What Is A Garden Maze Called

Garden mazes

Intro --- Pre-maze --- Cretan --- Roman --- Chartres --- Turf --- Garden --- Other --- Design --- Lay out --- Designer --- Games

O, what a tangled web we weave, when first we practise to deceive! Marmion by Sir Walter Scott (1771-1832)

Many of the mazes on this webpage are unicursal mazes, without choices or branches, and they are often flat. However, most people think of mazes as places where you can get lost, with choices and high walls so you can't see where you're going. These are the garden mazes or hedge mazes.

So how did multi-choice mazes come about? The photo on the left shows the garden of Provand's Lordship in Glasgow, centre, a parterre in the grounds of Kenilworth Castle, and right, a garden at Bridge End in Saffron Walden. Parterres and knot gardens were decorative designs, edged with small hedges, often box . Between the hedges was gravel, or flowers, or herbs, or grass. In some of these gardens itis possible to wander between the hedges. You can imagine that the hedges were made higher, which meant that you couldn't grow flowers or herbs any more. Finally, you have a full-size maze, where you can get lost. Daniel Loris's mazes show the stage where they were developing this idea.

Types of puzzle mazes

A name for a non-unicursal mazes is a puzzle maze. There are two types, branching and island mazes. Branching mazes are more old-fashioned. Once you know the trick, you can always find your way to the centre of a branching maze, but not necessarily by the quickest route. Because of this, modern mazes tend to be island mazes. However, when the first garden mazes were made, they hadn't quite worked out the principle of branching mazes, so they were often island mazes, I suspect by mistake!

A branching maze has just one path that leads to the end (which is usually the centre). There are other paths leading off the main path, and they may have further paths leading off them, but all these subsidiary paths are dead-ends. So if you make a mistake, at some point you have to turn round and come back again. This is why it's called a branching maze. If you untangled all the paths and laid them out in straight lines, you would get a tree-like structure. The trick to solve a branching maze depends on this. At the entrance, you put a hand on the nearest wall of the maze (it doesn't matter which hand). As you walk, you keep your hand gently touching the wall of the maze. Every time that wall is interrupted by a gap caused by a path branching off, turn down that path, keeping you hand always on the wall. When you meet a dead-end, move your hand round the end of the path, always keeping it on the wall. This will turn you round, and you will retrace your footsteps back to the main path, but with your hand on the opposite wall. In a branching maze, all the walls are connected, so this way, you will end up by touching both sides of every single wall in the maze, and end up at the exit again. At some point, you will have visited the centre. Some people suggest changing hand at the centre, but that will make you return to the start by the same path you came in, which seems boring.

An island maze is a maze with choices, but with more than one path to the centre. Island maze walls are not all connected. There are several different bits of wall, and these are the islands. This means that you may find the branching maze trick (with the hand touching a wall at all times) won't get you to the centre. You may go round and round without getting anywhere. It can be very hard to get to the centre of a good island maze; however, a bad one will have too many correct paths, which means that it becomes easy to blunder onto the centre without trying.

Some non-branching mazes are technically island mazes, but they are possible to solve with the 'hands-on-wall' method. Hampton Court is an example of this. It depends where the islands are.

Hampton Court maze

Britain's oldest surviving hedge maze is the maze at Hampton Court Palace. It is the oldest hedge maze in the world in continuous use, although there are descriptions of earlier hedge mazes and mazes which have been remade from old designs.

The maze was designed by George London and Henry Wise in 1690. The original design is on the left. I think the bits in the middle are two trees with two people sitting under them. There used to be two trees in the centre, but they've gone now. The maze was originally planted with hornbeam, but is now made of yew.

The diagram on the right is the modern maze (the other way up!) You can see that the original symmetrical shape of the maze is now lop-sided. The red bits in this plan were originally part of the hedge, but are now gaps. (Thanks to John Florentin for this piece of information.) I suspect that at least some of the gaps were caused by bushes in the wall dying. This is always a problem with hedge mazes; see the Dublin maze.

Lauren Royal, a historical novelist from California, has an extremely interesting snippet about the history of the Hampton Court maze in the description of her novel Rose: 'As for Hampton Court's maze, the one you can visit there now was built in 1690 for William III, but it possibly replaced an earlier maze, perhaps laid out for Henry VIII. In an inventory of Cromwell's goods at Hampton Court dated 1659, there is mention of a cistern that serves "the fountaine and Maze."' The source for this reference is "History of Hampton Court Palace in Stuart Times" by Ernest Law, published in 1888 in London. The entry in question can be found in the back of the book in Appendix C, "Inventory of Goods mostly claimed as belonging to Cromwell, at Hampton Court Palace in 1659." The exact mention is on page 306 under "Store Cisternes" and reads: "One large Cisterne under the Square stone Courte that serves the fountaine and Maze there." The book includes a fold-out map showing the Stuart era floorplan of Hampton Court, but not unfortunately the "Square Stone Court".

The maze was always an island maze, and the gaps have increased the number of paths to the centre. The plan on the left shows why an island maze is so-called. There is one main 'island' or hedge, and four small ones. Some island mazes cannot be solved by the 'hand on the wall' method, but Hampton Court maze can. The smaller 'islands' are very small, and don't affect your journey to the centre much.

