[r-t] Method symmetry

[Mark Davies]

Leaving the field of "row-pivot" symmetries for a while, how about this 
little number:

-1-1-3-1-3-1-1-3-1-1-1-5-1-1-1-1 = 31276854

It is a Differential with (3,5)-tuples at the leadhead, generating a 
15-lead plain course. The bluelines for both sets of bells are 
conventionally symmetric, but the method as a whole is not.

It would seem (unless I'm missing something) that, to achieve symmetric 
bluelines within an asymmetric method, you need to split the bells up 
into at least two groups. Then they can have their own independent 
reflection points, out of phase with each other. A 3,5-differential 
seemed like a natural and pleasant way to achieve this.

MBD