Having worked with digitized, microform, photocopied, and original versions of
16th-18th century documents in the AGI in Seville, I would advise that although
digitizing the microform versions of manuscript documents (by whatever
technique) might indeed provide you with a roughly useful copy of the original
documents for use in illustrations, etc., my own experience is that such
digital copies would be considerably less useful from a paleographic
perspective than the microfilm itself (which in turn is less useful than the
original document).  The AGI has recently developed an extensive collection of
their documents in digital format, all of which were scanned in directly from
the original manuscripts in the AGI.  While there are many advantages to this
procedure (ease of access, computer enhancement of images, protection of the
originals, etc.), important details in the handwriting are frequently lost due
to the "graininess" of the magnified image, despite the fact that the original
itself was scanned.  When I work with such digital copies (and printouts from
them), I almost always end up having to consult the original to clarify
paleographic details.
 
Although I realize that access to originals is only rarely possible for
researchers, in David's case I would recommend verbatim transcription of these
Jamaican documents directly from the microfilm.  Though somewhat
labor-intensive and time-consuming, these transcriptions would almost certainly
be more accurate than those which would inevitably have to be done by other
researchers from any digital copies of the microfilm version.
 
If I had to rank copy media for "faithfulness" to original manuscripts (my
opinion only) it would be as follows, from best to worst: microfilm from
original, good photocopy from original, digital version from original, and then
all other "2nd-generation" copies of these copies.
 
John Worth
Fernbank Museum of Natural History