You don't get trapped going round and round one island without getting onto the next. This is because there is continuous hedge from the entrance to the centre. The plan on the right shows the paths to the centre. Red is keeping your right hand on the hedge, and blue is left. Note that you don't follow every possible side of hedge. In fact, you leave out the 'islands'.

The 'hand on the wall' method of getting to the centre is not the quickest route, as you have to follow several dead-ends. It merely guarantees that you don't get lost. On the left, this plan shows the quickest route for the modern maze. In fact, this path is much the same as the quickest path before the gaps were introduced.

These photos are of a recent visit to the maze. We used my maps to find our way in (except my son who tried to get lost, and got to the centre before us) and used the "take every left turning" method to find our way out. One person we met said that using a map was cheating, another asked where we got the map from, and a group followed us out. The maze is about a third of an acre with, it's claimed, half a mile of paths, although it didn't seem that long. Other websites on the internet say that it takes 45 minutes for the maze, but it took us more like 30 minutes, including lots of photographing. But then we knew where we were going.

Jerome K. Jerome gives a very vivid account of the Hampton Court maze in Three Men in a Boat (not forgetting the Dog) written in 1889. Harris has a map of the maze, and yet gets lost. I wonder if he had the original plan, and the gaps were already in place (I am not quite sure when the gaps were inserted). That would make confuse him badly! The account does say "keep taking the first turning to the right", but he must have made a mistake, since that would have worked.

Jerome K. Jerome said that it cost two pence to enter the maze, and this is in 1889. These are old pence, of course. One of my correspondents used to live near Hampton Court as a kid, in the 1960s, and says that it cost less, only one old penny to get in. It now costs a lot more!

Hampton Court from the air (courtesy of Google maps)

Here is Jerome's account: (Click here if you wish to by-pass this.)

Harris asked me if I'd ever been in the maze at Hampton Court. He said he went in once to show somebody else the way. He had studied it up in a map, and it was so simple that it seemed foolish - hardly worth the twopence charged for admission. Harris said he thought that map must have been got up as a practical joke, because it wasn't a bit like the real thing, and only misleading. It was a country cousin that Harris took in. He said:

"We'll just go in here, so that you can say you've been, but it's very simple. It's absurd to call it a maze. You keep on taking the first turning to the right. We'll just walk round for ten minutes, and then go and get some lunch."

They met some people soon after they had got inside, who said they had been there for three-quarters of an hour, and had had about enough of it. Harris told them they could follow him, if they liked; he was just going in, and then should turn round and come out again. They said it was very kind of him, and fell behind, and followed.

They picked up various other people who wanted to get it over, as they went along, until they had absorbed all the persons in the maze. People who had given up all hopes of ever getting either in or out, or of ever seeing their home and friends again, plucked up courage at the sight of Harris and his party, and joined the procession, blessing him. Harris said he should judge there must have been twenty people, following him, in all; and one woman with a baby, who had been there all the morning, insisted on taking his arm, for fear of losing him.

Harris kept on turning to the right, but it seemed a long way, and his cousin said he supposed it was a very big maze.

"Oh, one of the largest in Europe," said Harris.

"Yes, it must be," replied the cousin, "because we've walked a good two miles already."

Harris began to think it rather strange himself, but he held on until, at last, they passed the half of a penny bun on the ground that Harris's cousin swore he had noticed there seven minutes ago. Harris said: "Oh, impossible!" but the woman with the baby said, "Not at all," as she herself had taken it from the child, and thrown it down there, just before she met Harris. She also added that she wished she never had met Harris, and expressed an opinion that he was an impostor. That made Harris mad, and he produced his map, and explained his theory.

"The map may be all right enough," said one of the party, "if you know whereabouts in it we are now."

Harris didn't know, and suggested that the best thing to do would be to go back to the entrance, and begin again. For the beginning again part of it there was not much enthusiasm; but with regard to the advisability of going back to the entrance there was complete unanimity, and so they turned, and trailed after Harris again, in the opposite direction. About ten minutes more passed, and then they found themselves in the centre.

Harris thought at first of pretending that that was what he had been aiming at; but the crowd looked dangerous, and he decided to treat it as an accident.

Anyhow, they had got something to start from then. They did know where they were, and the map was once more consulted, and the thing seemed simpler than ever, and off they started for the third time.

And three minutes later they were back in the centre again.

After that, they simply couldn't get anywhere else. Whatever way they turned brought them back to the middle. It became so regular at length, that some of the people stopped there, and waited for the others to take a walk round, and come back to them. Harris drew out his map again, after a while, but the sight of it only infuriated the mob, and they told him to go and curl his hair with it. Harris said that he couldn't help feeling that, to a certain extent, he had become unpopular.

They all got crazy at last, and sang out for the keeper, and the man came and climbed up the ladder outside, and shouted out directions to them. But all their heads were, by this time, in such a confused whirl that they were incapable of grasping anything, and so the man told them to stop where they were, and he would come to them. They huddled together, and waited; and he climbed down, and came in.

