Mark,
What you say certainly has validity.  More than once the single
specimen used to describe a species has later turned out to be a
form, or even a wierd individual variant (a "freak", so to speak) of
an existing species.  On the other hand, requiring the collection of
ten specimens before naming a species could take many years, perhaps
even a century or more in some cases.  Such current or former
rarities as Conus gloriamaris, Conus cervus, Cypraea valentia,
Cypraea leucodon, etc. were known from either a single specimen or
very few specimens for many years before the 10th specimen was
discovered.  Personally I would like to see at least a second
specimen emerge before a new species is described, but even that is
not always the case.
Paul M.
Rhode Island