I'm not a fan of Bob Triples, and try to avoid ringing it when I can. The dodges in 3-4 and 5-6 at the lead end completely spoil what Triples is supposed to be.

 

However, Bob Triples can still be an interesting method for devising extents (albeit, not nearly as fascinating as Grandsire or Stedman Triples. UPDATE: Actually, I think that Bob Triples is more interesting than Grandsire for devising extents! It can be twisted and turned in many more interesting ways.) Interestingly, you can get peal compositions of Bob Triples with no singles. This is possible for three overlapping reasons.

​

Firstly, you do not need singles to attain negative changes as you do in Bob Minor.

 

Secondly - and unlike in Grandsire - the lead ends of Bob Triples are alternately in-course and out-of-course, due to the fact that only two pairs of bells swap over at the lead end.

 

Thirdly, when the plain course is rung backwards it is still in-course (like in Bob Minor, but unlike Bob Major, or the B-blocks of Grandsire Triples.)

 

Because of this third reason, it is possible to come up with q-sets in which not all of the members are rung in the same direction. For example, instead of having the normal bobbed q-set group 423567, 342567, 234567, you could have 423567, 432657, 234567. These latter three are all bobbed as well, the only difference being that the change 1342567 from the first group occurs at handstroke in the second group, instead of at backstroke. I suppose these are similar to many q-sets used in Stedman Triples, where a third q-set member does not occur at the six end - a technique that makes the magic blocks for bobs-only peals possible. The existence of these 'Irregular Q-sets' is one of the few ways in which compositions of Bob Triples can be more versatile than Grandsire Triples - Grandsire has no such special Q-sets at all.

 

In Bob Triples, the fact that these special q-sets exist in which one of the members is rung in the opposite direction can be helpful in the following scenario. We have a peal composition of Bob Triples partioned into two separate round blocks that need to be joined together. In many methods singles would have to be used, as bobs cannot join together an even number of blocks. But in Bob Triples we can search for one of these irregular q-sets, with two members in one of the blocks, and another member in the other block. When the irregular q-set member is reached by the addtition of the extra bob, this switches the direction of the ringing, and the block is rung backwards until the next member of the new q-set group arrives.

 

This is an amazing trick which does three things. Frstly, it allows an even number of blocks to be joined without resorting to singles. Secondly, part of the reason this is possible is because some of a block can be rung in the opposite direction, but without falseness or collision. Thirdly, it passively shunts the bells into one of the required part ends, preventing the composition from coming round at the first part end (this is a crucial property for producing - for example - perfect five part compositions, which is difficult to do in Bob Triples as niether a bob nor a single affects five bells. These irregular q-sets can solve this problem.)

 

These irregular types of Q-set also exist in Bob Minor, and it is thanks to them that a bobs-only 360 in that method is possible (in contrast to the bobs-only 300s that are also possible.) To understand how they work, examine the Bob Minor 360 W, H, W x3, and compare it to the 300 W, W, W, H, H, H. Even in Bob Minor, the 360 passively shunts the bells into the part end, as described for Bob Triples.

 

John Holt, that self-taught pioneering genius of the 18th century, clearly knew about these irregular q-sets, and the special properties they yield. He used them to produce his brilliant bobs-only three part of Bob Triples, shown here in a notation that is more sympathetic to Holt's generation:

​

5040 Plain Bob Triples

by John Holt

234567

235746 1

645273 4

642357 1

526473 5

652473 6

543762 2

354762 6

643527 5

645732 1

235674 4

236457 1

475632 3

354726 5

435726 6

625473 4

562473 6

256473 6

543726 2

547632 1

546273 1

654273 6

523764 2

352764 6

235764 6

465273 4

754632 5

562374 2

256374 6

534726 2

453726 6

342567 5 (Plain lead)

​

Published in about 1753, this composition is typical of Holt's genius, being elegantly structured, with very few calls, and no singles. Some have speculated that the Grandsire Triples compositions Holt's ten-part and Holt's Original were in fact composed by Benjamin Annable, and that Holt plagarised them. I don't believe this, and I think the ingenuity and elegance of this composition helps corroborate the two Grandsire compositions as definitely being by Holt. They are all certainly neater than what Annable composed.