He was a young keeper, as luck would have it, and new to the business; and when he got in, he couldn't find them, and he wandered about, trying to get to them, and then HE got lost. They caught sight of him, every now and then, rushing about the other side of the hedge, and he would see them, and rush to get to them, and they would wait there for about five minutes, and then he would reappear again in exactly the same spot, and ask them where they had been.

They had to wait till one of the old keepers came back from his dinner before they got out.

Harris said he thought it was a very fine maze, so far as he was a judge; and we agreed that we would try to get George to go into it, on our way back.

Thomas Hill's mazes

In Thomas Hill's "Treatise", there is a design for a garden maze, which is a super Chartres design, bastions and all. However, they suggest that you plant fruit trees or rose bushes in the bastions! The book is dated about 1568, which is a lot earlier than Hampton Court maze (1690). It is, of course, a unicursal maze, without choices or branches, while Hampton Court maze is a puzzle maze. Still, it shows that the idea of a maze in a garden goes back at least that far.

Here is a simpler maze of his, similar to a Chartres design, but different.

This maze is a knot, where the 'walls' seem to weave over and under each other.

Batty Langley's mazes

Batty Langley was an architect and garden designer. He wrote an influential book called "New Principles of Gardening" in 1728. He liked Gothic in gardening, and seems to have had a soft spot for mazes. His first name, by the way, is not a description or nickname like Capability Brown, but seems to be a family name (see the International Genealogical Index).

This design on the left has been replanted in Castle Bromwich Hall Gardens. The design of the maze and the associated garden comes from "New Principles of Gardening". I used my maze designing webpage to reproduce the design. It is in blue as it shows the walls. The maze is basically the same as the Hampton Court maze but it has been massaged into a plain rectangle, and is also a mirror-image. I wonder if Batty Langley was working from an engraving, which reverses the image? Or perhaps he was pretending that his image was original!

I don't know if there is a modern example of this second maze, also in "New Principles of Gardening" but here is his design, in red, as it shows the paths. There are a few more wiggles at top and bottom which I haven't shown as they don't really add anything to the maze itself. I do find this maze more interesting. It does have branching paths, so is not a unicursal maze. However, it has practically no dead-ends - just one tiny one (which you would have thought would be obvious). It achieves its multiple paths through multiple entrances. There are three ways in and two ways to the centre. You might even enter via one path and exit via another without ever reaching the centre. In fact, it seems possible to get from any entrance or centre or any other. So perhaps it is rather an arrangement of paths, like parterres, rather than a true puzzle maze. Batty Langley lived 1696-1751, and Hampton Court maze was laid out about 1695, so this must have been a time when hedge mazes were the latest fashion.

Gonzaga maze

This is a Gonzaga maze, which I found on the web. The Gonzaga were an Italian family who ruled Mantua from 1328 to 1707, and had the labyrinth as their symbol. There are other Gonzaga mazes here and here.

Presumably it's a garden maze, marked out with hedges or herbs. In Elizabethan times in England, these gardens were called knot gardens, because they looked like embroidery. It is an excellent way to make a long walk in a small space. It gives you constantly changing view points and a good look at all the plants in the flower beds.

The original picture is a little small and blurry. I have tried to produce what the original looked like, but you're welcome to disagree. You have to walk through the whole garden twice.

Barcelona maze

I have been told by Adrian Fisher that the Gonzaga Hedge Maze is in Barcelona. He says that it is the finest hedge maze in Spain. I assume that this is it.

The parc del Laberint d'Horta (Labyrinth Park of Horta) is a historical garden in the Horta-Guinard� district in Barcelona. Works began in 1791 when marquis Joan Antoni Desvalls i d'Ardena, owner of the lot, created the design of a neoclassical garden in collaboration with Italian architect Domenico Bagutti. Execution was made under direction of master builders Jaume and Andreu Valls as well as French gardener Joseph Delvalet. The maze was created created around 1794.

The diagram on the left shows that this is a puzzle maze. There are 8 apparent ways to the centre, but 6 are dead-ends (green). On the other hand, it doesn't matter which of the other two you take. There are some dead-ends even on the OK path (pink), but there are also multiple paths to the centre. The centre is an odd shape in one diagram because I was trying to map the curves of the original into a square diagram, and it didn't fit very well!

Schloss Schonbrunn maze

The Maze at Schloss Schonbrunn was laid out between 1698 and 1740. During the 19th century the Maze was gradually abandoned until in 1892 the last remaining hedges were felled. In autumn 1998 a new maze was laid out taking the historical model into account where possible. The photo on the left is taken from Google Maps - enter "Schloss Schonbrunn" and choose Satellite.

Perhaps the maze in the film the Shining (see right) was inspired by this maze. It's hard to see as a tree obscures the centre of the maze.

Bridge End maze, Saffron Walden

Saffron Walden has a famous turf maze on its common. It has a second, puzzle maze in Bridge End Gardens. You enter at the bottom, and half way through, you come to a second entrance/exit at the top, where there is a summer house. You continue on to the centre. Here there is a standing stone (modern, I suspect!) and a viewing platform to see over the maze. The pictures were taken during the 2016 Saffron Walden Maze Festival. It was very tantalising to hear the musicians play in the centre while trying to solve the maze, with the music getting closer, then further away...

