Table 2
n=6 | 2,2,2 | | |
|
| | | |
|
n=7 | ~3 *,2,2 | | |
|
| | | |
|
n=8 | ~4 *,2,2 | 2,2,2,2 | |
|
| ~3 *,~3 *,2 | | |
|
| | | |
|
n=9 | ~5 *,2,2 | ~3 *,2,2,2 | |
|
| ~4 *,~3 *,2 | | |
|
| ~3 *,~3 *,~3 * | | |
|
| | | |
|
n=10 | ~6 *,2,2 | ~4 *,2,2,2 | |
|
| ~5 *,~3 *,2 | ~3 *,~3 *,2,2 | |
|
| ~4 *,~4 *,2 | ~3 *,2,~3 *,2 | |
|
| ~4 *,~3 *,~3 * | | |
|
| | | |
|
n=11 | ~7 *,2,2 | ~5 *,2,2,2 | ~3 *,2,2,2,2 |
|
| ~6 *,~3 *,2 | ~4 *,~3 *,2,2 | |
|
| ~5 *,~4 *,2 | ~4 *,2,~3 *,2 | |
|
| ~5 *,~3 *,~3 * | ~3 *,~3 *,~3 *,2 | |
|
| ~4 *,~4 *,~3 * | | |
|
| | | |
|
n=12 | ~8 *,2,2 | ~6 *,2,2,2 | ~4 *,2,2,2,2 | 2,2,2,2,2,2
|
| ~7 *,~3 *,2 | ~5 *,~3 *,2,2 | ~3 *,~3 *,2,2,2 |
|
| ~6 *,~4 *,2 | ~5 *,2,~3 *,2 | ~3 *,2,~3 *,2,2 |
|
| ~6 *,~3 *,~3 * | ~4 *,~4 *,2,2 | |
|
| ~5 *,~5 *,2 | ~4 *,2,~4 *,2 | |
|
| ~5 *,~4 *,~3 * | ~4 *,~3 *,~3 *,2 | |
|
| ~4 *,~4 *,~4 * | ~4 *,~3 *,2,~3 * | |
|
| | ~3 *,~3 *,~3 *,~3 * | |
|
|
Now we may calculate the number of generating links and
the number of all links derived from 2,2,¼,2
(Table 2). Every class, where ~n1*
occurs k1 times, ~n2*
occurs k2 times,¼,
~nm*
occurs km times, consists of
Õi = 1m |
æ
ç
ç
ç
è |
|
ki |
|
ö
÷
÷
÷
ø |
= Õi = 1m |
æ
è |
|
fq+ki-1
ki |
|
ö
ø |
|
|
generating links, where q=ni-2.
Replacing each ~ni*
by ni * , and every
~d * (ni)
by d * (ni)
(1 £ i £
m), we could enumerate all links of this class.
The derivation of generating links from the direct products
2,2,¼,2+ and 2,2,¼,2++
completely follows the preceeding derivation from 2,2,¼,
2. To obtain from them all links for n fixed (n £
12), we need to replace in the generating links obtained every ~k
* by k *
, and every ++ by 3,4,5,... pluses. Hence, for every n (6 £
n) we have the same number of prismatic (n+p)-crossing
links with p pluses (p ³
0).
According to this, the structure of S-world is
the following: all links of this world are derived from three infinite
classes of basic links or their direct products (S-, S×O-
and S×L-links) by R *
-replacements, resulting in SR*
-, SR * ×O- and
SR * ×L-subworld.
We could also analyze infinite families, representing
the "vertical" structure of every world, where each family is obtained
from certain generating link. For example, considering generating links
from the first column of Table 2, we obtain the infinite families:
p,q,r (p ³
q ³ r ³
2), p1,q,r (p ³
2, q ³ r ³
2), pq,r,s (p,q ³
2, r ³ s ³
2), p1,q1,r (p ³
q ³ 2, r ³
2), pq1,r,s (p,q ³
2, r ³ s ³
2), p1q,r,s (p ³
q ³ 2, r ³
s ³ 2), p13,q,r
(p ³ 2, q ³
r ³ 2), pq,r1,s
(p,q,r,s ³
2), p1,q1,r1 (p ³
q ³ r ³
2), etc.
Different combinatorial problems arise from here. For
example, for every n fixed we could try to calculate the number of different
classes occuring in the first column of Table 2 (beginning by series 1,
1, 2, 3, 4, 5, 7 for n £ 12), or the number
of generating links occuring there, or even the number of all links derived
from some generating link for every n fixed, etc. For example, the number
of different classes mentioned is equal to the coefficient corresponding
to qn-6 in [n-1,3], where by [n,r]
is denoted Gauss polynomial [36].
[n, r] = |
(1-qn)¼(1-qn-r+1)
(1-qr)¼(1-q) |
. |
|
All such problems belong to the theory of partitions with
a given symmetry group (P-partitions).
Let be given a permutation
group P of k objects and number n (n
Î N, n ³
k). To every object ki (1 £
ki £ k) is assigned
a natural number ni, where åi
= 1k ni = n. Two partitions
defined by assigned objects are equal iff there is a permutation
from P transforming one to another; we need to find and enumerate
different P-partitions. In some special cases, we could reduce problems
of P-partitions to the classical partition theory, but in
general this is an open question.