used in 1997 and later popularized by Danny Hillis, creator of the Long Now project. The most realistic solution so far is Brewster Kahle's superb Internet Archive and Wayback Machine. But who archives the archive... and the archive of the archive...?

It could be argued that the anarchic (bottom-up) Web is so redundant that you don't need an archive for much of it anyway. That's a subtle and very important point that the archivers tend to ignore, perhaps for dramatic effect or because exaggerating the problem helps them fund their own favoured solutions. Just as the brain is a distributed architecture that can survive damage to a really remarkable extent, so much of the knowledge encapsulated in the World Wide Web is dispersed rather than localized in particular places. Even if the BBC deleted its priceless archive of online news stories, many of them will have been copied onto other sites, and most of the stories will have been covered by other people; so an archive of BBC News is actually less valuable than it might appear. Not that I'm arguing against archives of the Web, simply pointing out that the fundamental structure and inherent redundancy cleverly avoids the problem a lot of the time.

175. Thanks to the British Library, the Gutenberg Bible has been digitally preserved, in perpetuity, but who knows if the digital facsimile will last as long as the paper original? See Gutenberg Bible, British Library, accessed 9 February 2014.

176 Ray Tomlinson has explained the thinking behind the first email. Why did he do it? 'Mostly because it seemed like a neat idea'. Why the strange at sign (@)? 'The primary reason was that it made sense. At signs didn't appear in names so there would be no ambiguity about where the separation between login name and host name occurred.' See The First Network Email, Raytheon BBN Technologies, accessed 9 February 2014.

177. There's more about the BBC project at BBC Domesday Reloaded: Story of Domesday, BBC, accessed 8 February 2014.

178. Radicati, S. 2012. Email Statistics Report 2012–2016. The Radicati Group, Palo Alto.

p.196. If you have some sympathy with my question "Modern life might be digitally dictated, but is it necessarily better that way", another book to put on your reading list is The Revenge of Analog: Real Things and Why They Matter by David Sax. Public Affairs, 2016.

Chapter 13: Blowing Hot and Cold

179. The weight of a typical iceberg is about 150,000 tonnes (165,000 tons) and its internal temperature is about −15°C (5°F). So the heat energy it contains (heating it from absolute zero) is equal to its mass × specific heat capacity of water × its temperature above absolute zero = 150,000,000kg × 4181J/kg/°C × (273−15) = 162,000 gigajoules. A very large cup of coffee might be about 0.5 litres and at a temperature of 90°C (194°F), so its heat energy is 0.5kg × 4181J/kg/°C × (90+273)°C = 760 kilojoules. The iceberg has about 200 million times more heat energy. My iceberg statistics come from Just the Facts: Icebergs, Canadian Geographic, March/April 2006.

180. Black holes don't have to be cold, however, and space scientists are currently working to make 'the coolest spot in the Universe' (with a temperature of just 1 picokelvin, 10-12K or 0.000000000001K). See Cold Atom Laboratory, NASA, accessed 16 October 2014.

181. Hand, E. 2012. Hot stuff: CERN physicists Create Record-Breaking Subatomic Soup. Nature, 13 August 2012.

182. Bingelli, C. 2009. Building Systems for Interior Designers. John Wiley & Sons, Hoboken, NJ, p.22.

183. Ice import: Norway's Ice to London. London Canal Museum, accessed 11 February 2014.

p.206 "... lack of any appreciable convection." There's obviously plenty of convection happening outside, but it doesn't carry heat to you from a fire the way convection does indoors, where the heated air is trapped and forced to circulate.

p.207 "Air conditioners... also function as heaters." I'm referring to combined HVAC (heating, ventilating and air conditioning) units here.

184. Mrs. Marshall's Liquid Air Ice Cream, Rowley's Whisky Forge blog, September 2010.

p.209 Going underground: An interesting 2015 BBC News article describes how Drammen in Norway generates huge amounts of power from an ice-cold fjord using heat-pump technology. See Anderson, R. 2015. Heat pumps extract warmth from ice cold water, BBC News, 10 March 2015.

185. I'm using figures for typical costs and savings compared to electric storage heaters calculated by the UK's Energy Saving Trust. See Ground Source Heat Pumps, Energy Saving Trust, accessed 16 March 2016.

186. Quoted in Fox, S. 2010. Superinsulating Aerogels Arrive on Home Insulation Market At Last, Popular Science, 2 April 2010.

