The species database/system was created at a time when there were not many specimens in the Essig specimen database. Even now a relatively small percentage of species in the museum are represented by digitized specimen records. The species database represents a list of species in the physical museum. People wanted to keep track of the species in the museum and not wait 10 or 30 years until the 6+ million specimens are databased. This list is used for loans and for tracking other information in the physical museum. It's also connected to the specimen database. The original species list was based on Excel files that a number of people in Essig created over the years. There was no other system of tracking this information before that that I know of. (Cheryl or Gordon can correct me if I'm wrong.)
When a record is added/edited in the specimen database, users are forced to enter the name into the species table if it's not already there. When a species name is edited in the species table, there is a warning message and a link to any existing specimens that may have the same name (genus and species). The specimen names are not changed automatically because such changes may be incorrect based on the circumstances.
The species list is also used to automatically fill in higher taxonomy for specimen records, in cases where a species name has already been entered into the system.
The communication between the specimen data and species list is not perfect. (Author name changes do not trigger any changes/messages between the two tables.) Quite are few name discrepancies were introduced during bulk uploads (which are not officially supported for the Essig database but have been done on a case-by-case basis in the interest of saving time).
Also there are names in the species table of species not in the Essig museum (because we have specimens in the specimen database from multiple institutions). Such names are tagged as not being in the Essig museum and are displayed differently (or not displayed at all, depending on who is viewing the records).