The genus Oenothera (Onagraceae) in Belgium † Krzysztof Rostański1 and Filip VerlooVe2 1 Faculty of Biology and Environmental Protection, Department of Plant Systematics, Silesian University (Katowice, Poland) 2 author for correspondence: Botanic Garden of Meise, Nieuwelaan 38, B-1860 Meise, Belgium [ilip.verloove@br.fgov.be] Illustrations by S. Bellanger (ig. 2-5), W. Vercruysse (7-8), R. Barendse (10: lowers and stem; 11; 17), J. Pottier (10: hairs), R. Jean (21), P. Kalinowski (22) and F. Verloove (the others). AbstrAct. – The genus Oenothera (Onagraceae) was thoroughly and critically revised in Belgium. This revision is based on the study of several hundreds of older and more recent herbarium specimens from all major public and some private herbaria as well as on additional ieldwork carried out since 2000. A brief history of the genus Oenothera in Belgium is provided. Dificulties with regard to taxonomy and nomenclature are discussed. The present treatment of Oenothera follows the ‘European school’ and is thus based on a narrow species concept for subsection Oenothera. In the present paper only species that have been recorded since 1950 are dealt with. As such 27 species (including stabilized hybrids) have been upheld: Oenothera biennis, O. cambrica (with two varieties), O. canovirens, O. delexa, O. depressa, O. ersteinensis, O. fallax, O. fruticosa subsp. glauca, O. glazioviana, O. hirsutissima, O. issleri, O. lindheimeri, O. nuda, O. oehlkersii, O. paradoxa, O. parvilora, O. pycnocarpa, O. rosea, O. royfraseri, O. rubricalyx, O. rubricaulis, O. rubricauloides, O. stricta, O. subterminalis, O. victorinii, O. villosa and O. wratislaviensis. Nine additional species were only recorded before 1950. Oenothera is represented in Belgium by the following sections, subsections and series: sections Gaura, Hartmannia, Kneifia and Oenothera, the latter furthermore represented by subsections Munzia, Oenothera (with the series Devriesia, Linderia, Oenothera and Rugglesia) and Raimannia. An identiication key and photographs (mostly of typical and/or widespread species) are presented. Each species is concisely compared with related or similar species and its current distribution and naturalization status in Belgium are assessed. résumé. – Le genre Oenothera (Onagraceae) en Belgique. Le genre Oenothera (Onagraceae) fut inventorié en Belgique. Cette révision taxonomique est basée sur l’étude de plusieurs centaines d’échantillons d’herbier (anciens ainsi que plus récents) et des prospections de terrain à partir de 2000. L’histoire du genre Oenothera en Belgique est brièvement présentée et des dificultés concernant la taxonomie et nomenclature commentées. La présente révision du genre Oenothera suit la dite ‘école européenne’ pour la sous-section Oenothera. Dans cet article seulement les 27 taxons (y inclus quelques espèces hybridogènes) récoltés après 1950 sont retenus : Oenothera biennis, O. cambrica (avec deux variétés), O. canovirens, O. delexa, O. depressa, O. ersteinensis, O. fallax, O. fruticosa subsp. glauca, O. glazioviana, O. hirsutissima, O. issleri, O. lindheimeri, O. nuda, O. oehlkersii, O. paradoxa, O. parvilora, O. pycnocarpa, O. rosea, O. royfraseri, O. rubricalyx, O. rubricaulis, O. rubricauloides, O. stricta, O. subterminalis, O. victorinii, O. villosa et O. wratislaviensis. Neuf taxons supplémentaires ont été récoltés uniquement avant 1950. Les taxons précités appartiennent aux sections, sous-sections et séries suivants : les sections Gaura, Hartmannia, Kneifia et Oenothera, ce dernier en plus avec les sous-sections Munzia, Oenothera (avec les séries Devriesia, Linderia, Oenothera et Rugglesia) et Raimannia. Une clé pour l’identiication des espèces est présentée ainsi que des photos des espèces les plus caractéristiques et/ou les plus répandues. Chaque espèce est comparée avec des espèces proches ou similaires et sa distribution et son statut (fugace/naturalisé) sont fournis. sAmenvAtting. – Het genus Oenothera (Onagraceae) in België. Het geslacht Oenothera (Onagraceae) werd kritisch gereviseerd in België. Deze revisie is gebaseerd op de studie van enkele honderden herbariumspecimens (uit de belangrijkste publieke herbaria van België en uit enkele kleinere private herbaria), alsook op veldwerk, uitgevoerd sinds 2000. De hisDumortiera 106/2015: 12-42 12 toriek van het geslacht Oenothera in België wordt geschetst en problemen met betrekking tot de nomenclatuur en taxonomie worden aangekaart. De huidige revisie volgt de zogenaamde ‘Europese school’ en hanteert dus een eng soortsbegrip voor de subsectie Oenothera. Alleen de 27 taxa (inclusief enkele gestabiliseerde hybriden) die na 1950 in België zijn waargenomen, worden uitgesleuteld en besproken: Oenothera biennis, O. cambrica (met twee variëteiten), O. canovirens, O. delexa, O. depressa, O. ersteinensis, O. fallax, O. fruticosa