1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Anthurium antioquiense  Engl., A. amnicola Dressler and others once believed to be members of of section Porphyrochitonium  have proven to be members of sect. Calomystrium  but warrant recognition as a distinct series, ser. Rupicola Croat (Croat & Whitehill, in prep.). This small group of riparian species differ in having more or less lanceolate leaves. Also included in this series are A. antrophyoides Killip, A. sytsmae Croat, A. rupicola Croat and two undescribed species from South America, A. werfii and A. palacioanum .

XV. BELOLONCHIUM Schott emend Engl. (Fig. 35-39)
This section is the most poorly known of all sections included in Engler's revision of Anthurium  and would appear to be a "dumping ground" for all cordate species not clearly assigned to other groups with cordate leaves such as sections Cardiolonchium  and Calomystrium .

There are at least two groups of seemingly related cordate species currently placed here and the section may ultimately have to be divided into several distinct sections. In one group the leaves are relatively coriaceous and dry brown. Examples of this group are A. bogotense  Schott, A. brownii  Masters, A. concinnatum Schott, A. gualeanum  Engl. (Fig. 35), A. incurvatum  Engl., A. rigidifolium Engl. and A. variegatum  Sodiro (Fig. 36). Quite a number of the species in this group occur at relatively high elevations and bear more or less hooded spathes (Fig. 37-39). These species fit Schott's original circumscription of the section. Still another group has thinner leaf blades which dry green. These include A. dolichostachyum Sodiro, A. lancetillense  Croat, A. ovandense  Matuda and A. umbrosum  Liebm.

Most species in the section Belolonchium have 2N = 30 chromosomes though one species (A. patulum Sodiro) was reported as having 28 + 1f chromosome (Sharma & Bhattacharya, 1976) in the diploid condition. Still another species believed to be in this section (A. supianum  Engl.) is reported to have ca. 90 chromosomes (Sheffer & Kamemoto, 1976).

Two species reported for this section by Sheffer & Croat (1983) were erroneously placed in this section. Anthurium cerrocampanense  Croat with 2N = 30 + 2B chromosomes is best placed in section Cardiolonchium  while A. signatum  C. Koch & Mathieu with 2N = 34 is best placed in section Semaeophyllium .

Little documented data is available about hybridization within the section. Based on fruit set only, intrasectional pollinations and intersectional pollinations between Belolonchium and Cardiolonchium or Calomystrium are in general successful.

XVI. SEMAEOPHYLLIUM Schott (Fig. 40-42)
This section as treated by Schott contained only the deeply 3-lobed species and would have appeared to be a relatively natural group. Engler's realignment, based on the shape of the inflorescence, is, in my opinion, an unnatural one. The section has species with generally short stems with short internodes. The blades are deeply 3-lobed but entire, not divided into leaflets. Examples include A. falcatum Sodiro, A. furcatum Sodiro (Fig. 40), A. garagaranum Standley, A. grexavium Madison, A. insigne Masters, A. rimbachii Sodiro, A. signatum  C. Koch & Bouché, A. truncicolum  Engl. (Fig. 41), A. tilaranense  Standley, A. tridigitatum Engl. and A. subsignatum  Schott (Fig. 42).

14