When describing a new subspecies (from a species that does not yet have described subspecies), the nominotypical subspecies remains the “default” subspecies (it does not need to be described separately). It is the taxon to which the nomenclatural type of the species belongs. It therefore has the same species rank name (as another proper name), and authorship of that name. This is a formal nomenclatural matter, unrelated to the objective similarity of these forms or their evolution.
For example, there is a fictional species Insectus mundus Linnaeus, 1768 with no subspecies described. If the fictional Dr. Smith describes the subspecies Insectus mundus americanus, there will be two subspecies:
Insectus mundus americanus Smith, 2024 (a new taxon with a new nomenclatural type and distinctive features) and
Insectus mundus mundus Linnaeus, 1768 (“everything that is Insectus mundus Linnaeus, 1768 and is not Insectus mundus americanus”).
No special description of Insectus mundus mundus Linnaeus, 1768 is needed.
This is approximately how it works.
The very concept of subspecies proposed by Ernst Mayr is not entirely unambiguous. I wrote about it on the forum earlier. But anyway, it is important to realize that taxonomy (division of living organisms into taxa) and nomenclature (rules for naming taxa) are not directly related. There are many alternative (rankless, numerological and so on) concepts of nomenclatures (almost not used practically). In this case, we are talking about the generally accepted binominal nomenclature, the rules of which are described in the Code.