Annual or biennials; not scapose; pubescent. Stems (simple or several from base), erect, branched basally and distally. Leaves basal and cauline; petiolate or subsessile; basal rosulate, petiolate, blade lyrate to pinnatifid, margins crenate-dentate; cauline subsessile or petiolate, blade (base not auriculate), margins dentate or pinnatifid. Racemes (corymbose, several-flowered), considerably elongated in fruit. Fruiting pedicels erect, stout. Flowers: sepals widely spreading or reflexed, oblong, lateral pair not saccate basally; petals yellow, obovate to spatulate, claw differentiated from blade, (apex obtuse); stamens tetradynamous; filaments not dilated basally; anthers oblong or ovate, (apex obtuse); nectar glands not confluent, median glands present. Fruits siliques, dehiscent, sessile, segments 2, linear, slightly torulose, terete or slightly 4-angled; (proximal segment not torulose, somewhat corky at maturity, 8-20-seeded; terminal segment indehiscent, 1- or 2-seeded, slightly swollen apically); valves 3(-7)-veined, usually glabrous, rarely sparsely pubescent; replum rounded; septum complete; ovules 10-22 per ovary; (style present); stigma capitate, entire. Seeds uniseriate, plump, not winged, globose; seed coat (smooth to finely reticulate), mucilaginous when wetted; cotyledons conduplicate. x = 7.

It is with some hesitation that I recognize this genus; it should perhaps be united with Erucastrum, as recently proposed for conservation by I. A. Al-Shehbaz (2005b). As clearly shown by S. I. Warwick and L. D. Black (1993), Brassica, Diplotaxis, and Erucastrum are artificially delimited genera, and a substantial revision of their boundaries is needed.