187. The Passivhaus Standard: What is Passivhaus?, Building Research Establishment, accessed 11 February 2014.

188. See for example Fischer, B. "How much does it cost to charge an iPhone 5? A thought-provokingly modest $0.41/year", Opower Outlier Blog, 27 September 2012.

p.215 "C-shaped Cray supercomputers": Unquestionably my favourite computers of all time! I'm referring specifically to the Cray-2 from the mid-1980s. The Computer History Museum has a wonderful online exhibit about the Cray-2, including a photo of Seymour Cray pictured with a model of the Cray-2's Fluorinert cooling system. There's also a great archived brochure [PDF format], in which the "liquid immersion cooling system" is explained in delicious detail: the Fluorinert coolant "flows through the circuit boards at a velocity of one inch per second and is in direct contact with the integrated circuit packages and power supplies". Now, how cool is that?

189. Without Much Fanfare, Apple Has Sold Its 500 Millionth iPhone, Forbes, 25 March 2014.

190. Greenpeace International. 2012. How Clean is Your Cloud, accessed 19 October 2014. Also worth a look is the earlier report Greenpeace International. 2011. How Dirty is Your Data Center: A Look at the Energy Choices that Power Cloud Computing, accessed 23 March 2015 [PDF format].

191. Desktop versus laptop, EU Energy Star, accessed 10 July 2015. (Archived link via the Wayback Machine.)

p.216 A couple of people have raised an eyebrow about my statement that computers get hot, noting that their own machines stay blissfully chilled. But getting hot is a basic problem of high-performance computing—and always will be. Even the mooted move to quantum computing (pioneered by controversial companies such as D-Wave) brings no relief: "Much of the D-Wave hardware's power consumption—slightly less than 25 kilowatts for the latest machine—goes toward running the refrigeration unit that keeps the quantum processor cool. The quantum processor itself requires a comparative pittance," according to How Much Power Will Quantum Computing Need? by Jeremy Hsu, IEEE Spectrum, 5 October 2015. From the same source, here's a great recent review of cutting-edge chip-cooling techniques: Four New Ways to Chill Computer Chips by Prachi Patel, 20 November 2015.

Chapter 14: Food Miles

192. As explained later in this chapter, I am adopting the convention of using a capital to represent nutritional Calories, where 1 nutritional Calorie = 1000 thermal calories = 4.2 kilojoules (4200 joules) of energy.

193. The data for this pie chart is compiled from a number of different sources, including Insel, P. et al. 2010. Nutrition (Fourth Edition). Jones & Bartlett, Sudbury, MA, p.342.

194. 'On average, efficiency ranges between 20 and 25% for walking, running, and stationary cycling.' McArdle, W. et al. 2010. Exercise Physiology: Nutrition, Energy, and Human Performance. Lippincott Williams & Wilkins, Baltimore, MD, p.208.

195. Walking briskly for an hour at about 6km/h (4mph) consumes about 450 Calories. A large egg contains 75 Calories, so you could walk for about 10 minutes on an egg and go about 1km (0.6 miles).

196. Column 1 is compiled mostly using nutritional data from Gebhardt, S. & Thomas, R. 2002. Nutritive Value of Foods, US Department of Agriculture: Agricultural Research Service, Home and Garden Bulletin Number 72, October 2002, accessed 17 February 2014. A few entries have been compiled directly from the nutritional labels on food from typical grocery store products.

197. My rough exercise equivalents are based very loosely on the figures suggested by Insel, P. et al. 2010. Nutrition (Fourth Edition). Jones & Bartlett, Sudbury, MA, p.344.

198. The winter energy consumption of a musk oxen is about 920kJ (219 Calories)/kg of body weight, while summer consumption (when the animals are much more active) rises by about a quarter to 1163kJ (278 Calories)/kg. See Kazmin, V. D. & Abaturov, B. D. 2011. Quantitative characteristics of nutrition in free-ranging reindeer (Rangifer tarandus) and musk oxen (Ovibos moschatus) on Wrangel Island, Biology Bulletin, Volume 38, Issue 9, p.935.

199. A hummingbird has a BMR to mass ratio of about 870kJ/hr per kg of body mass, while a human clocks in at just 77kJ/hr per kg, according to Sherwood, L. et al. 2012. Animal Physiology: From Genes to Organisms. Cengage, Independence, KY, p.720.

200. Mice eat 12g (0.4oz) of food per 100g (3.5oz) body weight per day according to Species Specific Information: Mouse, Johns Hopkins University Animal Care and Use Committee, accessed 19 February 2014.

