OFFSET
1,1
COMMENTS
Note that the starting point of these aliquot sequences are not in increasing order, since for instance we have: 392->463->1 and 324->523->1, that is, with 392>324 while 463<523.
One can observe that the "ever increasing aliquot" part in the definition is not really necessary. A prime is in the sequence if there is an abundant number whose sum of proper divisors results into this prime. So sequence could also be defined as: Primes resulting from summing up the proper divisors of an abundant number. - Michel Marcus, Jan 05 2014
If we try to build the revert sequence listing the starting points of the aliquot sequences, we would get the following terms in increasing order 324, 392, 784, 800, 2304, 2450, 2704, 3600, 3872. But then for n=5352, we'd hit a sequence that begins 5352->8088->12192->20064 and keeps rising to a point where the factors of the last known term are not known. Then later, there are several other such aliquot sequences like 9336->14064->22392 or 10344->15576->27624 that have the same behavior. Thus the only sure terms of the revert sequence would be the terms listed earlier. - Michel Marcus, Jan 11 2014
LINKS
Michel Marcus, Table of n, a(n) for n = 1..100
Factordb, Sequence starting with 14952
Michel Marcus, Aliquot sequences leading to these primes
Help needed for a sequence Thread on Aliquot Sequences Forum
EXAMPLE
The aliquot sequence that begins with 10712 is always increasing before reaching prime 12451: 10712->11128->11552->12451->1, hence 12451 is in the sequence.
20422951 also belongs here with the aliquot sequence that starts at 14952, so a 13-tuple abundant (see factordb link).
People at the Aliquot Sequences project have found longer sequences that reach higher primes.
PROG
(PARI) prev(n) = {for (i=1, n, if ((sigma(i) - i) == n, return (i)); ); return (0); }
lista(nn) = {forprime(p=2, nn, if (prev(p), print1(p, ", "); ); ); } \\ simplified by Michel Marcus, Jan 11 2014
CROSSREFS
KEYWORD
nonn,changed
AUTHOR
Michel Marcus, Dec 31 2013
STATUS
approved