Association Football, or 'soccer', is the major spectator sport in the world, as well the most widely-played. When space becomes a practical living space for the public to visit, it is likely that there will be a desire for sporting activities of some kind, and that these will be modelled on familiar sports. Since football or 'soccer' is the most familiar of all, it will be a strong candidate for adaptation to the micro-gravity conditions found in orbit. In recent years there has been growing interest in the feasibility of passenger space travel, or 'space tourism'; Nasa has acknowledged that it is likely that it will grow to become the largest business in space [1], the Japanese Rocket Society ( JRS) has published [2]; and a library of research papers on the subject of space tourism is available on the website http://www.spacefuture.com. As one possibility, the building and operation of 'sports centers' in space has received some attention by staff in the Japanese construction company, Hazama [3,4]. With the start of commercial tourism in space with the flight of Dennis Tito in April 2001, and the announcement in September 2001 of Mircorp Inc's plan to build a small space station for use for tourist accommodation, it seems timely to start to consider how football may be adapted for playing in orbit.

The aims of this paper are to consider what form such adaptation might take, staying as close as possible to the spirit and gameplay of the original game, and taking into account the restrictions and possibilities of a zero-gravity environment. These will fall into two parts, the first considering the practical aspects of adaptation, the second considering how the rules need to be altered to create a fair game, and one that will be entertaining to both players and spectators. As far as possible, the game will be considered without additional artificial aids for players in playing the game; the game will be played by using any part of the body except the arms and hands to control a single ball; and the game will use similar procedures to those in terrestrial, or Earth-gravity soccer wherever possible.

That is, it is likely that various forms of football will evolve in zero gravity, as they have on Earth as rugby, American football, Australian rules football, and so on. However, in view of the unique popularity of 'Association Football' we restrict ourselves to considering how close a game in zero-G may come to the terrestrial game of 'soccer' (referred to simply as 'football' in the rest of this paper).

Practical considerations include such things as the manoeuvring abilities of players, what limits there are on the size of playing field, what form that playing field might take, safety of the players, and how to referee the match - in other words, the physical and physiological parameters in which the game is played.

Sporting considerations include devising an appropriate scoring mechanism, how many players to a team, how much time to allow for the match, what constitutes unfair behaviour, and how fouls should be penalised - in other words, matters of fairness, competition and engagement of spectators.

The most important problem in adapting the game of soccer to zero-gravity is simply working out what motions are available to a player. For this, Newtonian motion is sufficient to describe the movements of players in terms of direction, and leads to the conclusion that once a player has launched himself from a fixed point, i.e. the side of the playing field, he will have little opportunity to change their direction until reaching the other side of the playing field.

The main exception to this will be through impact with other players; by working together, two players on a team will be able to achieve changes in direction while in open volume. Further considerations of this kind of teamwork are verging on the sporting considerations, and some implications of these options are considered in the second section.

Of more interest will be the capacities of the players to induce rotation around various axes through the centre of mass of their own bodies. This is an important aspect of how zero-gravity soccer could be played, as the potential for rotation of this kind has implications both for lining up to kick the ball, and for dealing with the forces involved with making the kick.

Footage of astronauts and cosmonauts in orbit generally shows little of this, as they tend to be in relatively small, confined areas. However, they do give some idea. More interesting in this respect is recent footage of dancers under free-fall inside an aeroplane, shown by the Open University [5], This includes a few short sequences in which dancers are clearly shown inducing rotation without the aid of other people or objects. The amount of change and the rate of change is dramatic and shows that zero-gravity soccer players will certainly have a great deal of scope for movement in this way.

What we find, therefore, is that players have a great deal of control in terms of their attitude (which way they face relative to direction of travel) but very little control over their direction having left the side of the playing field.

The limits on how large a playing area should be are governed primarily by the distances

that can be expected for a player to travel after pushing off from a surface. Unless the players are expected to wear bulky pressure suits and breathing apparatus, the playing field will need to be filled with breathable air, so this discussion is based on the aerodynamic properties of the human body in 1 atmospheric pressure, and the potential kick-off speeds that a body might attain. An important question is, how quickly will aerodynamic drag reduce the body's speed to zero or thereabouts, and thereby leave the player stranded?

