One famous integer sequence is the Fibonacci sequence, which starts with 
a pair of 1s and then each subsequent term is the sum of the two 
proceeding terms.

1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, ...

And the ratio of adjacent terms approaches the golden ratio as the 
number of terms goes to infinity.

Where the golden ratio, denoted phi, is the unique ratio for which a 
segment can be cut into two unequal parts such that the ratio of the 
longer part to the shorter part is the same as the ratio of the whole to 
the longer part. Phi is irrational, with a non-terminating, 
non-repeating decimale expansion, but is approximately 1.618.

As it turns out, any integer sequence where each term is the sum of the 
two terms before it converges to adjacent terms being in golden ratio, 
but the Fibonacci sequence is still special as it gives the best 
rational approximations of phi.

Phi also has the property of being 1 more than it's reciprocal and 1 
less than its square, and any integer power of phi is equal to the two 
powers before it.

Phi shows up a lot in nature, but I'm more interested in its 
generalizations.

One generalization tweaks the definition of the Fibonacci sequence..

If we start with a pair of 1s, treat everything before the 1s as 0s, and 
each term is the sum of the 3 proceeding terms, we get the Tribonacci 
sequence:

1, 1, 2, 4, 7, 13, 24, ...

And the ratio converges to the Tribonacci constant.

Summing four terms at each step gives the Tetranacci sequence:

1, 1, 2, 4, 8, 15, 29, ...

Five terms, Pentanacci:

1, 1, 2, 4, 8, 16, 31, 61, ...

And while I'm not sure how to prove it, the sequence of n-nacci numbers 
and their corresponding ratios seem to converge to the powers of 2, and 
the number 2 respectively, so, if true, one could call 2 the infininacci 
constant.

Another generalization of the golden ratio are the metallic means, for 
which the nth metallicmean has the continued fraction where all of the 
non-1 values are n.

The silver ratio is 1 + the square root of 2 or approximately 2.414, 
which is also the ratio of an Octagon's second diagonal to it's side 
while the golden ratio is the ratio of a pentagon's diagonal to it's 
side. It is an open question if any other metallic means show up in 
regular polygons.

But perhaps my favorite generalization to the golden ratio is the golden 
trisection. I don't fully understand the math, but in a certain sense, 
the golden trisection divides the whole into three unequal parts in a 
manner comparable to the golden ratio splitting the whole into two 
unequal parts.

As it turns out, the values of the golden trisection can be found in the 
lengths of the diagonals of a regular heptagon, and there's at least one 
source that denotes them rho and sigma, though I don't know how 
widespread this notation is, and the golden trisection seems to be 
somewhat obscure to begin with.

There is also a golden quadri-section, one solution for which can be 
found in the diagonals of an enneagon, though I believe it's an open 
problem if this solution is unique and if the pattern continues with 
further odd-sided polygons.

Though it does lead me to wonder if there is a "silver trisection" or 
even a family of metallic trisections. It also leads me to wonder if 
there is anything special about the 1:square root of 3: 2 ratio when 
treated as a trisection, though it appears not only in the diagonals of 
the hexagon, but in the sides of a 30-60-90 triangle, of which there are 
many ways of slicing a hexagon to make.

I also have to wonder if the golden trisection is algebraic or 
transcendental, and considering that a pentagram inscribed in a pentagon 
features several powers of phi, I have to wonder how many of the lengths 
found in either typ of heptagram can be expressed in terms of rho and 
sigma.

And as both the golden ratio and the golden trisection are related to 
the smallest diagonals of regular polygons, I think it worth noting 
that, taking the triangle formed by two adjacent sides of a polygon with 
side length 1 and the shortest diagonal, and applying the law of cosines 
and some simplification, you get.


square of the diagonal = 2 - 2cos(pi-2pi/n)

Though only a few inputs for n result in outputs that simplify nicely 
with the trigonometric identities at my disposal.

Does make me wonder if there are similar formulas for the other 
diagonals of a polygon.