Bridge End maze in Saffron Walden

The next three mazes, from Egeskov Castle, Chevening House and Hatfield House, were reproduced on Saffron Walden common as part of their maze festival in 2011.

Egeskov Castle maze

The estate of Egeskov Castle in Denmark contains 4 hedge mazes (see below), of which this one (left, below), the oldest, was created in the 19th century. The maze is formed of high bech hedges. Its design is based on the Hampton Court maze (see above). This is an island maze. Looking at some possible paths to the centre, I wonder if it is suffering from extra gaps, caused perhaps by beeches dying. The quickest path to the centre (red) doesn't even into into one half of the maze. The green path seems a more sensible second half of the solution. BUt perhaps they wanted a quick and a slow way to get to the centre!

The photos are taken from Google Maps.

Chevening House maze

Between 1818 and 1830 the 4th Earl of Stanthorpe planted a yew hedge maze at Chevening House in Kent to a design by the 2nd Earl who was a famous mathematician of his day. Unlike ealier historic hedge mazes, you cannot solve this maze using the hands-on-wall method, as this will take you right round the maze and back to the entrance without ever reaching the centre. Standthorpe's trick was to include 'islands' of hedges in the layout which have no connection to the outside perimeter (see below centre). This significantly increase the difficulty of finding the right path (below right).
The photo on the right is taken from Google Maps.

Hatfield House maze

The yew maze at Hatfield House was planted in 1840. There are two entrances, on each side. Each one takes a different path to the centre, so you can enter by one and return by the other. The maze is in fact split into two, so once you enter by one path, you must go through the centre before being able to access the second half of the maze at all.

Dublin maze

Dublin has beautiful parks (something to do with the rainfall?) The Iveagh Gardens are hidden away so not many people find them. They were designed by Ninian Niven in 1863 and include a rustic grotto, cascade, fountains, rosarium, archery grounds, wilderness, woodlands and a maze. Here is a diagram of the maze, in blue as it shows the walls. There is a sundial in the middle. The hedges are very low, and the maze is simple, so you can see all the blind alleys, and walk straight to the centre without a mistake quite easily. The sticks poking up in the photo are protecting new hedge plants, replacing those that have died.

Photo of Dublin Maze

Blackpool maze

Longleat mazes

Here are some excellent modern mazes at Longleat, courtesy of Google maps. The designs are copyright and should not be copied.

Soekershof mazes

Soekershof is a collection of mazes and botanical gardens in South Africa. The aerial view is from Google maps.The Klaas Voogds Maze is 13,870 sq metres, and there is a butterfly maze and a cactus maze. The photo above shows the cactus maze, which is a small Cretan maze.

Conwy Valley maze

Conwy Valley maze is a large hedge maze in the Conwy Valley, North Wales. The aerial view is from Google maps. The maze is over 2 acres. It is constructed using English yew and contains themed gardens.

Maze in Russell painting

This painting is of Lord Edward Russell, grandson of the first Earl of Bedford, by George Perfect Harding, after an unknown artist, 1573.

The maze seen through the window is enlarged on the right.

Daniel Loris's mazes

Finally a large collection of old maze designs. A delightful book called 'Le Thresor des Parterres de l'universe', by Daniel Loris and published in MDCXXIX (1629) can be seen in full here. It is intriguing as it seems to be right at the beginning of the fashion for garden mazes. In fact, much of the book is taken up with parterres. Most of his maze designs are unicursal, without choices or branches, but a few do have choices. I have given diagrams of his designs below. The blue diagrams are walls, and the red (and colours connected with red) show the paths. The different colours attempt to show how you walk through the maze, and so how complicated it is.

