Campanula jacobaea var. hispida Bolle

Gardère, Mathieu L., Florence, Jacques, Muller, Serge, Savriama, Yoland & Dubuisson, Jean-Yves, 2021, Codonographia Gorgonum, or the description of a pleiad of bellflowers (Campanula, Campanulaceae) from the Cabo Verde archipelago, Candollea 76 (1), pp. 13-40 : 36

Campanula jacobaea var. hispida Bolle in Bonplandia 9: 51. 1861.

Typus: CABO VERDE. Santiago : “loco natali specialiore ignoto”, s.d., Bocandé s.n. (not found) .

Notes. – We transcribe here the note of BOLLE (1861: 51) on the variety hispida : “ Hab. in insula Santiago , loco natali specialiore ignoto. Inter reliquias B. Bocandé inventam e pessimis exemplis imperfecte tantum novimus. Fieri potest, hanc propriam esse speciem; eximia setositate florumque exiguitate habitu saltem a stirpe Webbiana, quae probabiliter et Smithiana, magnopere recedit; tamen ab hac eam separare non ausi sumus ”; and we give here-after this loose translation: “Grows on the island of Santiago , the exact locality is unknown. We only know it [ var. hispida ] imperfectly from poor specimens of B. Bocandé. It may be a particular species. By the extreme hairiness of the flower and the weakness of the habit, it differentiates itself strongly from that of Webb and is probably that of Smith; however, we dare not separate it from that [ C. jacobaea ]”.

BOLLE (1861) described the var. hispida for the island of Santiago based on the Bocandé’s collection, currently untraceable. The specimens from B used by Bolle for the description of his varieties were destroyed in the fire of Berlin in 1943 ( HIEPKO, 1987) and probably alongside those of Bocandé. LEYENS & LOBIN (1995) treated the variety as a synonym of C. jacobaea but the same authors saw and identified a specimen of Bocandé conserved at “D” as C. bravensis . However, “D” does not correspond any indexed herbarium ( THIERS, 2019) and it is probably a typing error (W. Lobin, pers. comm.). The only known European Herbaria to hold Bocandé’s collections are B, FI-W and S (M.L. Gardère, unpubl. data) but S is temporarily closed for renovation. We were unable to check this specimen and to confirm or not the identification of LEYENS & LOBIN (1995). The description of Bolle is insufficient to place the variety hipida among the CVB species. Given that Santiago harbors both C. jacobaea and C. bravensis , we prefer a precautionary approach and rather consider it to be a nomen dubium.