Centrolepis Labill.

Plants annual or perennial. Leaves radical or distichous, eligular. Inflorescences a head consisting of 2 opposite, inequilateral, 1-11-flowered bracts. Flowers hermaphrodite, sessile, if several, alternate and inserted in 1 or 2 rows, flowering in a centrifugal sequence; each flower consisting of 1-3 glumes, 1 stamen and 2-20 carpels. Glumes 1 to 3, hyaline, unequal in length, erose at the apex, monostichous and dorsal, very rarely opposite (ventral and dorsal). Stamens 1, dorsal (in the median) between the carpels and the glumes; anther 1-celled, dorsifixed near the base. Carpels 2 to 20 to a carpophore, sac-like, superposed in 1 or 2 rows, free or connate, the pairs transversal; style one to each carpel, terminal, free or united at the base, twisted or curled at the tips, adaxial, apical part papillose, stigmatic. Fruiting carpels dehiscing with a lengthwise, lateral (outward) slit.

Lowland up to c. 4000 m, in marshy places, humid sandy grounds, open heaths, and on rocks, in Malaysia exclusively above 2100 m.I have examined and photographed the two mountain species in North Sumatra and Central Celebes in the field and have observed that both start as small pin-cushions which may occur scattered, but in other cases fuse together in a cushion or turf, which may be later invaded by seedlings of other plants (Fig. 3Fig. 4). Isolated tufts will grow and increase in size but in reaching a diameter over c. 10-20 cm, the central part of the cushion will decay (or mummify) into a bare hole, thus bringing about the remarkable ring-shaped cushion of which a specimen is shown in Fig. 5. I have observed the same phenomenon in several other high mountain plants in Malaysia, notably Oreobolns and Monostachya, with which Centrolepis is often associated in Malaysia. It occurs also in other caespitose plants in Europe and is obviously due to the polarity of the stems causing centrifugal growth with natural decay of the older, central (distal) parts. This puts the cause of the formation of the rings on the same level with those popularly known as 'fairy circles' in toadstools, rather than assuming that the central part of the plant decays because in this portion either necessary trace minerals are exhausted or the decayed matter contains self-poisoning substances. Field experiments should be made to ascertain the non-poisonous nature of the bare central part in the rings. — Van Steenis.