subsp. glauca, O. glazioviana, O. hirsutissima, O. issleri, O. lindheimeri, O. nuda, O. oehlkersii, O. paradoxa, O. parvilora, O. pycnocarpa, O. rosea, O. royfraseri, O. rubricalyx, O. rubricaulis, O. rubricauloides, O. stricta, O. subterminalis, O. victorinii, O. villosa en O. wratislaviensis. Negen soorten werden uitsluitend voor 1950 in België aangetroffen. Het geslacht Oenothera wordt in België vertegenwoordigd door de volgende secties, subsecties en series: de secties Gaura, Hartmannia, Kneifia en Oenothera, deze laatste bovendien door de subsecties Munzia, Oenothera (met de series Devriesia, Linderia, Oenothera en Rugglesia) en Raimannia. Een determinatiesleutel en foto’s (meestal van typische en/of wijd verspreide soorten) worden eveneens gepresenteerd. Elke soort wordt vergeleken met gelijkaardige of verwante soorten en de huidige verspreiding en inburgeringsgraad in België worden kort besproken. Introduction1 The genus Oenothera L. is not native to Belgium. Although O. biennis is naturalized in this country since at least the 18th century (Roucel 1792), the exact number and the true identity of the other Belgian taxa always remained unclear. This is primarily because of the very complex taxonomy of the genus, especially with regard to section Oenothera subsection Oenothera. For this group, there are two very divergent taxonomical views. According to Dietrich et al. (1997) the group is made up of only 13 species. This concept is usually referred to as the ‘American school’ but this is misleading. As early as 1958, the American botanist and geneticist Ruggles Gates already presented a taxonomic concept for Oenothera that for the greater part confers with the current ‘European’ species concept. Moreover, Dietrich is German. Rostański (1985), who represents the so-called ‘European school’, claimed there were 85 species and 23 hybrids. Moreover, in the past decades several additional new taxa have been described (e.g. Jehlík & Rostański 1995, Delipavlov 1998, Rostański & Ramst 2001, Rostański et al. 2004, Rostański 2007, Deschâtres et al. 2013). As a result of a peculiar breeding mechanism (permanent translocation heterozygosity2) species from subsection Oenothera cross freely, producing off-spring that differs from either parent and developing populations that consist of many different but in fact continuous genotypes. As such, numerous, closely similar but morphologically well-delimited entities are formed. These newly arisen The results presented in this paper are almost exclusively based on the determinations by the irst author. The text, however, was written by the second author and approved by Dr. Adam Rostański, son of the irst author. 2 Permanent translocation heterozygosity is a cytological phenomenon in which an organism is true breeding for a set of translocations involving all or part of the chromosomes. 1 plants breed more or less completely true. They may be intermediate but are usually nearer one parent than the other and the degree of resemblance may vary from character to character. Some hybrids even resemble one parent species more strongly in one stage of development and the other parent at a later stage (Gates 1933). Also, when Oenothera species spread to new habitats, new ‘species’ (or mere morphotypes?) adapted to new conditions, may arise (Tokhtar & Wittig 2003). These ‘microspecies’ were included in one of the 13 recognized species (or hybrids of these) by Dietrich et al. (1997). As such, for example for O. biennis, there are no less than 68 ‘synonyms’. According to Rostański, whose work was for the greater part based on the earlier publications of Renner (1942, 1950, 1956) these mutants and hybrids are, once more or less established, preferably considered as independent species. The choice between both concepts is not straightforward and merely a matter of opinion and there is no consensus whatsoever. However, this species concept is perhaps not applicable in North America, where the genetic variation is greater than in Europe. Wisskirchen & Haeupler (1998) presented, without judgement of their respective merits, both concepts next to one another. In our view, we consider that Dietrich et al. (1997) presented a too broad species concept. Under O. biennis, for example, numerous, easily differentiated species, are grouped together (e.g. O. rubricaulis) that altogether often hardly resemble O. biennis s.str. Other examples: O. perangusta is included with O. oakesiana, whilst the (possibly) conspeciic O. ersteinensis is included under O. biennis. Also, O. delexa is included in O. parvilora whilst both are not in the least related and actually belong to different, relatively distinct series. See also a more lengthy discussion in Rostański et al. (2010). In Europe, since Renner (1917), who was the founding father of the European school, a narrow species concept has been applied most of K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 13 the time although recently, in Flora Iberica, an exception was made (not surprisingly, the genus Oenothera being prepared by Dietrich; see Dietrich 2000). Van der