Integer Sequence A112088
Idea: Add the same number of edges as there are leafs in the previous step.
This sequence has been entered as A112088 in the On-Line Encyclopedia of Integer Sequences.
2, 3, 5, 7, 11, 16, 24, 36, 54, 81, 122, 183, 274, 411, 617, 925, 1388, 2082, 3123, 4684, 7026, 10539, 15809, 23713, 35570, 53355, 80032, 120048, 180072, 270108, 405162, 607743, 911615, 1367422, 2051133, 3076700, 4615050, 6922575, 10383862
1st step: 2
. . . . . . . o o
2nd step: 2, 3
.
. .
. .
. .
. o
. .
. .
. .
o o
Results in 3 leafs
3rd step: 2, 3, 5
. . . . . . . . . . . . . . . . . . . . . o o o o . . . o
Results in 5 leafs
4th step: 2, 3, 5, 7
. . . . . . . . . . . . . . . . . . . . . . . . o . . . . . . . . . . . . . . . . . . o o o o o o
Results in 7 leafs
5th step: 2, 3, 5, 7, 11
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . o o o o o o . . . . . . . . . . . . . . . o o o o o
6th step: 2, 3, 5, 7, 11, 16
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . o o o o o o o o o o o o o o o o
N'th step: 2, 3, 5, 7, 11, 16, ...
Originally I (Simon Strandgaard) submitted this Ruby oneliner:
a, c=2, 4; p Array.new(99){c-=x=(a*4-c+1)/2; c+=2*a*=2 if c<0; x} #=> [2, 3, 5, 7, 11, 16, 24, 36, 54, 81, 122, 183, ...]
And recently Graeme McRae has made this formula for it:
a(1)=2; a(n)=floor((5+sum(a(1) to a(n-1)))/2)
Also recently, I found that it was referenced in MathForum, discussing the Josephus problem. I have never heard about Josephus, and I have no idea if there exists a connection.
Story about A112088
I have always been facinated by things that mimics the prime numbers. While playing with these things I made up this sequence. My friend Johan Gade told me about The On-Line Encyclopedia of Integer Sequences, where this sequence was entered. It felt really good :-)