Perennial with a musty or aromatic smell, extremely variable in almost all characters. Stems erect or ascending, 30-120 cm, much branched or rarely simple, leafy, with a range of hair types, from soft pilose to crisped to retrorse. Leaves very variable in size, colour and indumentum, 2-8 x 0.5-3 cm, broadly ovate to lanceolate to narrow oblong-elliptic, green, white or discolorous, serrulate to coarsely dentate, cordate, rounded or cuneate, ± acute, sessile or lower leaves shortly petiolate, densely to lightly gland-dotted on abaxial surface. Inflorescence of usually numerous congested many-flowered verticillasters forming terminal spikes, few or numerous. Bracts linear to linear-subulate, as long as to longer than flowers. Calyx 1.5-3 mm, narrow campanulate, softly pilose with eglandular hairs, with or without oil globules; teeth subequal, narrow triangular to linear-subulate, shorter than to as long as tube. Corolla 3-4.5 mm, light purple, violet, mauve or white. Nutlets c. 0.8 x 0.6 mm, ovoid, apically rounded, pale to dark brown, delicately reticulate, slightly mucilaginous on wetting.

Fl. Per.: May-November.

Lectotype (selected by Tucker et al. 234 and fig. 3): Burser specimen (UPS).

Distribution: Europe and Asia (except Far East and SE Asia) southern Africa.

A remarkably variable species in which, throughout its range, a huge number of infra-specific taxa - or independent species - have been recognized. Briquet in the Pflanzenfamilien recognized 21 subspecies and 150 varieties! In our area, with the substantial quantity of specimens at hand, the taxonomic choice is, as with Mentha royleana, between recognizing an endless number of intergrading varieties, elusive to categorize, or else accepting one unusually multiform species. Although the latter course might seem defeatist, it is, with the available evidence, the only practical one to follow in a regional Flora. Breeding systems, cytology and chemistry certainly have a very important role to play in understanding patterns of relationship, but information from these sources is scarcely even fragmentary for our plants.

Rechinger (l.c.) did not accept Mentha longifolia as a species in our area (cf. Mentha royleana), but recognized 8 varieties in the overall Flora Iranica area (he also gave a good synopsis of its history of classification and the problems of typification). Among the varieties dealt with in Flora Iranica was one which Soviet botanists recognize as the widespread C.Asiatic species Mentha asiatica Boriss. Tscherneva (in Grubov, Pl. As. Central. 5: 92. 1970) indicates the species from Kashmir.

The correct nomenclatural citation of this species is, to some extent, a matter of opinion. I have favoured Mentha longifolia (L.) L. but W. T. Stearn cites it as Mentha longifolia (L.) Huds. (cf Stearn, Magnol’s Botanicum Monspeliense and Linnaeus’s Flora Monspeliensis in Geck & Pressler, Festschrift fur Claus Nissen, p. 629-630. Wiesbaden 1973).