DELTA home

The grass genera of the world: descriptions, illustrations, identification, and information retrieval; including synonyms, morphology, anatomy, physiology, phytochemistry, cytology, classification, pathogens, world and local distribution, and references

L. Watson, T.D. Macfarlane, and M.J. Dallwitz


This package is generated from a DELTA database (Dallwitz 1980; Dallwitz, Paine, and Zurcher 1993). It comprises an interactive identification and information retrieval system using the program Intkey, descriptions, illustrations, references, and other subsidiary material (see

The original database contained detailed morphological, anatomical and physiological descriptions of over 800 grass genera (Watson and Dallwitz 1981; Watson, Dallwitz, and Johnston 1986; Watson 1987), incorporating extensive original observations. It has been updated, recently in particular (November 2016) to account for the worldwide phylogenetic classification of the Poaceae of Soreng et al. (2015), involving incorporation of over 50 additional genera recognised by them. We choose to retain numerous genera treated by them as synonyms, on the grounds that (a), many of them are likely to be resurrected by other taxonomists; (b), it is far easier to coalesce sensu stricto descriptions than to restore them by breaking down sensu lato versions; and (c), their retention in the present context better informs on sampling in connection with important ‘esoteric characters’ such as anatomy and physiology. We have been at pains to prepare detailed, fully comparative generic descriptions, but users interested in ‘esoteric characters’ (notably anatomy, physiology, pathogens) need to be aware of inevitable underestimates of intrageneric variation reflecting sampling limitations. Sampling information is sometimes directly incorporated in the descriptions, but limitations imposed in earlier times by lack of computer storage space necessitated listing species directly examined for leaf blade anatomy in a separate file. The latter is provided here, and users should note the qualifications set out in its introduction (see

It is easy, using Intkey, to obtain lists of genera exhibiting or lacking particular features or combinations of features; to cross reference attributes with geographical distributions and taxonomic groupings; to compare the existing tribes and subfamilies with one another, and with proposed alternative groupings, in terms of the available descriptive data (with statistical details of character state distributions); to obtain lists of genera for which information on particular features is lacking; to locate generic synonyms; and to obtain printouts of all such information. The Soreng et al. classification is seriously deficient in lacking group descriptions, although these could easily have been prepared in detail from our readily available interactive package, by applying the ‘Summary’ command of Intkey in ‘Advanced Mode’ (cf. our 1998 classification: For information on the more sophisticated uses of it, see ‘Hints on using Intkey’ (available within the Intkey package). The notes provided there, and published examples (Watson et al. 1989, Bruhl et al. 1992) exemplify the kinds of questions you can pursue using Intkey, and illustrate applications of the most commonly used directives. We may as well emphasize that so far as we are concerned, the interactive (Intkey) component of this package ( with its compiled and original comparative descriptive content and illustrations were and remain the main justification for preparing it; so it is exasperating to find modern grass systematists routinely ignoring it and instead quoting from a hard copy book that was automatically generated from the Delta data in 1994.

Since 1977 we have tried, principally via the Delta character list, to set a standard for preparing detailed, fully comparative taxon descriptions or grasses. Our descriptive terminology is mostly in line with normal agrostological usage, as set out in modern textbooks, monographs and regional floras and sometimes modified and extended by us (e.g.,. Hubbard 1968, Hitchcock and Chase 1950, Gould 1968, Jacques-Félix 1962, Bor 1960, Clayton and Renvoize 1986, Chapman and Peat 1992, etc.; and for anatomy, Metcalfe 1960, Clifford and Watson 1976, Ellis 1976 and 1979, and Watson and Dallwitz 1988). In addition to morphology, anatomy, cytology and ecology, our character list and generic descriptions incorporate comparative data on other aspects of ongoing or potential interest to research biologists in various disciplines; e.g., fungal pathogens, physiology, and pollen antigens and allergens. Detailed, written Character Notes have been entered for too few of the characters, but the copious character illustrations should facilitate differentiating between character states.

As with our other Intkey packages, we have tried to avoid encumbering potential users with copyright-restricted images, and have tried of course to ensure that all original sources are (a), succinctly acknowledged in the Intkey displays and (b), cross referenced in the HTML descriptions and the References file ( All images for which sources are not given can be assumed to have been prepared in Watson’s R.S.B.S Taxonomy Lab in association with the present endeavour. External sources include numerous derivatives from high resolution scans of plates from early taxonomic classics, mostly downloaded from the excellent Biodiversity Heritage Library (; notably line drawings from de Beauvois (1812), Kunth (1835) and the Hookers’ Icones Plantarum, as well as coloured plates from the Sowerhy et al. English Botany (1872 edition). These are historically fascinating and very rich in technical taxonomic details, but the present application has inevitably required laborious nomenclatural cross referencing, as well as much editing of contents, revising of legends, and re-arranging image layouts to conserve space. We have also liberally incorporated scans from Gardner’s (1952) accout of Western Australian grasses, and from Grasses of Southern Africa (Gibbs Russell et al., 1990). For the rest, we offer scans of about 550 coloured and black and white photos and line drawings, illustrating spikelet details and leaf anatomy, prepared in Watson’s R.S.B.S. Taxonomy Lab between 1972 and 1994, most of which have not been published elsewhere. Numerous unattractive photos of leaf blade sections prepared from herbarium material are included, because they often provide useful descriptive information.We also include photos of serological slides associated with an investigation of grass pollen allergens and antigens conducted in the Taxonomy Lab, many of which were not included in the related publications of Watson and Knox (q.v.).

The same suite of images is currently employed to illustrate both characters and taxa, with critical features ‘hot-spotted’ in the interactive identificatory context (locate ‘hotspots’ by moving the mouse cursor over text boxes or images). Character states are often exemplified by several images, in attempts to indicate the ranges of variation they encompass, and some genera are represented by numerous taxon images. *Note that in some contexts, such as photos depicting grass pollen antigen and allergen serological reactions, detailed legends can be accessed ONLY via the ‘Notes’ offered on the interactive (Intkey) displays*.

The Taxonomy Lab slide collection is now housed at the Western Australian Herbarium under the care of T.D. Macfarlane, along with the notebooks detailing information on the source material.