Firstly, how fast can a human being kick-off from a flat surface? In Earth gravity conditions a simple measure of the fastest kick-offs is the athletic event of the high jump. In this men regularly clear the 2 metre mark (this appears to be the limit of the range for female jumpers). Kick-off speed is given by:

V²₀ = 2Gh

Where V is take-off speed and h is height.

2 metres gives √ 2×10×2 = 6.3 m/s

It should be pointed out at this stage that the high jumpers using the technique known as the Fosbury Flop' effectively cheat in that their centre of gravity passes below the height measured by the bar; the error this induces is of the order of 10%. However, comparison with gymnasts featured at the top level suggest from a naked-eye estimate that the leaps that these competitors are able to perform come close to the heights achieved by high-jumpers. Since it is not to be expected that zero-gravity sportsmen will necessarily be as powerful as high jumpers or gymnasts, needing to focus some areas of training on other aspects of physical prowess, an estimate for maximum take-off speed of 5 m/s seems reasonable as a working hypothesis. (However, it should be noted that gymnasts use a springboard, and it would be possible to use a similarly elastic surface for the walls of an orbital football stadium.) This leads to the question: How quickly would a player, having launched at 5 m/s, reach 0 m/s?

The rate of deceleration speed depends on the player's drag, on their surface area, and on the square of their speed. Players will be able to alter their drag by approximately a factor of ten by altering then: attitude, in order either to slow down as fast as possible or to travel as far as possible. Following the analysis of flying in zero gravity published in [6], we estimate that, starting with a speed of 5 m/s they will lose speed at between 0.3 m/s and 0.03 m/s, that is between 6% and 0.6% per second. Even at the higher rate of deceleration it will take 1 minute to slow to 1 m/s, during which time the player will travel more than 100 metres. At minimum deceleration a player would travel considerably further. Thus a playing field of about 100 metres would allow players to travel the full length unaided, even allowing for different take-off angles. Conversely, there is no real limitation on how small the playing field might be, save to ensure that there is sufficient room for the players themselves within the field of play.

Having covered size, the next question is the shape and properties of the enclosure. It is intuitive that a genuinely 3-dimensional game of soccer such as is being proposed here will need to have walls to mark off the playing field, since otherwise determining what is or isn't in play will become virtually impossible. It is also necessary to have some fixed points from which the players can kick off, and the most intuitive form for that to take would be in the shape of the barriers denoting the enclosure in which play takes place.

Initially, it is possible to envisage a box of some kind, which in analogy to the rectangular pitch of 2-dimensional soccer would be cuboid in shape. However, other shapes are possible, perhaps the simplest to build practically being elliptical in a cross-section taken lengthways. However, the main effect that these would have would be on tactics of the game, and that is an issue that belongs more in the sporting considerations.

The composition of the walls is of greater importance. The first matter is that the players will clearly need to be able to hold onto the walls at times, so a means to make that possible will be required. Positioning for each jump into the volume would be necessary as well, so moving about the walls will need to be possible. Therefore, some form of grid of bars or cables could provide the required grips. Alternatively, one can imagine a game in which players run along the walls as much as they float in zero G. For this they would need equipment such as 'velcro'-soled shoes in order to have traction on the wall's surface. This might help to preserve the 'feel' of terrestrial football, with plenty of action taking place on the 'walls' of the pitch. One possible drawback of this is that more force will go into pulling clear of the walls, and therefore there will be less speed available to players once in flight.

The walls will also need to be flexible so that the force of a player's impact will be dissipated to some extent, to reduce the likelihood of injury. This will mean that there will need to be some elastic properties, and that the frequency of its vibration be low enough to give players the chance to gain a handhold if they need to, and so as not to break a player's speed too quickly (which could cause injury). However, if the wall has too much elasticity, players will find that too much of their energy will go into the wall and not enough into launching themselves (although good timing might be able to use the oscillations of a flexible wall to gain extra impetus). Another drawback to having too much elasticity would be that the impact of a player elsewhere on the same wall might destroy the momentum that a launching player would otherwise gain, which would have serious implications for gameplay, and leading to very different tactics.