Comment	Walls	Paths
Labyrinth 1 This resembles a Roman maze. Each quarter is walked in turn, which makes it rather a dull maze to walk, but its symmetry makes it quite attractive to look at. It's not quite a Roman design, since they rotate the design through 90 degrees each time, whereas this design reflects each quarter. You could enlarge it quite easily.
Labyrinth 2 This is also based on a Roman maze, but it has included some choices. Some of these give you an alternate path onwards, but some would take you backwards. I mark one path in red, and the alternatives in pink. This labyrinth also finishes at an edge rather than the centre.
Labyrinth 3 This is based on a Chartres maze, although it is not quite the same design. The shape is interesting as well. Some Chartres mazes are octagons, and some are round with bobbles on the corners. This seems to use both ideals, but still fits within a square. The paths design, going from red to pink to purple to dark red, shows how the walk progresses.
Labyrinth 4 This is also based on a Chartres maze. It has some dead-ends and alternate paths (in green). The shortest path does the bit near the centre (red), then the left side (pink), then the right side (purple). You can't get into the centre of the squares in the corner at all.
Labyrinth 5 This looks very complicated, but the basic idea is still a Roman maze. You complete each corner before going onto the next. You can visit each corner garden and the garden in the centre, and end up on the edge, by the entrance. The path goes yellow, gold, red, pink, purple, dark red (or the other way round, of course).
Labyrinth 6 This does each of the corners in turn, then zig-zags into the centre. The green path is an alternative, which in fact gets to the centre quicker, missing out the bottom two corners. I'm not sure whether this was intentional, or whether he just wanted to fill out the pattern! The main path goes yellow, gold, red, pink, purple. The other path is yellow, gold, green, purple. If you take the green path in the wrong direction, it will lead you backwards.
Labyrinth 7 This is a more interesting maze. It doesn't have a centre. First it does the top two corners in order, rather like a Roman maze (these are the pink and purple). Then it wiggles to a branching point (marked green). Here the path splits into two, but this is the beginning and end of a loop. So you can go out by the red and back by the dark red, or the other way round. When you've done this, you retrace your path through the purple and the pink back to the entrance (or of course make a wrong choice at the green and round and round the loop!)
Labyrinth 8 This maze has two entrances, which lead to two different paths (pink and red). These meet at the green spot, and then there is a single path to the centre (dark red). Of course, if you take the wrong path at the green spot, you will go out without ever seeing the centre.
Labyrinth 9 This maze has also two entrances, but perhaps you should think of it as an entrance and an exit. The top half of the labyrinth is the same as the bottom half, rotated. In each case, you do the centre of one corner, then wiggle back round the outside, then between the two corners, then outside the second corner before filling its centre. There is no (single) centre to the maze and eventually you find yourself on the other path, with no choices, part from whether to go into the middle of the four corners or not. The colours are red, pink, purple, dark red, gold, yellow (or the same in the opposite direction).
Labyrinth 10 This is a standard Chartres maze design. It is a triple level one (or Super Chartres) and square. Chartres mazes are usually round. However, the underlying pattern is the same.
Comment	Walls and Paths
Labyrinth 11 This is a very large and rather dull maze. I have combined paths and walls together, as I can't imagine that anyone would ever want to reproduce it. The red is the shortest path. The dark green are dead-ends and the pale green are loops and alternate paths. There are alternates off the alternates, but I haven't marked those. It is interesting that there are dead-ends (although I suspect that he put them in just to fill out space!) But it wouldn't work as a real maze. The dead-ends and alternatives are so long, that once you made a wrong decision, you would end up in entirely the wrong part of the maze and so wander for a long time before you had the chance to correct yourself. Even the correct path is too long. It might be acceptable if there were short hedges that you could jump over, or flower beds or grass marking the walls, so once you had got thoroughly fed up, you could just walk back to the start in a straight line.
Comment	Walls	Paths
Labyrinth 12 This looks like an attempt to give some genuine choices. The pale greens are alternate paths which are almost loops. There is one dead-end, but that's off an alternate path, which is a bit of a waste (since you'd have to make two mistakes to find it). However, it's not obvious what the point of the maze is, since there's no centre, and only one entrance. The centre triangle has no entrance (this is the dead-end) and this isn't a mistake, as otherwise the path to the centre would be too easy. I presume that you wander round until you leave by the entrance, and the green paths make your walk slightly longer.
Labyrinth 13 The second triangular maze is larger, but not as interesting. Again it has no centre and only one entrance, so you walk a loop. You can walk the red and purple, or the red and pink, or even all three colours. None of them can be called mistakes though, unless you start taking the same path more than once.
Labyrinth 14 A strange shape! I'm not sure if it's supposed to be a boat or a fish. The maze has several choices, with the mistakes leading to longer alternatives or loop. The quickest path is red, pink, purple. Pale green is a longer alternative, and dark green a loop. This isn't really a good puzzle maze, as the shortest route only has two choice points. Much of the maze space is used up with the results of mistakes. If someone doesn't make either of those two mistakes, he has a short trip to the centre. What's more, the correct path at choices tends to point closer to the centre, which makes it too easy.
Labyrinth 15 This is similar to a simple Roman maze. It's round rather than square, and has eight sections rather than four like the Roman mazes, but it's the same principle - do each section in turn before moving onto the next. Rather dull to walk, but attractive to look at, especially in this round form. There is one walk to the centre (red) and another to return (purple).
Labyrinth 16 This is a development of the previous maze. The path to the centre has been moved to line up with the entrance, which means that you walk the whole maze to get to the centre. This means that there needs to be some extra paths so each section goes both to the edge and the centre. The path goes red, pink, purple, dark red.
Labyrinth 17 This is similar to a Chartres maze. The design is not quite the same, but it's the same principle. I'm pretty sure that the dark blue square (a gap in the original) is a mistake. Having a choice there really makes no sense - it just gives a short alternate path. There's nothing wrong with short alternates, but you need more than one in a maze!
Labyrinth 18 This is an interesting design, but spoiled by details of the design. It's basically circular, but there is a rectangular grid superimposed on it, which means that the zigzags go in different directions. However, there are just too many choices (green squares). The shortest path (red) is just a fraction of the maze. If you get the first two choices right (and both times, you should chose the path towards the centre, which seems obvious), then you are unlikely to find most of the maze (pink). Still it would be quite attractive as a garden with multiple paths.
Labyrinth 19 This is more of a parterre than a maze. There is a choice (green) at the beginning. One path (red) takes you round the circles in the middle and the other (pink) takes you round the edge.
Labyrinth 20 This is another maze with a lot of choices (green) but this one is better designed. The quickest path to the centre is longer and there are more potential wrong choices on it. However, the best path still tends to be the one pointing towards the centre, which is what people would naturally chose. There are no dead-ends, so strictly speaking the 'wrong' choices aren't wrong, but merely longer paths to the centre, unless you end up walking out the entrance instead or going round and round in a loop.
Labyrinth 21 This maze has far too many choices. All you have to do is zigzag to the centre.
Labyrinth 22 Another simple Roman maze, although rather an odd shape. You walk to the centre on the red path, and back again on the pink path. The junctions of the paths are green.

