Hypericum, , Nurk & Blattner, 2010

Robson, Norman K. B., 2016, And then came molecular phylogenetics-Reactions to a monographic study of Hypericum (Hypericaceae), Phytotaxa 255 (3), pp. 181-198 : 191

2. Hypericum L. subgenus Brathys (Mutis ex L.f.) N.Robson, comb. & stat. nov.

Basionym:— Brathys Mutis ex Linnaeus (1782: 43 , 268). Type: Hypericum juniperinum Mutis ex Kunth (≡ Brathys juniperina Mutis ex L.f.).

Dark hypericin-containing glands absent. Stem 4–6- (rarely 2-) lined, not terete. Perianth and androecium 5–4(–3)- merous. Stamen fascicles all united in a ring, except when stamens are few or single. Placentation incompletely axile to usually parietal.

This includes sects. 20, 29, 30, Triadenum and Thornea .

As stated above, the ‘ Elodes ’ syndrome has appeared in several independent evolutionary lines in both subgenera: Santomasia and Adenotrias from sect. 1. Campylosporus ; Elodes from sect. 27. Adenosepalum subsect. Caprifolia ; Triadenum from sect. 20. Myriandra ; Lianthus possibly from sect. 9. Roscyna near H. przewalskii . This syndrome is therefore not plesiomorphic, but would seem to be the result of ‘evolutionary recall’ by the operation of a switch gene in response to pollinators.

Sections 25. Adenotrias and 28. Tripentas were retained in the monograph and their species therefore have accepted names in Hypericum , viz. H. aegypticum L., H. russeggeri and H. aciferum in sect. 25 and H. elodes L. in sect. 28; and Santomasia was returned to Hypericum in Part 9 in sect. 1a. Santomasia as H. steyermarkii . But those taxa having flowers with the Elodes syndrome that were originally excluded from the monograph should now be included. All species already have names in Hypericum ( Table 2). However, two new combinations at sectional level in Hypericum are necessary to accommodate these taxa.: