Rose and others;
Tom Eichhorst (who is either away, or too modest to speak up!!) has
written a very good article on how to describe a new species, including
how to chose the appropriate Latin ending! You can find it at http://manandmollusc.net/Naming.htm.
Regarding how many specimens are required, i don't think there is any
hard and fast rule: If you use any less than three, you are treading on
thin ice, but even today, if you have a fully adult specimen which
doesn't look in any way like a freak or abnormal beastie, and if it is
different enough from any similar species you can find the types for, it
may be worthwhile to publish it - but this would be an exeption, not the norm!!
Full-fledged summer here in Bluenose central!!,
Ross.