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

Because of its concentration in eastern Brazil on the ancient, eroded Guyana shield and Atlantic Coastal slopes ranging from Baia to Río Grande do Sul, the section Urospadix  is possibly one of the oldest sections in the genus.

The section is apparently based on 2N = 30, however a number of B chromosomes or fragments have been reported. The chromosome number for A. harrisii (Grah.) Endl. is especially confusing with both fragments and B chromosomes being reported. Anthurium bellum  Schott has counts of 2N = 28 and 56 (Bhattacharya, 1977 and Mookerjea, 1955 respectively). Anthurium lucidium  Kunth with a chromosome number of ca. 124 is the highest known chromosome number in Anthurium.

Little information about crossability in this section is currently available.

XII. CARDIOLONCHIUM Schott (Fig. 24-28)
This apparently natural group is relatively well known horticulturally. It is a medium- sized section characterized by plants with short stems and short internodes and with peduncles and petioles which are often striate or ribbed (Fig. 24 & 25). The leaf blades are generally conspicuously velvety on the upper surface while the tertiary veins on the lower surface are inconspicuous (Fig. 26 & 27). The condition on the lower surface results from the epidermal cells which are raised, sometimes even more or less conical. The velvety species which have been investigated are also distinct chromosomally, all having additional B-chromosomes (small chromosomal fragments) in addition to their regular count of 2N = 30 (Sheffer & Kamemoto, 1976).

Some of the well known examples in this section are A. crystallinum  Linden & André, A. magnificum Linden, A. regale Linden and A. warocqueanum  J. Moore. Anthurium cerrocampanense  Croat and A. Rubrinervium (Link) G. Don Schott (Fig. 28) are aberrant members of section Cardiolonchium lacking velvety leaf blades but having winged petioles and B-chromosomes.

Species long considered to be members of sect. Cardiolonchium , namely A. clarinervium  and A. leuconeurum  Lem. from Mexico may not belong in this section and may indeed belong in a section of their own. Breeding studies by John Banta in Florida have shown that while these species readily interbreed with a number of other Mexican species they do not cross with more typical members of sect. Cardiolonchium . Among Mexican species fruits with viable seed have been obtained for example between A. yetlense  Matuda, once believed to be a typical member of sect. Belolonchium (Croat, 1983) and A. clarinervium  but I now believe that these two species are more perhaps closely related and belong in the same section.

In general where intrasectional interspecific pollinations have been attempted, the species of section Cardiolonchium are intercrossable, and numerous reports have been made of intersectional crosses involving section Cardiolonchium  with sections Belolonchium , Calomystrium and Semaeophyllium . Hybrids between A. cerrocampanense . A. crystallinum and A. regale Linden and A. subsignatum  Schott (Semaeophyllium ) have been made.

12