I'm attaching the graphical output from searching the genomic sequence with
the longest pogo sequence (X59837). It looks (satisfyingly) as we had found - the
~2.1 kb are the largest with a bunch of elements between 1.1 and 1.5 kb
having different internal deletions, and lots and lots of tiny elements
around 0.2 kb.
I've not looked in detail at the sequence alignments - we did find some
polymorphisms between individual elements anyway. The apparent
discontinuity around 250-300 is not real - just BLAST taking out a region
of relatively simple sequence (mostly a run of T).
So I guess the 2.1 kb sequence can probably be called complete. David
Finnegan did take these clones and express the encoded protein and showed
that it bound the ends of pogo:
Wang H, Hartswood E, Finnegan DJ (1999) Pogo transposase contains a
putative helix-turn-helix DNA binding domain that recognises a 12 bp
sequence within the terminal inverted repeats. Nucleic Acids Res  27:455-61 .
Kevin