Of great concern is the question of what injuries players might sustain, and how might the more serious of these be prevented. During play it is inevitable that collisions between players will occur, and similarly collisions as a player meets the side of the playing area. Although as mentioned above, the walls of the playing field would need to be designed to consider these problems, approaching the wall with an unfortunate attitude could still cause serious problems. For example, if the side of the head were to make first contact against the wall, the neck could be put under a great deal of stress.

There are two main ways in which injuries might be gained, colliding with walls and colliding with players. In colliding with a wall, the main concerns are about the angle of approach, as in the example given, and the ability to fend oneself off using the arms or legs as shock-absorbers. Under normal conditions, it would be expected that a player could control their attitude so as to maximise their ability to control the impact; however, since collisions with other players might disrupt their control, and on a fairly frequent basis, then it becomes a serious concern to prevent injuries. The other concern regards the possibility of serious injury caused by having an ankle, leg or wrist caught in the mesh from which the walls are formed. The clear solution to this is that the mesh size must be small enough to prevent this possibility. Beyond this, the main worry from collisions with the wall has to be the possibility of serious neck injury. The amount of protection that can be worn without seriously impeding gameplay, in that for effective team sports awareness of the surroundings is vital, is a matter of conjecture for the moment, and it is likely that neck injuries would constitute one of the more common zero-gravity soccer injuries.

Whether intentional or not, it is inevitable that when there is a single target for the attentions of players, i.e. the ball, then there will be collisions between players contesting control of the ball. The ways in which injuries might be inflicted when players collide with one another are almost limitless in scope, so that trying to protect against them all would be pointless. One of the most serious possibilities is whiplash caused by being brought suddenly to a stop by such an incident - as mentioned above, zero-gravity footballers may need some form of neck protection. Beyond this, the main concern should be to give protection to important areas of the body without impeding the manoeuvrability of the players. Helmets akin to cycling helmets, a chest-guard, shin-pads, lower-arm guards, and a box to protect the genitals should cover the essentials and the most common impacts. A gum shield, elbow and knee pads may also be worn. As in the terrestrial game, it is the referee's duty to ensure that a player is properly attired before he is allowed to play; this includes the regulation protective equipment.

One other concern regarding injuries is that of bleeding. It is already required in the terrestrial game that a player who is losing blood must leave the field until the bleeding has been stopped [7]. This is even more important when there is no gravity to control the motion of the fluid, and the game itself might have to be halted for quite some time while attempts are made to clear the area of stray droplets.

The final practical consideration is that of how the referee will observe the game. In the terrestrial game, there is no difficulty with having the referee on the field and moving with the same degree of motion as the players. The referee is assisted by two officials on either side of the playing field whose job it is to signal offences that the referee is not in a position to adjudicate upon. In the case of zero-gravity soccer, however, it seems likely to be the case that the referee's assistants will be more important than a man observing from within the field of play.

If a referee in the playing field were to move using the same manoeuvrability as the players, then his reaction times would be similar to those of the players, that is, he will have to wait until he reaches a wall before he is able to change direction to keep up with the play. For the adjudicator to be limited in this way places severe restrictions on how fairly the game might be judged, since it will not be possible for him to follow events closely enough to make accurate, informed decisions. For this, he would need to rely on observers outside the playing field.

Alternatively, the referee within the playing field could have an extra form of propulsion, but what could this be? The most effective would be propeller of some kind setting up extra air-currents. But this could end up having a much wider effect on the game in very direct ways (not merely through the 'butterfly effect') by deflecting the ball in its flight, and conceivably deflecting the players as well (depending upon how powerful the referee's motors are). Therefore, it seems that refereeing the game will have to take place entirely from the sidelines. Given that incidents will occur in which it is hard to judge without a closer view, it is also likely that video-refereeing will play a much more important part than it does in most terrestrial sports (though it should be noted that it is playing an ever-increasing role on Earth).