201. 1 litre of gasoline contains 34.8 megajoules (34,800 kilojoules) of energy, according to Dunlap, R. 2015. Sustainable Energy. Cengage, Stamford, CT, Table 16.2. So 1 UK gallon (about 5 litres) contains about 160,000 kilojoules.

202. At the time of writing, and using UK prices, a litre of petrol costs about £1.30, which works out at about 0.00374p per kilojoule. A typical fast-food burger might cost £4.00, which is more like 0.2p per kilojoule (over 50 times more for the same energy). My cost for electricity is based on a unit price of 20p per kWh that doesn't take account of the annual standing charge. That works out at 0.006p per kilojoule.

203. Estimated Energy Requirements, Health Canada, 8 November 2011.

204. Breiter, M. 2008. Bears: A Year in the Life. A & C Black, London, p.157.

205. Percent of Consumer Expenditures Spent on Food, Alcoholic Beverages, and Tobacco that were Consumed at Home, by Selected Countries, 2012, US Department of Agriculture Economic Research Service, accessed 15 February 2014.

p.227. "It would, however, be interesting to know if we use more energy now (fuelling machines more and our brains and bodies less) than we did back then..." It turns out that exactly this calculation has been done by Earl Cook (in the 1970s) and, more recently, by Ian Morris of Stanford University. The Guardian has a great interactive demonstration at the start of its article Energy: Our Giant Leap, which (as the name suggests) shows that there's been a massive increase in total energy consumption. According to Morris's calculations, per-capita energy consumption (for food, industry, agriculture, and everything else) has increased over 100-fold since hunter-gatherer times. You can find the original research in Morris, I. 2013. The Measure of Civilization: How Social Development Decides the Fate of Nations. Princeton University Press, Princeton, NJ. Thanks to Anthony Faber for bringing this to my attention.

206. Wrangham, R. 2010. Catching Fire: How Cooking Made us Human. Basic Books, New York.

207. Chew More to Retain More Energy, Institute of Food Technologies, 15 July 2013.

p.228. For a great science-based review of how we could improve our cookers, take a look at Nathan Myhrvold's Recipe for a Better Oven by Nathan Myhrvold and W. Wayt Gibbs. IEEE Spectrum. 30 June 2014. Myhrvold is a former Microsoft CTO and long-term cooking enthusiast.

208. Wolke, R. 2008. What Einstein Told His Cook: Kitchen Science Explained. W.W. Norton, New York. McGee, H. 2004. On Food and Cooking: The Science and Lore of the Kitchen. Scribner, New York.

p.229. There are some splendid pictures of Rayleigh-Bénard cells in Ecoulements thermo-convectifs de Rayleigh-Bénard-Marangoni. The text is in French.

p.230. For further fascinating explorations of microwave science, see: Microwave Oven Diagnostics with Indian Snack Food by Lenore Edman, Evil Mad Scientist Labs, 24 August 2011; Ingenious Infrared Microwave Shows Your Food Change Color As It Heats Up by Brent Rose, Gizmodo, 11 February 2015; and, of course, a favourite experiment for kitchen scientists everywhere Measure the speed of light using chocolate, Planet Science, accessed 11 February 2015.

209. Unless, of course, we can find a way of making and digesting nuclear food. Einstein's equation E=mc2 allows a pill weighing 1.5g to produce a theoretical 135 trillion joules of energy.

Chapter 15: Stirring Stuff

210. Dust ranges widely in size, but the smaller it is the more likely we are to breathe it in and the bigger the danger it poses. Typical dust ranges from a few microns (millionths of a metre) in diameter to a hundred microns or more.

211. This is correct in theory; in practice, the atmosphere is much more variable and complex and with phenomena such as inversion layers, the wind speed can actually decrease with height.

212. In theory, if you double the hub height (the height of the point around which the blades spin), the wind speed you tap into is about 10 per cent greater because (as a rule of thumb) wind speed increases with altitude to the power of 1/7, and that produces an increase in power of about a third. In practice, it's more complex. As the turbine rotates, each blade experiences higher wind speeds when it's at its highest point than when it's at its lowest point, so the load on the turbine increases too. Not only that, but bigger and taller turbines weigh more, so they lose energy by doing more work against gravity. See: Assessment of Research Needs for Wind Turbine Rotor Materials Technology, Committee on Assessment of Research Needs for Wind Turbine Rotor Materials Technology, Commission on Engineering and Technical Systems, Division on Engineering and Physical Sciences, National Research Council, 1991.

p.238. (Top) You can see an example of this on my Flickr photo laminar flow on a beach.