​

                                                Other bobs-only peals

As mentioned above, 5-part peals are made possible in Bob Triples thanks to the irregular q-set groups, and these are usually bobs-only. There are several examples of these 5-parts around, although they are probably all variations of the first one ever to be composed, which is almost certainly this extremely neat and ingenious peal by old-time composer Henry Hubbard. I have never seen a B-block peal of Bob Triples published before, so realised that a bobs-only 5-part made up from B-Blocks was a potential gap in the market, and produced this peal. I then turned this 5-part into a palindromic 10-part 10080 with a single halfway and at end, rather like Holt's 10-part of Grandsire. Unfortunately it isn't quite a perfect palindrome, as the two 5040s link up a course apart.

 

In 2009, Don Morrison published two intriguing bobs-only peals. One of them is a bicycle three-part, with Queens and Tittums part ends. The other is an extension of the bicycle group to form an 18-part peal. Another 18-part, by F. T. Blagrove has been published, though this does have singles.

 

                                                        20-part peals

I became interested in the idea of a 20-part composition for Bob Triples, where the 20 part ends are the cyclic rotations of Rounds, Queens, Back Rounds, and Tittums, with the 7th in its home position for all part ends. These are the same part ends used for Thurstans' Masterpiece for Stedman Triples, and other compositions in that method. In 2011 I also came up with a peal of Grandsire Triples on this plan.

 

In Bob Triples, I have so far found three part-blocks which have the potential to form 20-part peals, though neither are especially co-operative (as mentioned above, B-block peals of Bob Triples are still a gap in the market, so there is scope for more research here.)

 

Here is the first 20-part block I came up with:

​

  234567

- 235746

  372654

  763425

s 674532

s 765243

  627354

  236475

s 324567

  253746

s 527634

  265473

  642357

- 643725

  476532

  754263

  527346

- 523674

  265437

This should be mutually true to the 20-part ends described above, but it is joining the parts together which is difficult, and this block isn't particularly helpful with offering joining points. And it is difficult to shunt five-part transpositions, as bobs and singls do not affect five bells.

​

Another difficulty for 20-parts of Bob Triples (and some other part plans) is the method's internal falseness. The course starting 1243657 is false against the plain course, as both courses are the same but rung in opposite directions. As can be seen, 24365 is 23456 with two pairs of bells swapped over - 3,4 and 5,6. This means that part plans where two pairs of bells are swapped over offer potential duplication of courses.

​

For example, in a 20-part two of the part ends are 23456 and 43265. But this means that when the course 34256 occurs in the composition, so will 32465 in the same relative place - but both courses are false partners of each other!

​

However, there is a solution to this problem. Thanks to the fact that Bob Triples has six leads in a P-block, it is possible to ring exactly half a course. What's more, you can ring exactly half of 34256, and the same equivalent half of 32465, which produces a true holistic course. To do this you simply call a single when the 7th makes seconds, and a single Home three leads later. You can see in the block above where this happens. It also creates a two-part shunt.

​

As of yet I haven't been able to turn this block into a 20-part as intended. However, I have been able to turn it into a palindromic 10-part, in which the ten parts with Queens and Tittums part ends are rung backwards and super imposed into the Rounds and Reverse Rounds parts which are rung forwards. This is another nice feature of Bob Triples composition: due to the way that the falseness works, the singles, and some of the features described above, palindromic structures can be had very easily. (An interesting six-part palindrome that I composed in late 2010 or early 2011 can be found here.)

​

Here is the palindromic 10-part I came up with, derived from the 20-part block above:

​

5040 Plain Bob Triples

RBP

W        M  H

- sT,sB     s 32456|  |  

  sF          64235|A |

-        -  s 25643|  |

  sF        s 35624|  | C

  sB,sT  -    56324|B |

-        -  s 34562|  |

     3C       62345

      A*      65432

     4C       26543

      A       24356

      B*      64532

      B       23456

In A* and B* the final single is omitted.

Composed 15th June 2018.

​

The following part-block is interesting, though it is one lead short in each part (when the 7th is seconds place bell) meaning that q-sets would have to carefully chosen, allowing for the missing leads to be rung as B-blocks. 

​

  234567

  352746

  573624

- 576432

  745263

  427356

s 243675

s 426537

  254763

  572346

s 753624

  567432

s 654273

  526347

  235764

- 237456

  342675

  463527

I did manage to turn this into a 1008 that would be mutually true for a perfect five-part peal, but it comes round at the first part end, and the options for linking the parts together are cumbersome and not very statisfying.

​

Here is the third 20-part block that I came up with, again on 15th June 2018:

​

  234567

  352746

  573624

  765432

  647253

  426375

s 243567

  452736

  574623

  765342

  637254

s 362475

s 634527

  356742

  573264

s 752436

  547623

s 456372

s 543267

This doesn't have as much character as the first two, but probably more potential for linkage points to other parts, and possibly irregular q-sets.