Many of the areas covered in the practical considerations also have implications in the sporting considerations section, and the conclusions already presented will be re-addressed in the light of the requirements for a fair and exciting game.

The game of football is regulated by a set of 17 laws that cover all aspects of the game, which are outlined by the international football authority, FIFA, and overseen in England by the Football Association [7]. Their 1998 publication of the laws shall form the basis for discussing how football under zero-gravity might retain the qualities that draw support for the terrestrial form of the game. These laws shall be discussed not in numerical order, but in order of importance to developing a form of football for zero-gravity.

The first law deals with the dimensions and markings of the field of play. Under the practical considerations, a practical limit on the width of the pitch was discovered to be 100 metres, allowing for variations in angle of take-off. However, from a sporting perspective, this is unsatisfactory since a player traversing the distance of 100 metres will take 1 minute of time, during which time they will be largely unable to react to the events around them. This will result in a slow and uninteresting game. A more important consideration for devising a playable sport, is the reaction time afforded to players in flight. In zero-gravity, this can never be as quick as it can be on the ground, but by making the field of play small enough, reaction times can be brought down to a few seconds.

A limitation on how small the field of play can be is given not only by the size of the players, but also the fact that they need to be able to perform several actions while in flight. These are, launching themselves, checking where the ball is, positioning themselves to kick the ball, then positioning themselves ready for the impact on the opposite wall. In addition to these, they will need to be able to react to their immediate surroundings to avoid breaking one of the laws (e.g. handball).

The reaction times for each stage of a player's flight might be estimated as 1 second each. That would put a bare minimum of 4 seconds of flight as the lowest acceptable. Conversely, a 10 second flight time would be the absolute maximum before reaction to the game's events becomes too slow to keep up. Using the estimate of 5m/s take-off speed, and regarding the deceleration over 4 seconds to be negligible (in fact this will be of the order of 1 metre per second lost by the time the player reaches the opposite wall) then a minimum size of width 20 metres is acceptable. The speed actually lost in flight, will then mean that reaction times are slightly slower, giving more time for the essential manoeuvring that the players must achieve during flight. Towards the longer duration, 8 seconds of flight leads to approximately 2 m/s of speed lost. Since the rate at which deceleration has changed is very small, the average speed is close to 4m/s. That gives rise to a maximum size of width 32 metres. For the sake of convenience, this can be rounded to 30 metres. Assuming a square cross-section, two of the three dimensions have been determined.

The length of the pitch is a different concern, the main requirement being that it afford the opportunity for good tactical play. Too short, and there is little need for tactics as every player has equal chance of scoring a goal; too long and advancing the ball up the field becomes almost impossible. The dimensions given for the 2-dimensional pitch suggest that any ratio of side to length between 1:1 and 1:3 is acceptable. For the zero-gravity game, the tunnel shape of the pitch suggests that the figure directly between these two extremes is more desirable, that is a ratio of 2:3. That means that the length should be 30 metres at the smallest cross-section, ranging to 60 metres at the largest cross-section. The difference between these two figures is very notable: as has been seen, 30 metres can be travelled in 8 seconds and is a reasonable flight for a player to make. However, 60 metres will take a lot longer, of the order of 20 seconds. The different sizes would require very different tactics for teams to play effectively in them, but since there is no analytical way to choose between them, this issue can only be decided by practical experience of playing in zero-gravity.

Markings of the pitch shall be discussed where they become relevant to the discussion of other rules, since their reference to these rules will make them very different to the traditional markings.

Under law 10 of the laws of football, scoring is done by passing the whole of the ball into the opponent's goal. The goal's dimensions are given as 7.32 metres wide by 2.44 metres tall (or the equivalent in Imperial measurements). This gives a cross-sectional area of approximately 18 metres . In zero-gravity, there is no difference between height and width, so it seems appropriate to adjust these measurements to give a square of equivalent cross-sectional area. This gives a square of side 4.24 metres. Since a man at full stretch will not be able to stay in one place to protect the entire area, this provides the requirement of goal-keeping skill that is an important part of the game of football.