Comment	Walls	Paths
Labyrinth 23 This is definitely showing off. It is loosely based on a super Chartres maze. It has no choices at all. I would think that it would be fairly difficult to lay out on the ground! The path goes red, pink, purple, dark red, gold, dark grey, pale grey, yellow.

The book, 'Le Thresor des Parterres de l'universe', has an introduction in Latin, French, German and English. I give the English part below. I have changed the spelling to modern. The original (dated 1629) used the long 's', and 'i' for 'j', and 'u' for 'v', and 'vv' for 'w', which can be hard to read until you get used to it. I have left the vocabulary and grammar. The original is here. I have also given what I think he is saying, at the side of each paragraph. I have also made the occasional comment on it, in brackets.

Original

My comments

The Manner of dressing banks and beds in gardens

Although the figures and images of our beds be represented with no less labour and cunning than with the proportion required, so that it were easy to draw them upon the paper, or to compass them upon a piece of ground, nevertheless that this might be more easily understood and profitable to any soever will be pleased to see them or use them, it seemed good to set down here the manner and following instruction for it.

It is easy to lay out a garden using these plans, but here is how you do it.

It is required ere to come to the direction of the banks and beds to know how many feet containeth the compass of the place, then to keep it to the square until it be reduced unto a true square, either upon a piece of paper, or upon the ground, which being not done, hardly could they well be measured. Afterward thou shalt with a drag make the ground so clean, small, and even, that no turf or clod, nor stone that could hinder it, may be found in it.

Before starting to lay out the design on the ground, work out how big the design will be. Whether on the ground or on paper, draw a square with all the corners true right angles so you end up with a true square. Work the ground to a fine, even tilth.

Moreover it is also required in the dressing of beds that the piece of ground or square be broad of four feet and the alleys of two, which the gardeners think to be the fairest and most profitable, as it shall appear by the following figures and images. But if thou desirest to dispose it to thy own pleasure, and according to the commodity of thy garden, thou may add or take some away, in making thy banks, so that at least the ways between be by the half narrower than the square.

The alleys are the paths, and between the paths are the beds. You can make the paths two foot wide and the beds four foot wide, which most gardeners prefer. However, if you want, perhaps because your ground is too small, you can make the widths smaller, but the paths should still be half the width of the beds. (This is only for parterres. See later for labyrinths.)

Take heed also that two squares, or two alleys do not touch themselves one another, but that one alley or way be always found between two pieces of ground.

In the design, there should always be a path between two beds.

If therefore thou desirest to dress some banks, thy piece of ground first being squared in all the corners, thou must within this little compass choose a proportionable figure of the same space and compass as thy ground, and such as shall seem thee fitter for thy turn, then thou shalt draw it upon some paper in the following manner.

Draw out the design on paper before starting work on the ground, so you know what you're doing.

Lay down in one corner a branch of the compass and take the breadth of about four feet, where having made a point, that shall already serve for a piece of the square, then with another compass thus far by the half (which are two feet for the alleys) thou shalt again make a point; Then thou shalt take again a piece of ground, until thy square be equally divided into so many feet as thy figure requireth, referring itself exactly to the corner right over against whom thou meanest to make thy banks. That being done done thus the same must thou make of the three other sides. After this thou shalt put thy line from point to point right over against, and with a bone or leaden style draw so many broken lines as the points done by thee shall show thee till the end of the place. Do the same in the other sides, as that seen in the first figure noted by E and F. That being done thy broken lines of a right figure shall be all perfected.

You should already have a drawn square (several paragraphs ago). Using a pair of compasses (or possibly dividers) use one distance to represent four feet, for the beds, and half that to represent two feet, for the paths. (We would probably just use a ruler.) Mark along the edge of the square, the points representing the beds and the paths. (This will tend to be bed, path, bed, path, bed, etc. starting from one corner.) Do this along all four edges. Then join opposite edges with dotted lines (in pencil, I suggest, rather than a bone or lead stylus!) to make a grid.

The manner of going further shall be easily known by the figure Numero A, in whom thou shall see all the projects, how it is meet to dispose the piece of ground by beds, and the broken lines by little points.

This seems to be the diagram that he's talking about. The letter E shows a bed and the letter F is at one end of a path.