 

Update: Since writing this, I have now (18/4/19) turned this block into some peal compositions - possibly the first ever 20-part 5040s in the method. By replacing two of the singles at Home with bobs causes the last course to be rung backwards, and so producing a 5-part shunt rather than a 2-part shunt. The 5-part shunt is far more useful for producing peal compositions. The first peal has the 6th as the half-hunt observation (note that the A Block is one of the twenty parts) and the second peal is a rotation with the 4th as half-hunt, producing an arguably neater arrangement. Straight afterwards I discovered that the blocks could be arranged differently, producing a very structured peal not dissimilar to Dexter's no. 2 arrangement of Stedman.

​

                                       Peals on the sB,sH plan

Bob Triples has in-course falseness, namely that the 24365 course is false against the plain course. This is exactly like the B falseness in Treble Dodging Major methods. This does not pose a great problem, as peals can still be composed around the falseness.

 

However, by ringing exactly half of the course in-course, and the other half out-of-course, the in-course falseness can be eliminated completely. You can do this by calling sB,sH in every course, producing a 2-part course with 1243657 the halfway course end. I like to think of these as the Bob Triples equivalents of the John Lancashire s1.s10 courses in Stedman Triples. I had known about these sB,sH courses and their special properties for years, but had never really exploited them in a peal composition (except in my 20-part peals above, where the group of part ends falls foul of the in-course falseness, but the sB,sH courses can come to the rescue and so make 20-part peals possible.)

 

Recently I used the sB,sH courses to produce a diverse range of new compositions. I discovered the beautiful systems that these courses offer, including perfect palindromes.

 

Being only two courses long, the sB,sH block is quite a small building bock for peal compositions. Calling W,sB,sH in every course produces a 5-course building block. Unlike ordinary P-Blocks (plain courses), this block is not palindromic. The result is that when some of these blocks are rung in reverse (which is sometimes necessary), the calling becomes sB,M,sH.This can be clearly seen in my first composition on the plan, where there is a patterned alternating of the forward and reverse blocks. Extra calls at 3rds and 4ths are used to link the blocks together. The same courses are used in my second composition, but instead of using the extra calls to link up the blocks, omits at Home are used.

 

This W,sB,sH block moves the 6th to all relative positions within the block ('transitive' is the posh term for that.) This entails that - if we treat the 6th as the secondary observation bell - the calls at Wrong which don't affect the 6th can be omitted. The resulting block is sB,sH,W,sB,sH,W,sB,sH,sB,sH,W,sB,sH. I used it to produce a peal with more omits than the previous two. Two singles at Home are replaced by bobs to link up the forwards and reverse blocks. Again, a very patterned composition is formed.

 

By calling W,sB,M,sH in every course, you get a palindromic block. This entails that the reverse courses are the same as the standard courses - very neat. But at only two courses long it is a very small building block, which will make repetitive patterns less likely. I used it to compose a 10-part peal with all in-course part ends: 35264, 56342, 64523, 42635, 24365, 46253, 65432, 53624, 32546, 23456:

​

5040 Plain Bob Triples

RBP

W  B  M  H

-  s  -  s 43265

   s  -*   23654

-  s  -# s 63245

-  s  -    26354

   s  -  s 35642

   s  -  s 64523

10 part.

In part 1, call a single at -# to produce -*.

In part 6, call a single at -* to produce -#.

Composed 28th May 2019.

 

It is wonderful that these special courses make in-course 10 parts - and other part plans - possible, as such a group of part ends would be useless for the ordinary P-Block compositions. I have never seen an in-course 10 part peal of Bob Triples before (though such 10-parts are of course two subgroups of my 20-part peals.)

 

I then realised that the earlier block - W,sB,sH - would also work for a 10-part peal, and would have a more repetitive structure. The resulting peal looks extremely neat on paper, the main part (the 'A' section) being completely symmetrical.

 

As is the case with many 10-part peals of Grandsire and Stedman Triples, the two halves of these in-course 10 parts have to be linked up irregularly. But by using the same basic block, a perfect palindromic exact 5-part can be had. A perfect palindromic 3-part (based on a 6-part) is also possible, with a neat alternating of the forward and reverse courses which looks very neat on paper.

 

I then produced a 5-part of St Simon's Bob Triples, where M,sB,W,sH produces a 3-course building block that is different to the ones in Bob Triples, and consequently is not a variation.

 

                                                   Equal-call 5040s

See this page for information about my 5040 containing 120 each of bobs, singles and plain leads.

​

​

​

​

​

​