Aspidostemon

Werff, Henk Van Der, 2006, A revision of the Malagasy endemic genus Aspidostemon Rohwer & Richter (Lauraceae), Adansonia (3) 28 (1), pp. 7-44 : 8-9

publication ID	https://doi.org/ 10.5281/zenodo.5186799
persistent identifier	https://treatment.plazi.org/id/03AE87F6-5D57-FFC5-167C-7630FCC5FB52
treatment provided by	Carolina
scientific name	Aspidostemon
status

Treatment

CHARACTERISTICS OF ASPIDOSTEMON View in CoL

The differences in wood anatomy between Aspidostemon and Cryptocarya are presented in Rohwer & Richter (1987) and Richter (1990) and will not be repeated here. The most obvious morphological differences are opposite leaves and flowers with 3 or 6 stamens in Aspidostemon vs. alternate leaves and flowers with 9 stamens in Cryptocarya . Tepals and stamens persist on top of the fruits in Aspidostemon , but are deciduous in Cryptocarya . The stamens of Aspidostemon are all quite similar, regardless whether there are three or six; the staminodia vary in shape and size from whorl to whorl, but are uniform within a whorl. I consider A. parvifolium (Scott-Elliot) Van der Werff as the most primitive species; it is the only member of the genus with the tepals spreading at anthesis and in which the outer 6 stamens have a distinct, short filament ( Fig. 13C View FIG ), similar to the condition found in Cryptocarya . In all other species of Aspidostemon the tepals are erect at anthesis and the stamens do not have a distinct filament. In some species the stamens have become tepaloid, similar in shape and texture to the tepals ( Figs 2B View FIG ; 7A View FIG ). In these species the locelli are located on the inner face of the stamens (introrse).Tepals are about as wide as long in these species. In another group of species the stamens become shorter and have a flat, thick apex; locelli move in this group upwards and become apical, rarely lateral, but do not remain introrse ( Figs 2C View FIG ; 7C View FIG ). The stamens appear thus as short, broad stubs with apical locelli. In the staminodia one finds a parallel change. In A. parvifolium the staminodia of the third whorl (staminodia III) are columnar, with a somewhat enlarged tip and are not united. A fourth whorl, staminodia IV, is present, stipitiform, pubescent and can be difficult to find. In the other species the staminodia III have a progressively larger, flat apex; these become tightly pressed together and may become fused, forming a shield-like structure that completely covers the ovary and sometimes part of the stamens ( Fig. 2A View FIG ). As the staminodia III become larger, there is less space for the staminodia IV and these disappear or become fused with the staminodia III. Finally the stamens of whorl II also become staminodial, tightly pressed against staminodia III and difficult to recognize ( Figs 11D View FIG ; 13D View FIG ). Rohwer & Richter (1987) observed that in some species the staminodia II have rudimentary locelli, which are visible as light-coloured spots, sometimes with flaps, but smaller than those of the stamens. I noticed in three species that some flowers appear to have three stamens and others, on the same specimen, six stamens. My interpretation of this is that in young flowers only the outer three stamens open and that only in mature flowers the outer six stamens have opened locelli. Most species with free staminodia III have 6 stamens ( Figs 9B View FIG ; 14B View FIG ) and most species with fused staminodia III have 3 stamens ( Figs 9C View FIG ; 16D View FIG ), but there are a few exceptions to this general rule. The flower shape is variable as well: in the species with free staminodia III flowers are more or less globose or elongate ( Fig. 4C View FIG ), but in some (though not all) species with fused staminodia III the flowers become depressed globose to mushroomshaped ( Fig. 4D View FIG ). In other species the flowers are narrowly to broadly cone-shaped. Through all these changes in the androecium, the pistil remains more or less the same: spindle-shaped, glabrous, sunk in the hypanthium, but not fused with it. Fruits are not known for all species and, to the degree they are known, tend to be uniform, ellipsoid, with a small crown of floral parts ( Figs 11A View FIG ; 16A View FIG ). An inconspicuous character, but one worth noting, is the presence of bracts covering the inflorescence buds. As the buds elongate, the bracts fall off, but leave scars at the base of the inflorescences. Cryptocarya species do not seem to have these bracts.

The most important characters used for species delimitation are those of the flowers: number of stamens, shape of the stamens and staminodia, shape of the flowers, and indument of terminal buds, twigs and inflorescences. There are also differences in leaf shape and texture between the species, but these differences are on the whole not sufficient by themselves to recognize individual species.Therefore, sterile or fruiting specimens cannot be identified with much confidence.

All species of Aspidostemon have opposite leaves. Unfortunately, species exhibiting this character occur in other genera of Lauraceae on Madagascar as well and this can create confusion. There is one group of species of Beilschmiedia Nees with equally strictly opposite leaves ( Van der Werff 2003). Of course these species differ in flower and fruit characteristics from Aspidostemon , but the identification of non-flowering specimens has led to confusion (see note under A. scintillans (Kosterm.) Rohwer ). In general the Beilschmiedia species with opposite leaves have leaves with a dense, raised reticulation and black twigs. A few species of Ocotea Aubl. have subopposite leaves, with the distal leaves opposite and the older leaves becoming alternate, a character that allows identification to genus. One species of Cryptocarya , C. louvellii Danguy , reportedly has opposite leaves. The holotype (Louvel 253, P) consists of short, terminal twig fragments, none more than 4 cm long with one or two pairs of opposite or subopposite leaves. The flowers fit well in Cryptocarya , with nine 2-celled stamens and the inflorescences do not have bract scars at their base. The common name (Longotra mena) suggests Aspidostemon . Still, I prefer to treat this as an unusual (by its opposite or subopposite leaves) species of Cryptocarya and not as an unusual (by its nine stamens) species of Aspidostemon .

Capuron commented several times on his labels that the bark of Aspidostemon was platanoid or peeling in large, roundish patches. I have also noticed this in the field; one can recognize Aspidostemon easily by its bark which is soft and cheese-like.

The leaves of Aspidostemon species are not infrequently acuminate, with an acumen that is folded into a short tube, not flat. On opening a few of these inrolled apices I noticed egg-cases similar to those found in leaf domatia; epiphyllous hepatics were also found. Thus, it seems that these inrolled apices function as domatia and shelter mites that clean the leaves.