One concern regarding goal-keeping is the need for the goalkeeper to remain in a narrow region of space between the edges of the goal opening. If the goalkeeper should miss the opposite edge for any reason, he would end up drifting far from the goal, and be unable to do his job of guarding the goal. A possible solution would be to use an elasticated cord attached to the goalkeeper's back, with the other end attached to the back of the goal. While the goalkeeper is in position between the edges of the goal, the cord should be slack, but if the goalkeeper moves too far from position, it would reverse his momentum, allowing him to return to the goal opening. This also means that the goalkeeper can deliberately come forwards from the goal to catch the ball while it is being passed to a close attacker.

The other main issue is one of momentum: if the ball is moving sufficiently quickly, then when the goalkeeper catches it, it will carry the goalkeeper backwards into the goal, allowing a goal to be scored even though the goalkeeper was skilful enough to stop the shot. However, this is not as big a problem as it might at first seem. The ball weighs approximately 0.4 kg and is in the terrestrial game propelled at a rough maximum of 30 m/s [8]. For reasons that will be explored in more detail below, it is likely that the ball in zero-gravity football will have a smaller mass and be moving slower. A heavily built man weighs perhaps 80kg. The relevant calculation shows that the speed imparted to such a goalkeeper by the impact of a terrestrial football would be of the order of 0.15 m/s which should provide plenty of time to return the ball to play, with thought for where his team-mates are placed. Even if the effect is of the ball rebounding from the goalkeeper, the goalkeeper should be able to reach one side or the other of the goal before he is carried too far back into the goal to be able to defend it properly.

The offside rule is one of the most frequently discussed laws of football, and its purpose is to prevent a team from gaining an unfair advantage by placing a player so that there are no defenders able to impede his progress towards goal. This does not prevent an attacking player from being in such a position, but depends on other events on the field.

The offside position is described as follows:

A player is in an offside position if: he is nearer to Ms opponent's goal line than both the ball and the second last opponent. A player is not in an offside position if: he is in his own half of the field of play or he is level with the second last opponent or he is level with the last two opponents.

Offside is called when an attacker is in an offside position when the ball is passed forwards by one of his team-mates. Offside might not be called if the referee judges that the player was not affecting the course of play at that time. It is assumed that the goalkeeper is the last defender, although this need not be the case in terrestrial football. In zero-gravity football, as seen in the previous section, it will be necessary to ensure that the goalkeeper is the last defender.

The key issue with adapting the offside law to zero-gravity is the fact that any player away from the walls has no way of changing his direction of movement; that means that the provision of the offside position must be extended to include this fact; this means that an offside position must be described as '...if he is nearer to his opponent's goal wall than the second last opponent in contact with a wall'. At this point, it seems that it will be very easy for defenders to have attackers called offside, by means of kicking away from walls before the pass is made by the opponent. In actual fact, this need not be the case. What will be necessary instead is to strengthen the element that allows for discretion on the part of the referee. Since attackers as well as defenders cannot change direction while away from the walls, if a pass forwards is made in such a way that it is impossible for the player in an offside position to reach it, then that player cannot be playing an active part in the game at that point and offside should not be called. Only a player in contact with the wall, or who is the direct recipient of a pass forwards, should be penalised as being offside.

The laws of football provide a list of fouls which are punishable in various ways.

The list is as follows:

1. kicking or attempting to kick opponent
2. tripping or attempting to trip opponent
3. jumping at opponent
4. charging opponent
5. striking or attempting to strike opponent
6. pushing opponent

1. tackling opponent, making contact with opponent before ball
2. holding opponent
3. spitting at opponent
4. handling the ball deliberately, except for the goalkeeper
5. playing in a dangerous manner
6. impeding the progress of an opponent

Some of these are implicitly only defined for the terrestrial game, by reference to actions dependent upon a gravity environment (e.g. tripping an opponent). Others need no interpretation to see that they are relevant to zero-gravity football. These are: a, e, f, h, i, j, k. Items c and d can be interpreted as the act of launching from the wall in a deliberate attempt to collide with an opponent.