If thy figure seems to thee well drawn upon paper, as it hath been showed, and thou desirest to form it upon the ground, thou shalt do in the divisions of the broken lines the same thou hast done upon the paper, except that instead of thy compass, thou must use a pole, and upon that thou shalt mark four feet, then two. Measure with it the feet marked from one of the corners to the other, and where the four feet shall meet upon the ground set there a nail, and near the two feet another, and so consequently, until thou come back again to the corner, where thou hadst begun. Thy nails being thus planted thou shalt lay a line from every nail right over against the nearest, stretching and bending, that it may be right, in the same manner thou hast done with the broken lines.

Now to lay it out on the ground. You should already have your square. Take a stick and mark on it four feet and two feet (or use a tape measure!)

Mark along the edges of your square the same way you did on the paper, except you drive in a nail rather than make a pencil mark. I'm not quite sure whether he marks out the grid with string or whether he puts nails in at all the junctions, and then winds string round the nails to lay out the pattern (see example).

That being thus performed on every side, then thou shalt judge of thy banks by the figure which was drawn upon the paper, no work remains to be done but to compass thy square about with a box border, or any other than seems thee good, as it will appear by thy figure.

Then, all you need to do is plant box hedges, or any other plant, along the parts of the grid to make the pattern. (I do like that off-hand way of describing a lot of hard gardening! Remember that he is describing mostly parterres. The beds might be filled with plants, or grassed over, or even filled with gravel, perhaps of different colours. The word 'banks' suggests that they might be heaped. The paths themselves would be grass or gravel or other paving material. Labyrinths are different - see below)

And if thou will divide a figure or piece of ground by lines half across use likewise the right figure, which has formerly been taught, so that it be compassed in so many parts as every square requireth. Thou shalt therefore lay one of the branches of thy compass in a corner and stretch it out to eight or ten feet, making two points with it in both corners toward A and C whom thou shalt join together with a line as that may be seen in the figure, Numero 2, in the corner marked by B. After that, thou shalt draw yet from the corner B & D, a line overthwart to the both corners, then in the same place where they shall cross themselves, there thou shall make a point, from whom thou shalt measure at every corner two or three points two feet far from another, thou shalt also do the same in the other sides. Afterwards thou shalt join together the points, whom for better instruction's sake, I have noted in the first figure upon the line E F with 1,2,3,4, until 8, and for better understanding of that, take before thee the figure B, which shall give you the plain direction of it.

Some of the patterns are laid out on a slant, so the square grid won't help you. You need to draw diagonal lines.

Draw the lines which cross in the middle first. Then move in from the corner about 8 or 10 feet, draw a line at right angles to the diagonal, and mark out the two feet markers on this. (Or, I suppose, 4 feet markers if necessary). Do this on both diagonals to get a diagonal grid.

As for that which concerns the disposition of the figures, which are wholly in crossing lines, the manner of dividing in right form, cannot serve much in that, as in those which are in half overthwart lines, but one must first seek the middle, which being found, thou shalt draw from it in manner of a cross, one upon another a single broken line. Or if it happens that a square should have one alley in the middle, thou shalt afford, as we have said, two feet, I say of the single lines a foot at every side, thus much must thou do upon the corners, according to that which hath been showed thee in figures which are in half crossing lines. Seeing then these figures in their divisions are not like themselves, as the former, and that in many the circle is much in use, so that a true division of broken lines, which are most useful, cannot be found. Therefore the division of the numero 3 (after whose fashion the most part may be formed) as also the figure C shall give a sufficient instruction of it.

To draw out the diagonal grid, you must start by drawing the two diagonals of the square first.

If there is a bed in the corner, then the fist parallel diagonals either side of the main diagonal are two feet either side, making four feet between them. If there is a path in the corner, then the first parallel diagonals either side of the main diagonal are one feet either side, making two feet between them.

(I must point out that the diagram does not show this!)

Moreover, there are many squares, which for the most part are made with the circle, and needs not to be compassed, and specially those which those which are formed after the French fashion, having but few single lines broken, or points which the compass shall show by the use. In the meantime my mind is to teach how it is meet to use the compass upon the ground.

Provide then a wooden compass as great as the piece of ground shall seem thee to be, or more, to measure withall the space of it.

Some designs, especially the French designs, use curved lines rather than straight lines (see example). This means that you need to use a pair of compasses big enough to mark out the circles on the ground (presumably two poles jointed on one end to make such a large pair of compasses).

After that provide too one or more wedges, of three, or four fingers broad or thereabout sharp downward and flat upward, with a little hole, where thy compass may stand. Pitch the wedges in the ground, where thou shalt need a point to draw from thence somewhat with the compass either in the middle, or elsewhere. Put there a branch of thy compass within the said wedge, and with the other draw the roundness of thy circle as much as is required. Afterward thou shalt lay further thy wedge, where it will be needfully, which is very easy, without much turning to and fro, and treading the ground, but thou shalt hold fast near the points, and may draw with the compass whatsoever thou desirest.