213. Swimming in the ocean doesn't really compare with swimming in a pool, for all sorts of reasons. Waves make it more turbulent and quite a bit of your energy is wasted tumbling through them. You might be more inclined to keep your head above the water in a cold, rough sea, so you'll adopt a much less streamlined posture and, consequently, create more drag that makes every stroke harder work. Swimming in a cold sea means you need to wear a wetsuit, gloves, boots and probably a hat – all of which keep you warmer but make it harder to move. Also, salty seawater is very slightly more dense than freshwater and colder water is more dense than warm water; the added density of cold seawater (compared to warm pool-water) doesn't help, but I think it's much less important than the fluid dynamic factors.

214. NASA research in the 1970s found a 24 per cent reduction in drag at 90 km/h (55mph), according to Lamm, M. 1977. Popular Mechanics, August 1977, p.81. Fairings that improve drag around wheels are described in Rahim, S. 2011. Plastic Fairings Could Cut Truck Fuel Use. Scientific American, 10 February 2011.

p.240. You can watch DeMoss and Cahill's experiment laminar flow on YouTube.

215. Einstein, A. 1926. The Cause of the Formation of Meanders in the Courses of Rivers and of the So-Called Baer's Law. Die Naturwissenschaften, 14.

p.246. How skyscrapers can blow you off your feet. I've recently found a couple of good articles about this:

216. In some sufferers the pharynx stays shut, producing a terrifying condition known as obstructive sleep apnoea (OSA). It can be treated with a marvellous device called a continuous positive airway pressure (CPAP) machine, which gently pumps air in through a face mask to keep your pharynx open. See What is CPAP?, National Heart, Lung and Blood Institute, 13 December 2011.

217. Fajdiga, I. 2005. Snoring Imaging – Could Bernoulli Explain It All?, CHEST, Volume 128 Number 2, p.896.

Chapter 16: Water, Water

218. The Water in You, US Geological Survey, accessed 7 March 2014.

219. A brilliant example of wave science that goes by the technical name of thin-film interference.

220. Cavendish is often credited with discovering the composition of water, though James Watt claimed the same thing. There's an interesting discussion in Miller, D. 2004. Discovering Water: James Watt, Henry Cavendish, and the Nineteenth Century 'Water Controversy'. Ashgate Publishing, Farnham.

221. Indoor Water Use in the United States, US Environment Protection Agency (EPA), accessed 16 March 2016.

222. Toilet flushing figures come from the US EPA web page in note 4. Figures for the Siemens iQ300/iQ500 clothes washing machine are 7 litres (2 gallons) of water per kilogram of dry laundry and a load capacity of 5–7kg (11–15lb), giving my estimated 35–50 litres. In March 2016, when I revised these notes, Siemens was quoting 57 liters as the water consumption for the iQ500. See Siemens Washing Machines, Siemens, accessed 16 March 2016.

223. Terry, N. 2011. Energy and Carbon Emissions: The Way We Live Today. UIT Cambridge,Cambridge, p.47.

p.250. There is a typo in the last paragraph: "When the waiter carries that jug of hydrogen dioxide..." This should, of course, read "dihydrogen oxide" (as it says correctly higher up the page).

224. This is a variation on a classic physics exam question known as Caesar's dying breath, in which you're asked to calculate the likelihood of a breath of fresh air containing at least one molecule from the last lungful of the doomed Roman dictator. No calculation is required: all you really need to know is that there are more molecules in a glass of water (600,000,000,000,000,000,000,000 molecules in every 18g or 0.6oz of it, which is called a mole) than there are glasses of water on our planet (or in a lungful of air than there are lungfuls in Earth's atmosphere). Like the water, this amusing factoid has been endlessly recycled. You'll find it in Dawkins, R. 2008. The God Delusion. Houghton Mifflin Harcourt, New York, p.410, which itself recycles an earlier discussion by Lewis Wolpert noting that 'there are many more molecules in a glass of water than there are glasses of water in the sea'. I'm simply doing my duty by recycling it again.

225. The specific heat capacity of iron is about 0.47 kilojoules per kilogram per degree, or roughly nine times less than the specific heat capacity of water. That means it takes nine times more energy to heat up 1kg (2.2lb) of water by a certain temperature than 1kg of iron (or that you get nine times the temperature rise, if you prefer).