Meijden (2005a) originally also opted for this broad species concept for the 23rd edition of Heukels’ Flora (van der Meijden 2005b), yet eventually he accepted O. delexa and O. fallax as two taxa, which were not recognized by Dietrich et al. (1997). In the most recent edition of the Belgian lora (Lambinon & Verloove 2012) numerous microspecies are accepted in taxonomically similar, complex genera (e.g. the apomictic genera Rubus and Taraxacum). For these reasons, it would be at least useful and perhaps even advisable to follow the European school, applying a narrow species concept for Oenothera in Belgium and neighbouring territories. This will allow tracking of changes in range and frequency and to distinguish patterns and mechanisms of distribution; application of a wide species concept would lead to considerable loss of information (Rostański & Karlsson 2010). Worthwhile mentioning is that molecular phylogenetic studies have been performed in the genus Oenothera (e.g. Krakos et al. 2014) but not yet in subsection Oenothera. In 2000 the irst author began a systematic revision of the Belgian Oenothera collections, stored in the main public herbaria (the herbarium of the Botanic Garden Meise, BR; the herbarium of the University of Liège, LG; and the herbarium of the University of Ghent, GENT) and some smaller private herbaria. Since then the second author attempted to document, as much as possible, the variation of Oenothera currently encountered in favourable habitats in Belgium (e.g. in seaports, coal mining spoil heaps, sand pits, landills, railway infrastructure, sand depots, etc.). In this paper, all species reliably recorded since 1950 are keyed out and briely discussed. Those species that exclusively occurred as wool aliens in the Vesdre valley and species that were last collected before 1950 are excluded because they are, for the present day ield botanist, of lesser interest. These species are merely cited in the classiication scheme (preceded by an asterisk) and their respective records are included in the appendix of this paper. An overview of the studies concerning Oenothera in Belgium and Europe As previously mentioned the genus Oenothera was not well documented in Belgium. The majority of loras of the 19th and the irst half of the 20th century mentioned only O. biennis and O. muricata (auct. belg. non L.) (e.g. Durand 1899), occasionally along with a few ephemeral aliens such as O. laciniata or O. rosea. Goffart (1945) added O. glazioviana; later, it turned out that this species was already found in Belgium since at least 1868 (Verloove 2006a). De Langhe et al. (1967) listed the following species: O. biennis, O. erythrosepala (syn.: O. glazioviana), O. parvilora and the ephemeral aliens O. laciniata and O. stricta. Around the mid-seventies of the 20th century the Belgian Oenothera collections of the herbarium of the Botanic Garden Meise (BR) were revised for the irst time by an expert (Jean 1975). He conirmed the reported species of De Langhe et al. (1967) and added the following: O. ammophila, O. atrovirens (syn.: O. cruciata), O. fallax (as O. biennis × O. lamarckiana), O. hungarica (syn.: O. depressa) and O. longilora. Recently, it became clear that O. ammophila as well as O. longilora were misidentiications, the corresponding collections belonging respectively to O. parvilora and O. afinis (Verloove 2006a). In the subsequent editions of the Flora of Belgium the results of Jean’s revision were maintained. Moreover, in Lambinon et al. (1998) O. nuda was added, a species observed in northern France (Jean 1990). In Lambinon et al. (2004) some provisional results from the current study were already included (mostly based on Verloove 2002). Most notable was that the Belgian populations of the ‘small lowered evening primrose’ were attributed to O. delexa and no longer to O. parvilora (see further). Verloove (2006a) presented an overview of all observed species in Belgium prior to 2005 and inally, in the most recent edition of the Belgian lora (Lambinon & Verloove 2012), an extra species was keyed-out (the relatively widespread and well deined O. fallax). The list of ephemeral adventives was also updated. Elsewhere in Europe, in the last decades, a reasonably good picture has been formed of the diversity and spread of the genus (mostly as a result of the continuous revisions by the irst author). For many European regions there are overviews available by now. The most important (with identiication keys and/or good illustrations) are presented in table 1, chronologically listed per country or region. These publications are often relevant for the Oenothera lora in Belgium as well. Especially north, central and east European revisions usually include species that are also found in Belgium. In contrast, for southern Europe (e.g. Italy, Portugal and Spain) there are proportionally more taxa that are not found in Belgium (and vice versa). Finally, Dietrich (1999) offers an interesting and upto-date