The taxonomy of this Sumatrana Group is difficult due to taxa having partially correlated character traits.   Previous taxonomists have recognized only one taxon in this group, either as a variety (Cheng & Fu 1975, 1978; Li & Fu 1997), or species (Florin 1948; Handel Manzzetti 1929; Hu 1964), but they have not applied the correct name.  For example, de Laubenfels (1978) applied the name T. sumatrana to all yews in SW Asia, but this antedated by T. wallichiana.  Further, this very broad concept of T. wallichiana makes it difficult to distinguish T. cuspidata and its allies (Baccata group).

For practical reasons I recognize more than one species in the Sumatrana Group based on differences in the leaves. The abaxial leaf surfaces of T. sumatrana and T. mairei usually have truncated midribs that are slightly channeled in contrast to rounded (keeled) midribs in T. kingstonii.  Taxus mairei is then differentiated by the abaxial midrib having short trapezoidal, inflated epidermal cells, compared to nearly rectangular cells in T. sumatrana (L-sect.), but young leaves of T. mairei may have nearly rectangular cells; thus, I regard the inflated short epidermal cells as the derived (apomorphic) trait.  Taxus celebica is primarily identified by the lanceolate acuminate shape of leaves that usually remain green and flat upon drying, whereas T. sumatrana has darker colored leaves—green to blackish green—that become puckered and revolute along margins upon drying.  Seed shape and seed color, which are variable in this group, could prove valuable to species delimitation; unfortunately, most specimens lack seed. Taxus speciosa Florin is treated as a variety of T. mairei because of their close similarity in leaf anatomy and geographical distribution.

Leaf anatomical differences in other species groups seem less significant because possibly of more frequent hybrids between temperate and subtropical species, although geographical differentiation of taxa is evident in other morphological characters.  For example, zigzag isodichotmous branching characterizes T. mairei in SE China (Fig. 63) in contrast to relatively short unequally much-divided branchlets of T. biternata in NE China and adjacent temperate areas (Figs. 145, 268).   This is in further contrast to longer divaricate branchlets of T. umbraculifera var. microcarpa in SE Russia (Fig. 165).  These differences appear to be a product of adaptive radiation with branchlets becoming more unequally divided from west to east.  Similarly, T. sumatrana in Indonesia and T. biternata (Baccata Group) in Japan may be synapomorphic with regard to their two ranked leaf arrangement and elliptical shape of epidermal cells.  The most logical time for diversification in branching and leaf arrangement is during the Paleocene when the northern latitudes possibly supported a subtropical flora (Hsü 1983; Manchester 1999; Wolf 1975).  On the other hand, a closer similarity in branching between some specimens of T. umbraculifera var. microcarpa in SE Russia (Fig. 288) and T. mairei in its northeastern most range may relate to hybridization or a later history of geographic and climatic differences since the Miocene.