Portulaca
Description
Mostly succulent, copiously branched herbs.
Leaves linear to orbicular, in most species with axillary hairs.
Flowers in (1-)2-30-flowered, terminal capituli;
Sepals occasionally keeled or hooded, persistent or caducous with the petals, stamens and style.
Petals 4-6(-8), mostly obovate, occasionally emarginate or mucronate.
Stamens 4-∞, in 1 whorl.
Ovary half-inferior, occasionally apparently inferior;
Capsule with a caducous circumscissile operculum.
Seeds ∞ (in some Australian spp. 1-4).
Distribution
worldwide present
Mainly tropical and subtropical all over the world, not more than c. 40 spp., possibly several very polymorphous, a few worldwide anthropochorous weeds.
Morphology
In my opinion () the capitulum is a condensed cyme. It is surrounded by 3-18 involucral leaves, not to be confused with the "Involukralblätter" in the sense of VON POELLNITZ. The nerved scales between the flowers are considered bracts and bracteoles and have, like the vegetative leaves, axillary hairs. In P. quadrifida the small white tubercles at the base of a tuft of hairs are supposed to represent the bracts and bracteoles.
The axillary hairs are mostly homologized with stipulae, but in my opinion (l.c.) they are probably not of stipular nature. They are mostly found in 2 tufts in the leaf-axils. In Malesian spp. they are confluent at the base. In P. quadrifida they are present in a whorl around each node.
The axillary hairs are mostly homologized with stipulae, but in my opinion (l.c.) they are probably not of stipular nature. They are mostly found in 2 tufts in the leaf-axils. In Malesian spp. they are confluent at the base. In P. quadrifida they are present in a whorl around each node.
Taxonomy
In my precursor (l.c.) I have modified and simplified the infrageneric subdivision of the genus. Of the two subgenera, one is confined to Australia (subg. Portulacella).
As to the species concept, the one adopted here is distinctly larger than currently accepted, including that used by VON POELLNITZ and LEGRAND. As I have shown, and experimentally checked in all four genera, selfing in bud seems the rule in the family, which explains that in nature pure lines are formed, and that populations may consist of several pure lines keeping distinct for the small characters in which they differ. For those who want to give a name and rank to any constant difference there is hence ample opportunity for distinguishing varieties, subspecies, and even species, especially in variable complexes such as those of P. olerocea and P. pilosa. In view of this genetical situation it seems that according to a normal specific concept there are far less good species than distinguished by VON POELLNITZ and LEGRAND who recorded some 130 for the world and c. 62 (+21 varieties) for the Americas only respectively.
As to the species concept, the one adopted here is distinctly larger than currently accepted, including that used by VON POELLNITZ and LEGRAND. As I have shown, and experimentally checked in all four genera, selfing in bud seems the rule in the family, which explains that in nature pure lines are formed, and that populations may consist of several pure lines keeping distinct for the small characters in which they differ. For those who want to give a name and rank to any constant difference there is hence ample opportunity for distinguishing varieties, subspecies, and even species, especially in variable complexes such as those of P. olerocea and P. pilosa. In view of this genetical situation it seems that according to a normal specific concept there are far less good species than distinguished by VON POELLNITZ and LEGRAND who recorded some 130 for the world and c. 62 (+21 varieties) for the Americas only respectively.