35.1. A new name or combination published on or after 1 January 1953 without a clear indication of the rank of the taxon concerned is not validly published.
35.2. A new name or combination published before 1 January 1953 without a clear indication of rank is validly published provided that all other requirements for valid publication are fulfilled; it is, however, inoperative in questions of priority except for homonymy (see Art. 53.5). If it is a new name, it may serve as a basionym or replaced synonym for subsequent combinations or avowed substitutes in definite ranks.
Ex. 1. The groups "Soldanellae", "Sepincoli", "Occidentales", etc., were published without any indication of rank under Convolvulus L. by House (in Muhlenbergia 4: 50. 1908). The names C. [unranked] Soldanellae, etc., are validly published but they are not in any definite rank and have no status in questions of priority except that they may act as homonyms.
Ex. 2. In Carex L., the epithet Scirpinae was used in the name of an infrageneric taxon of no stated rank by Tuckerman (Enum. Meth. Caric.: 8. 1843); this was assigned sectional rank by Kükenthal (in Engler, Pflanzenr. 38: 81. 1909) and may be cited as Carex sect. Scirpinae (Tuck.) Kük. (C. [unranked] Scirpinae Tuck.).
35.3. If in a given publication prior to 1 January 1890 only one infraspecific rank is admitted, it is considered to be that of variety unless this would be contrary to the statements of the author himself in the same publication.
35.4. In questions of indication of rank, all publications
appearing under the same title and by the same author, such as
different parts of a flora issued at different times (but not
different editions of the same work), must be considered as a
whole, and any statement made therein designating the rank of
taxa included in the work must be considered as if it had been
published together with the first instalment.