How large is the difference between ACO and GAs

Hi all,
Before I write anything - this post is not intended to start a flame
war.GAs are an interest of mine, and I am not very experienced in this
field, so please be kind on the responses. I was reading an article on
DDJ about Ant colony optimizations
http://www.ddj.com/dept/ai/191800178?pgno=1 and from what the author
has mentioned regarding the various parameters that make ACO work, it
looked to me that most of them can be mapped to a typical GA scenario
(I am not very clear about how exactly GAs and GP differ, but that is
for another post). For example, from the article, the author mentions
that the pheremone levels are slowly evaporated - this being done so
that a particular route which is taken releatively lesser number of
times doesn't get chosen because of the residue pheremone levels. This
is what I think, mutation achieves in GAs. Also, the greedy approach of
the ants to choose a particular path, can also be done via reproduction
and crossover.
Is it that there is a very thin line between ACOs and GAs ? And if so
how does one identify the line ?

Thank You
Gangadhar

.