226. For the sake of a simple explanation, I've described home central heating as though it's a 'series' circuit with each radiator fed in turn from the boiler – which is how older systems were often fitted. In practice, newer systems have much more efficiently designed 'parallel' (trunk and branch) circuits with the water splitting up and taking different paths to different radiators instead of visiting each radiator in turn.

227. Operating temperatures for nichrome (nickel, chromium alloy) heating elements are about 750°C (1380°F).

228. Lifting the Lid on Computer Filth, BBC News, 12 March 2004; and Keyboards 'Dirtier Than a Toilet, BBC News, 1 May 2008.

p.260. "Plumbers sometimes need...": If you'd rather not try this and risk a drenching at home, watch an example on YouTube here: How to Quickly Get Water Out of Your Toilet Bowl . You can actually drain the bowl completely if you're lucky!

229. See note 223.

230. World Health Organization. 2003.Guidelines for Safe Recreational Water Environments Volume 1: Coastal and Fresh Waters. WHO, Geneva, p.45.

231. Botcharova, M. 2013. A Gripping Tale: Scientists Claim to Have Discovered Why Skin Wrinkles in Water, The Guardian, 10 January 2013.

232. Dyson Airblade V Technical Specification,, accessed 16 March 2016. There's also a video on Dyson's YouTube channel, here: Dyson Airblade Mk2 hand dryer explained.

Chapter 17: Stain Games

233. Bakalar, N. 2003. Where the Germs Are: A Scientific Safari. John Wiley, New York, p.54.

234. Handwashing: Why are the British so Bad at Washing their Hands?, BBC News, 15 October 2012.

235. Most People Washing their Hands, Guinness World Records, accessed 25 February 2014.

236. Bakalar, N. 2003. Where the Germs Are: A Scientific Safari. John Wiley, New York, p.53.

237. US figures for 1997 quoted in Carpenter, R. 1999. 'Laundry Detergents in the Americas: Change and Innovation as the Drivers for Growth.' In Cahn, A. (ed) 1999. Proceedings of the 4th World Conference on Detergents: Strategies for the 21st Century. AOCS Press, Urbana, IL. European figure quoted in Garratt, B. 2010. The Fish Rots From The Head. Profile Books, London.

238. Wilson, E. 1992. The Diversity of Life. Harvard University Press, Cambridge, MA, p.142.

239. Estimates vary depending on the breed, but 50 million seems a good, average figure, according to Cook, J. 1984. Handbook of Textile Fibres: Volume 1: Natural Fibres (Fifth Edition). Woodhead Publishing, Cambridge, UK. p.89.

240. The idea of a 'universal solvent' dates back to the alchemists, who sought (in vain) a substance they termed the Alkahest that would dissolve everything else. Water is still the closest thing we have. See Chapter 1. Introduction. In Reichardt, C. & Welton, T. 2011. Solvents and Solvent Effects in Organic Chemistry. Wiley-VCH, Weinheim.

241. Assuming each home uses 5,000 litres per year, you'd need 200 homes to use 1 million litres. A few hundred homes would give you several million litres, which is roughly the volume of an Olympic-sized swimming pool. That might not sound too bad, but imagine it scaled up for an entire town or city and you can quickly see the problem.

242. Say the diameter of the drum is 55cm (22in), giving a radius of 0.225m (11in) and a circumference of 1.41m (56in). If the drum rotates at 1000rpm, the circumference moves 1,410m (4,624ft) per minute = 85km/h (approximately 50mph).

p.260. Concrete in washing machines: You'll find some pictures here.

243. MacKay, D. 2008. Sustainable Energy Without the Hot Air. UIT Press, Cambridge, p.54.

244. Most people would say that the water is removed by centrifugal (literally, 'centre-fleeing') force, but science teachers don't like that term and would prefer to explain things a different way. They'd say the washing machine drum provides centripetal (literally 'centre-seeking') force that keeps the clothes moving around in a circle. Because there are holes in the drum, there's nothing to give the water inside the clothes centripetal force so it moves in a straight line, separates from the clothes, and escapes.

245. I tested this out on a cold December day a few years ago, when there was snow on the ground across much of the UK. I weighed a load of laundry before I hung it outside, left it blowing in the bitter easterly wind for about five hours, then brought it in and weighed it again. After washing and before any drying, it weighed 5kg (11lb); after outside drying, it clocked in at just 3.5kg (7.7lb). Then I dried it fully, indoors, and weighed it a third time. This time, the scales registered 3kg (6.6lb). Assuming that was the dry weight, you can see that the outdoor drying removed about 75 per cent of the water.