The wedges have a hole in them for one end of the pair of compasses. The point of them seems to be that the wedge will stick in the ground, and won't shift even when you are drawing the arc of the circle, whereas just ramming one end of the pair of compasses straight into the ground might. Also, if your static end of the compasses is firm, you can just twiddle the compasses easily and while standing still, rather than having to hold the compasses steady, which might mean walking along the line of the arc. I don't know why you need more than one wedge, though.

There is yet another means, with the line, although no compass may be had, in which line is made a running knot, who is tied to a staff fastly turning it about, afterward the staff must be planted in the ground, where there is need of a point, then with another staff thou shalt take the end of the line, and with all shalt mark the compass of the ground as much as shall be required. But because the line is sometimes uncertain, and that it easily looses itself, thou must run much, and trample the ground: I think it will not be needful to make further mention of it.

It seems obvious to drive a stick into the centre of the circle, and tie a string to it, and use the string to mark the arc. But it isn't as easy as you think. (Interestingly enough, Tom Baxter found problems in using this technique, but he came up with an intelligent solution to the problem! Of course he didn't mind walking over the ground, as it was grass.)

As for that which concerns the fourth division of the first figure, seeing that it could not serve in many figures, it hath not been set there to any other purpose but for the labyrinths, and that if any figure or square to be divided, which could not be referred to the other divisions, should occur, this might be used.

As for the construction of labyrinths, that may be sufficiently understood by the figures heretofore represented, wherefore there is no need to speak much of it.

This diagram (with equal widths for paths and beds) is to be used for labyrinths. (Labyrinths are different from parterres, since there are no borders between the paths or alleys, and the walls or beds. The 'beds' ARE the borders, so the width of the 'beds' or walls should be two feet rather than four, and the hedges planted in the beds rather than between the beds and the paths.)

There are three other figures, where the first two serve for the gardens of things to eat; which teach the manner of compassing them properly, and profitably, needing no box borders, but every year may be dressed in such form. It is meet to give to every alley a foot and a half of breadth, and to the piece of ground, four. Which being so, from the alleys, one may cleanse the said piece of ground without treading upon. As for the third figure, it serves to set rightly and in equal distance the trees in orchards, to walk therein by any way thou desirest. It is also taught by these broken lines in small points within this figure how the lines are to be drawn.

If you are making a vegetable beds, then allow less room for the paths (one and a half feet rather than two), and four feet for the beds. This means that you can clear up the bed (and plant, weed and harvest it, incidentally!) without treading on the soil. You can also plant out orchards in the same way (although the spacing may be different, I should imagine, depending on the type of tree).

Moreover, is seen the space and compass of the figures, to wit, how many feet every figure of it containeth, where the letter P notes a foot. As for the banks after the French fashion, and the labyrinths, every one may dress them at his own pleasure, greater or lesser, for they are not compassed according to the feet.

Some of the designs give how wide they are, e.g. 46 P or 46 feet. (The French had a unit of measure at the time called Pied du Roi, or Foot of the King. This was nearly the same as an English foot.) Labyrinths and French parterres (as opposed to German parterres) do not conform to the spacing that he has already described. (French parterres have curving lines in them. Labyrinths have paths and walls, or beds, the same width, say two feet. In fact, later, he describes labyrinths having cottages in the centre! His plans don't allow room for this.)

This little work, for a better instructions sake, is briefly and distinctly divided in three books.

The diagrams are divided into three groups.

In the first book are represented in many fair and profitable figures in the manner they use to be divided in Germany in the gardens of princes and other gardens of pleasure. Wherefore they are named by the gardeners figures or banks after Dutch fashion.

(The German parterres tend to have more rectangular beds, see left). 'Dutch' doesn't mean belonging to Holland, it means Deutsch or German. This website doesn't give these patterns. See this website.)

In the second book follows the figure after French fashion which are woven between, so that the ways cross themselves mutually like little lovers knots, or other things mixed one with another. And because they are very common in France, therefore they are called figures after French fashion, or woven figures.

(The French parterres have curving paths, see left. This website doesn't give these patterns. See this website.)

In the third are referred many fair and most chosen labyrinths, in whom, (as also in the other figures) according to the space and commodity of the middle you shall find fifty most rare and well built cottages, to frame some, like them. Which nevertheless hath not been done, as if needfully they should be set in those same places and pieces of ground, but only to the end this little work might be so much shorter and richer by it. Whosoever will adorn and beautify his own garden by those cottages, he may do it according to the commodity of the place, and as he shall find to be most convenient.

Some of the labyrinth designs have cottages in the middle, some don't. You can add a cottage in the middle of your labyrinth if you want, and if there's room.
(The labyrinth designs are on this website, see above. I've left out the cottages, though!)

And indeed the pictures represented within this little volume may much serve, not only to the gardeners, but yet to other trademen, who do tapestry, embroidery, works and pictures, as also to carpenters, saddlers, and others who will make use of it.

These designs can be used in tapestries, embroidery, pictures, wood (possibly as inlays, or carvings) and leather-work.

By this brief introduction the reader being informed of the mode and manner of compassing the gardens and making banks, he will not find that hard to do at any time he will be pleased to make the trial of it, who is desired to take this work in as good part as we have been desirous to bring him profit and contentment by it.

The reader can now easily make his own gardens using these designs.