Items b, g and 1 do not appear to have meaningful interpretations into zero-gravity.

Of particular note are the implications of items a, f and i. In the zero-gravity game, pushing an opponent is much more serious than simply leaving them off-balance: their entire flight through the field of play will be affected, and the same goes for kicking or striking an opponent. This can gain a team a huge temporary advantage, quite apart from the possibility of injury to the opponent. The possibility of spheres of saliva floating in zero-gravity is not a pleasant one, and certainly worse than the offence caused on Earth. A directed ball of spit could be quite nasty.

In addition to the above fouls, the following offences are described in the laws of football as resulting in cautions or a player being told to leave the field:

Caution:

1. unsporting behaviour
2. dissent (to the referee) by word or action
3. persistent infringement of the Laws of the Game
4. delaying the start of play
5. failing to respect the required distance when play is restarted by corner kick or free kick
6. enters or re-enters the field of play without the referee's permission
7. deliberately leaves the field of play without the referee's permission

Sending Off:

1. serious foul play
2. violent conduct
3. spitting at an opponent or any other person
4. denying an obvious goal-scoring opportunity to an opponent moving towards the offending player's goal by a foul punishable by a free kick or a penalty kick
5. using offensive, insulting or abusive language
6. receiving a second caution in the same match

These points are much more to do with the spirit in which the game is played, rather than the mechanics of how the game is played; therefore, there is no adaptation necessary save to observe that the zero-gravity field of play will be completely enclosed while play is in progress, so that entering or leaving the field of play without the referee's permission would be impossible to achieve.

Free kicks and penalty kicks are the ways in which fouls are punished in the terrestrial game of football. They work by handing control of the ball to the opposing team. A free kick is taken from the point at which a foul was committed. If a foul is committed within a certain distance from the goal, marked by a rectangle called the penalty area, then a penalty kick is awarded, which gives a clear goal-scoring opportunity to the opposing side.

Both free kicks and penalty kicks require the precise placing of the ball; in the case of penalty kicks, a spot is marked on the playing field, from which the kick is taken. However, the reliable placing of the ball in 3-d under zero-gravity conditions is unlikely to be achievable, especially in the case of the free kick, when it must be indicated by the referee where the ball should be placed. Since it is very difficult (as explained in the physical description of the playing field under Practical Considerations) to unequivocally mark out boundaries of areas in 3-d, marking the spot and the penalty area are unfeasible. This means that alternative procedures need to be devised to perform the same functions as closely as possible.

The purpose of the free kick is to transfer possession from one team to the other, as close to the point of the foul as possible. The only way this can be achieved in zero-gravity football is to allow a player of the team to which the free kick has been awarded, to throw or kick the ball from the point on the walls that is closest to the point of the foul. The way in which the free kick is taken is that all players on both sides must remain on the walls until the throw is taken. The ball is back in play as soon as another player touches it.

The purpose of a penalty kick is to penalise a team for denying what is assumed to be a latent scoring opportunity, because the ball is close to the goal of the offending team. The team is penalised by turning a latent scoring opportunity into a clear one.

In terms of clarity, the only way in which a penalty area can be marked in the zero-gravity game is by having a single line running around four walls, at a fixed distance from the goal-wall. On the goal-side of this line is the penalty area (or 'penalty volume' since the game is in 3-d). One method by which a penalty procedure could take place is as follows:

A player of the attacking team (against which the foul was committed) is positioned on the line at a position of his or his team's choosing, and holding the ball. All other players are positioned outside the penalty volume, and on the walls. On a signal from the referee, this player holding the ball throws it into the penalty volume. Only one other player of the attacking team is allowed to move at this point, in order to strike the ball thrown by his team-mate. No defenders save the goalkeeper are allowed to move at this point. Once the second attacker or the goalkeeper has made contact with the ball, then the ball is in play and all players may move. A goal cannot be scored until the ball is in play.

