The ketembilla is a better fruit than its congener the um-kokolo, but the plant is somewhat more limited in its distribution. From its native home in Ceylon it has been brought to the Western Hemisphere, where it may now be found in a few gardens in Florida, Cuba, and California; elsewhere it is little known. Since it is more tropical in its requirements than the umkokolo, it is not suitable for cultivation in the Mediterranean basin, except perhaps in the most favored situations.
In growth and habit the plant is less robust than its congener, although it reaches about the same ultimate height, 15 to 20 feet. The branches are slender, often drooping under the weight of their fruit, and the thorns are long and sharp, but not so formidable as those of the umkokolo. The leaves are lanceolate or oval in outline, acute, entire or subserrate, and 2 to 4 inches long. The fruit is of the same size and form as that of the umkokolo, but maroon-purple in color and more velvety on the surface. The purplish pulp is sweet and luscious, with a flavor resembling that of the English gooseberry, a fruit which the ketembilla suggests so strongly in appearance and character as to give rise to the common name Ceylon-gooseberry. Aberia Gardnerii, Clos., is a botanical synonym.
The plant does not withstand drought as well as the umkokolo, and is injured by temperatures considerably above 20°. While it succeeds in southern Florida, the climate of most parts of southern California has usually proved too cold for it. It likes plenty of moisture, both in the atmosphere and in the soil, and under proper conditions bears enormous crops of its attractive fruits.
The distribution of the sexes is the same in this species as in the umkokolo, and it is, therefore, necessary to insure the proximity of staminate and pistillate plants if fruit is desired. It has been reported that isolated plants of both species are sometimes fruitful, which suggests that they may in occasional instances produce perfect flowers and not require cross-pollination. If plants of such character are found, they should be propagated by budding or grafting, since they would be of considerable value. P. J. Wester reports that shield-budding is successful. He says: "Use petioled, preferably spineless, not too old budwood with tomentum still present; cut buds an inch to an inch and a quarter long; age of stock at point of insertion of buds unimportant." Propagation by seeds is easily effected, as with the umkokolo.