Cactaceae Juss. sec. APG (2009)

Warning message

This concept of the taxon Cactaceae Juss. sec. APG (2009), Gen. Pl.: 310. 1789 is not contained as an accepted taxon in the currently chosen classification, but in this one: Caryophyllales

Cactaceae Juss. sec. APG (2009)


Cactaceae Juss. sec. APG (2009), Gen. Pl.: 310. 1789



Cactaceae comprise about 120 to 130 genera and some 1450 to 1870 species (Hunt 2006; Nyffeler & Eggli 2010b). Most species are highly modified perennial stem succulents which conserve water to survive temporary dry periods. Only some two dozen species of the genera Pereskia, Pereskiopsis and Quiabentia have a shrubby or tree-like habit with more or less fleshy leaves. All species of the family bear characteristic spine clusters (i.e., areoles), representing short shoots with leaves transformed into spines already at the stage of primordia. Some taxa are spineless and even lack areoles at maturity but all species bear areoles as seedlings. This characteristic is a true synapomorphy of the entire family.
Cacti are native to the Americas, except for the widely distributed Rhipsalis baccifera (Sol.) Stearn that also occurs in tropical Africa, Madagascar, and on islands in the Indian Ocean. Several species from different lineages have been introduced worldwide as crop plants or ornamentals and have become naturalized, and are classified as invasive aliens in several areas, including Australia, Southern Africa, and the Mediterranean.
For a long time in the past, the classification into genera and suprageneric groups was based on form characteristics of vegetative and reproductive structures, culminating in the fine-grained classifications of Backeberg (1958-1962, 1966) or Buxbaum (1962) and Endler & Buxbaum (1974). Many of the highly modified structural features are associated with the succulent life strategy (e.g., Nyffeler & al. 2008), and hence provide particular challenges in the interpretation of a classification based on purported relationships. The consensus classification initiative as reported by Hunt & Taylor (1986) and subsequent papers helped to overcome the deviating systems used in the second part of the 20th century, but also fell short in not being based on further and expanded data sets of comparative data for reconstructing relative relationships.
However, the molecular phylogenetic studies (see the introduction and Nyffeler & Eggli 2010b) provide the base for an increasingly stable backbone classification for major suprageneric clades. At the same time, unexpected novel placements are suggested by such studies for several species or genera, such as Blossfeldia (Nyffeler 2002) or Lymanbensonia (Korotkova & al. 2010), while long-established genera, such as Echinocactus and Ferocactus but also Mammillaria have been found to be polyphyletic (Bárcenas & al. 2011; Hernández-Hernández & al. 2011; Vázquez-Sánchez & al. 2013). To use these findings for updating the generic classification of the family is a pronounced challenge (Hunt 2006; Nyffeler & Eggli 2010b).