Cardamine pratensis L. in the Netherlands 242 specimens of the Cardamine pratensis aggregate from 58 Dutch localities were morphologically and cytologically studied. Two forms could be distinguished, viz. a form with usually lower chromosome numbers (2n = 28—32, rarely up to 52), and another with higher chromosome numbers (2n = 56—84), which, judging by the criteria applied by LÖVKVIST (12), correspond with C. pratensis sensu stricto and C. palustris (Wimm. & Grab.) Peterm. respectively. The frequent occurrence of ‘transitional’ forms and of specimens which are difficult to place, in conjunction with a number of cytological, genetical, and geographical arguments, make the present authors feel that, chiefly for practical reasons, it is preferable to treat the two forms as subspecies. The two taxa can be clearly distinguished in many, though not in all, instances. The following differential characters are considered to be of diagnostic value: C. pratensis L. subsp. pratensis (fig. 1, a and 2, a—e): rosette leaves smaller with a mean number of pinnae of 4.1; segments of cauline leaves in most cases sessile and smaller; stem usually straight and erect, but rarely over 30 cm tall; mean number of flowering lateral shoots 3; mean value of length of sepals 3.6 mm, of width 2.0 mm; mean length of petals 10.3 mm and mean width 5.5 mm; mean length of filaments 3.4 and 5.2 mm and of anthers 1.7; mean length of siliqua 27 mm; chromosome number usually varying between 28 and 32; flowering season starting relatively early; a nocturnal nastic change of position (‘sleeping’) of the flower; vegetative propagation organs rarely developed; prefers a relatively drier habitat, mainly in grasslands and in forests on basic soils. C. pratensis L. subsp. palustris (Wimm. & Grab.) Janchen (fig. 1, b—d and 2, f—j): rosette leaves larger with a mean number of pinnae of 5.5, segments of cauline leaves usually petiolulate and broader, to broadly elliptic in outline; stem thicker, stouter and more fleshy and frequently showing a torsion about the longitudinal axis; number of flowering lateral shoots upon the average 1, in addition vegetative side shoots often present; mean length of sepals 4.7 mm, mean width 2.5 mm; mean length of petals 12.7 mm, their mean width 7.6 mm; colour of corolla often lighter (to almost completely white); mean length of filaments 4.5 and 6.6 mm and of anthers 2.3 mm; mean length of siliqua 38 mm; chromosome number 56—84; flowering season starting one to two weeks later; no nastic changes in position (‘sleeping’) of flower; adventitious shoots common on rosette leaves, sometimes (also) on stem leaves; prefers wetter habitats, especially in riparian and ‘floating island’ vegetation and in elder carr. The nature of these differences is characteristic of many similar cases of polyploidy (disploidy) and has no specific qualitative indication value. It is feasible that after a more detailed analysis ‘cytoclines’ can be recognized. There is some evidence for the view that the cytological and morphological range in variation is relatively large in habitats which are instable in an ecological sense and relatively small in stable habitats.