Hey JT-
Nice opening post, and definitely a point of worthy discussion I haven't seen mentioned in the forum before.
One thing you have to consider when you think about the evolution of a biological process or structure, (in this case linear chromosomes), is the simplest form of that process, and not the process as it exists in a mammalian cell.
In this case, linear chromosomes exist in some viruses and bacteria, without the complex telomere maintenance you describe above.
In some cases, repeated and palindromic sequences at the end of these chromosomes fold into complex hairpin loop structures (they fold back on themselves and bind complementary sequence) to protect free strand ends. (Searching for "linear bacterial chromosome" at
http://www.pubmed.org should give you lots of interesting abstracts to look at).
Thus it appears that the sequence itself is sufficient to protect the end of a linear chromosome in some situations, without any protein involvement.
In certain bacteria, the chromosomes are unstable/reversible, in that they commonly go from circular to linear and back again.(
reference) While such instability is negative in one sense, it also serves as a mechanism to "accelerate" evolution, since it provides more opportunities for rearrangements and duplications.
Also, the fact that many simple linear chromosomes use palindromic hairpin loops as simple telomeres supports a plausible course of evolution for the linear chromosome. The palindromic sequence could have arisen within the stability of the circular chromosome (perhaps from a small duplication event), allowing for potential hairpin loops to form. Subsequent breakage of the circular chromosome near such loops would result in a linear chromosome having a "telomere"
before it was even linear...
There are still many points that deserve to be discussed from your opening post, but complex telomere maintenance does not appear to be required for linear chromosomes.