Sources page biographical material

6.W.1. THREE PIECE BURR
Most of these have three pieces which are rectangular in cross-section (1 x 3 x 5) with slots of the same size and some of the pieces have notches from the slot to the outside. When one piece is pushed, it slides, revealing its notch. When placed properly, this allows a second piece to slide off and out.

In the 1990s, a more elaborate type of three piece burr appeared. These have three 3 x 3 x 5 pieces which intersect in a central 3 x 3 x 3 region. Within this region, some of the unit cubes are not present, which allows sliding of the pieces. Some versions of the puzzle permit twisting of pieces though this usually requires a bit of rounding of edges and the actual examples tend to break, so these are not as acceptable.

Crambrook. 1843. P. 5, no. 4: Puzzling Cross 3 pieces. This seems likely to be a three piece burr, but perhaps is in 6.W.3 -- ?? It is followed by "Maltese Cross 6 pieces".

Edward Hordern's collection has examples in ivory from 1850-1900.

Hoffmann. 1893. Chap. III, no. 35: The cross keys or three piece puzzle, pp. 106 & 139 = Hoffmann-Hordern, pp. 104-105, with photo. One piece has an extra small notch which does not appear in other versions where the dimensions are better chosen. I have recently acquired an example which appears identical to the illustrations but does not have the extra notch - this came from a Jaques puzzle box, c1900, and Dalgety has several examples of such boxes with the solution, where the puzzle is named The Cross Keys Puzzle (cf discussion at the beginning of Section 11). The photo on p. 105 is an assembled version, with verbal instructions, by Jaques & Son, 1880-1895 (but Jaques was producing them up to at least c1910). Hordern Collection, p. 67, shows Le Noeud Mystérieux, 1880 1905, with a pictorial solution and this does not have the extra notch.

Benson. 1904. The cross keys puzzle, pp. 205 206.

Pearson. 1907. Part III, no. 56: The cross keys, pp. 56 & 127 128.

Anon. A puzzle in wood. Hobbies 31 (No. 795) (7 Jan 1911) 345. Three piece burr with small extra notch as in Hoffmann.

Anon. Woodwork Joints. Evans, London, (1918), 2nd ed., 1919. [I have also seen a 4th ed., 1925, which is identical to the 2nd ed., except for advertising pages at the end.] A mortising puzzle, pp. 197 199.

Collins. Book of Puzzles. 1927. Pp. 136-137: The cross keys puzzle.

E. M. Wyatt. Three piece cross. Puzzles in Wood, 1928, op. cit. in 5.H.1, pp. 24 25.

Arthur Mee's Children's Encyclopedia 'Wonder Box'. The Children's Encyclopedia appeared in 1908, but versions continued until the 1950s. This looks like 1930s?? 3-Piece Mortise with thin pieces.

A. S. Filipiak. Burr puzzle. Mathematical Puzzles, 1942, op. cit. in 5.H.1, p. 101.

Dic Sonneveld seems to be the first to begin designing three piece burrs of the more elaborate style, perhaps about 1985. Trevor Wood has made several examples for sale.

Bill Cutler. Email announcement to NOBNET on 27 Jan 1999. He has begun analysing the newer style of three piece burr, excluding twist moves. His first stage has examined cases where the centre cube of the central region is occupied and the piece this central cube belongs to has no symmetry. He finds 202 x 10⁹ assemblies (I'm not sure if this is an exact figure) and there are 33 level-8 examples (i.e. where it takes 8 moves to remove the first piece); 6674 level-7 examples; 73362 level-6 examples. He thinks this is about 70% of the total and it is already about six times the number of cases considered for the six piece burr (see 6.W.2).

Bill Cutler. Christmas letter of 4 Dec 1999. Says he has completed the above analysis and found 25 x 10¹⁰ possibilities, which took 225 days on a workstation. The most elaborate examples require 8 moves to get a piece out and there are 80 of these. He used one for his IPP19 puzzle. He has a website with many of his results on burrs, etc.: www.billcutlerpuzzles.com .