246. Cost Benefits of the Xeros Technology, Xeros Ltd, accessed 11 January 2015.

247. The US EPA has a brief but useful summary of the ingredients in a typical laundry detergent and their environmental impacts. See Key Characteristics of Laundry Detergent Ingredients, US Environmental Protection Agency, May 1999.

248. Endocrine disruptors, present in many chemicals (not just detergents), cause a significant proportion of fish to change sex. See Tackling Fish Endocrine Disruption, US Geological Survey Toxic Substances Hydrology Program, 4 August 2009, quotes 18–22 per cent; Pollution changes sex of fish, BBC News, 10 July 2004, quotes 'a third of male fish'; Report: Pollutants in D.C. area drinking water, The Washington Times, 12 November 2009, quotes '80 per cent of fish' showing sex changes in Washington's Potomac River. For a much fuller discussion, see Kime, D. 1998. Endocrine Disruption in Fish. Kluwer Academic Publishers, Norwell, MA.

Chapter 18: Dressing to Impress

249. Simpson, W. S. & Crawshaw, G. H. (Eds.) 2002. Wool: Science and Technology. Woodhead Publishing, Cambridge. p.303.

250. This is called Newton's Law of Cooling. One of its most interesting applications is in helping crime scene pathologists to work out an approximate time of death from the temperature of a dead body.

251. Plowman, S. & Smith, D. 2007. Exercise Physiology for Health, Fitness, and Performance. Lippincott Williams & Wilkins, Baltimore, MD, p.418.

252. The formal scientific name for this is heat of sorption. For much more about the science of wool, see the excellent booklet: Leeder, J. 1984 Wool: Nature's Wonder Fibre. Australasian Textiles Publishers, Ocean Grove, Victoria.

253. Stainless steel comes in at 170–1000 MPa; wool manages 70–115 MPa. Data from Ashby, M. 2012. Materials and the Environment: Eco-informed Material Choice. Elsevier, Waltham, MA. p.466/592.

254. Get to Know the GORE-TEX® Membrane, GORE-TEX, accessed 16 March 2016. The GORE-TEX website quotes the membrane as being 20,000 times smaller than a drop of water, but 700 times bigger than a molecule of moisture vapour. You might wonder how these figures square with my estimate of a water drop having millions of trillions of molecules? As far as we can tell, Gore's scientists arrived at these figures by comparing linear dimensions rather than areas or volumes. It's important to remember that water drops vary hugely in size. We quickly ran the numbers for this book and arrived at a range of estimates from about 130 trillion up to several billion trillion water molecules in a drop.

255. Apart from discussing the fascinating similarities between structural materials and clothes, James Gordon shows how metal sheets can shear along a diagonal in exactly the same way as dresses and tablecloths, which is why you sometimes see a 'crease' in the fuselage of a jet plane or a helicopter. Gordon, J. 1978. Structures. Penguin, London. p.248–259. As Gordon notes, the bias cut was pioneered in the 1920s by French fashion designer Madeleine Vionnet.

256. Chapter 10: Energy Expenditure During Walking, Jogging, Running, and Swimming. In McArdle, W. et al. 2010. Exercise Physiology: Nutrition, Energy, and Human Performance. Lippincott Williams & Wilkins, Baltimore, MD, p.210.

257. For a good introduction, see Fields, K. 2009. Opening the Gait. Training & Conditioning, 25 May June 2009.

258.Chapter 10: Energy Expenditure During Walking, Jogging, Running, and Swimming. In McArdle, W. et al. 2010.Exercise Physiology: Nutrition, Energy, and Human Performance. Lippincott Williams & Wilkins, Baltimore, MD, p.210.

Further Reading

Some extra suggestions that have occurred to me..

  • Stuff Matters by Mark Miodownik (Penguin, 2014). Unfortunately, I didn't get around to reading this until my own book had gone to press, so it's a notable omission from the Further Reading. But I heartily recommend it as a friendly introduction to the fascinating world of materials science, which is a far more interesting subject than its practitioners make it seem. Stuff Matters is much more accessible than James Gordon's books (more familiar, less scientific)—and you can graduate to those afterwards if this whets your appetite.
  • Science and the City: The Mechanics Behind the Metropolis by Laurie Winkless (Bloomsbury, 2016). Another great little introduction to materials science, starting from the things we find in modern cities. A simple and very engaging read that I'm sure will appeal to people who like my own books.