This preserves the important aspects of the terrestrial penalty kick procedure, but it does increase the complexity of what is required by the attacking team. However, this system does allow for some tactical thinking that is denied a team in the terrestrial game.

The corner kick and throw-in procedures are for use when the ball leaves the field of play; however, as previously explained, the zero-gravity field of play will be fully enclosed, so that the ball will bounce off the boundaries rather than crossing them, therefore removing the need for these procedures.

The remaining points to discuss are the number of players, the duration of the match, how play should be started and re-started, and the properties of the ball.

Very little can be said on the number of players and the duration of the match, without seeing in practice the effects of varying these quantities. The number of players might be estimated in many ways, but the best estimate so far is that 10 players plus a goalkeeper should make a good game. This has the desirable factor that it is the same number as in a terrestrial football team. The duration of the match may possibly be found to be best if it is shorter, since there will be much more work with the arms and players may tire more quickly than they do in the terrestrial game. The duration of the game might be between 60 minutes and 90 minutes (the latter figure being the current duration of the terrestrial game).

Starting the game is achieved by nominating one side or the other to take the first kick, and then requiring that they kick the ball from the centre of the field of play, backwards into their own half of the field. While positioning the ball in the centre of the zero-gravity football field of play would be unfeasible, it is certainly possible to take the kick from the centre of any one of the four long sides, and have the kicker kick the ball into their own half. Since football is played in two halves, the kick off is taken by one side in the first half and by the other in the second half.

Sometimes it is necessary to re-start the game after a player has been injured and play has been stopped so that he can be removed from the field of play. In the terrestrial game, this achieved by the dropped ball procedure, which gives either side equal chance of winning control of the ball. It is hard to imagine a comparative procedure in zero-gravity football, so the best option is to allow a throw to be taken from as close as possible to the point of last contact with the ball, by the team whose player last touched the ball.

Finally, the physical properties of the ball in the terrestrial game are unsuitable for the size of stadium outlines above. As described under the section on scoring, the terrestrial football can be struck at up to 30 m/s. At this speed, a ball struck at an angle of 30 degrees travels most of the length of a pitch, which is a minimum of 90 metres. In the zero-gravity game, this will mean that it retains a great deal of its speed even at the longest suggested dimension for the zero-gravity field of play. This means that greater deceleration must be built into the ball's properties, for example, by increasing the drag.

There are many ways in which sports might be developed with the aim of playing in zero gravity conditions. By looking at the particular problems presented by adapting a familiar sport on Earth to be played in these conditions, some of the issues involved in creating such a sport can be assessed. Certainly it is feasible to adapt the game of football to zero-gravity conditions so that sporting interest can be maintained. However, other sports may adapt more naturally to zero-gravity and therefore prove more popular with players and spectators.

1. D O'Neil et al, 1998, "General Public Space Travel and Tourism Volume 1 - Executive Summary", Nasa/Space Transportation Association, Nasa report no. NP-1998-03-11-MSFC.
2. P Collins and K Isozaki, 1996, "JRS Research Activities for Space Tourism", Advances in the Astronautical Sciences, AAS, pp 521- 528.
3. P Collins et al, 1994, "Zero Gravity Sports Centers", Proceedings of Space 94, ASCE, Voll.pp 504-513.
4. P Collins et al, 2000, "Orbital Sports Stadium", Proceedings of Space 2000, ASCE, pp 604-616.
5. Anon, 2001, "Blue Sky", video produced by Uden Associates for the Open University.
6. P Collins and J Graham, 1994, " Human flapping-wing flight under reduced gravity", Aeronautical Journal, May 1994, pp 177 - 184.
7. FIFA & The Football Association, 1998, "Laws of Association Football (Guide for players and referees 1998-99)", Pan Books
8. Anon, 1992-2001, Sky Sports Football coverage, Sky Sports