overview of the genus Oenothera in cultivation. Problems concerning the identiication of Oenothera Unfortunately, the complex taxonomy is not the only problem in the identiication of Oenothera species. Some important diagnostic characteristics disappear on drying or become less visible: the presence or not of hairs with a red bulbous base, the colour and the size of the petals and their length and width, the red pigmentation of (parts of) the stem, the colour of the sepals, the colour of the midrib of the leaves, etc. For these reasons, it is of the utmost importance that specimens are either identiied from fresh material, or with the help of ield notes (especially dimensions and colour of the different plant parts). Moreover, the identiication of many species needs to be carried out on lowering and fruit bearing plants. It also is important to know that the identiication of Oenothera species only makes sense during the irst low- K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 14 Table 1. An overview of revisions of Oenothera from Europe. Countries or regions Publications Europe Rostański et al. (2010) Germany Hudziok (1968), Otto (1970), Gutte & Rostański (1971), Rostański & Schnedler (1991), Gutte (1998), Haeupler & Mür (2000), Lang (2001), Jäger & Werner (2005), Rostański & Meierott (2006) France Rostański et al. (1994), Tison & de Foucault (2014) Great Britain Rostański (1982), Sell & Murrell (2009) Austria Rostański & Forstner (1982) Eastern Europa (Baltic States, Russia, Belarus, Ukraine and Moldova) Rostański et al. (2004) Scandinavia (Denmark, Rostański (2006), Rostański (2007), Finland, Iceland, Norway Rostański & Karlsson (2010) and Sweden) Czech and Slovak Republic Jehlík & Rostański (1979), Jehlík & Rostański (1995), Jehlík (1997) Serbia and Montenegro Zlatković et al. (1998); Rakaj & Ros­ tański (2009) Spain Dietrich (2000) Portugal Rostański (1991) Italy Soldano (1980), Soldano (1983a, b), Soldano (1993) ering period (for most species this is from mid-June to the end of July). During the second lowering period, the petals are almost always (much) smaller (especially in largelowered species), the number of glandular hairs tend to increase while the number of stiff, eglandular hairs diminishes, the branching pattern changes (i.e. development of side branches below the main inlorescence or of oblique stems arising from the leaf rosette) and other characters also are often less obvious. It is inally recommended to collect only ‘typical’ examples of a population. In many Oenothera populations – certainly those where two or more species grow in close proximity – plants with intermediate characteristics are not uncommon. These locally arisen, not-ixed hybrids will, without doubt, fail to key out properly. In this perspective, it is an illusion to think that it is possible to assign a name to each individual Oenothera plant. However, the identiication of more or less homogeneous populations is generally feasible provided one has some ield experience. In fact, in a mixed population of Oenothera species the dificulty mostly is to assess which plants species are the ‘pure’ ones and which are the hybrids. Even if the above is taken into consideration, the identiication of Oenothera species often remains a dificult task. The correct identiication of herbarium material without accompanying ield notes is, even for experts, not easy. The Belgian Oenothera collections of the herbarium of the Botanic Garden Meise have been reviewed by Raymond Jean (see Jean 1975), Werner Dietrich and the irst author of this paper. That the conclusions of Dietrich and Rostański differ stems (mostly) from differences in taxonomical concepts. Jean, however, followed the narrow ‘European’ species concept and yet sometimes came to conclusions which were divergent to those of Rostański. For instance, plants which were named by Jean as “close to O. ersteinensis” were inally attributed to O. delexa and O. parvilora. A collection named as “O. hungarica” (= O. depressa) is here ascribed to O. oakesiana although both belong to different series. Without doubt, the most striking was that nearly all O. parvilora collections – one of the few more or less widely spread and fully established species in Belgium – turned out to be ascribable to (the not even remotely related) O. delexa. This important correction was already picked up by Lambinon et al. (2004). The repeated confusion, even amongst experts, was clearly illustrated with respect to a collection of Oenothera oakesiana from BR (Fig. 1); this collection was Figure 1. Identiication of species in genus Oenothera has often been dificult, even among experts. This collection of Oenothera oakesiana from BR was originally identiied by A. Lawalrée as O. biennis and later annotated by Dietrich, Jean and the irst author. It inally bears four names of species that belong to three different series. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 15 Figure 2. Flower morphology: A, stigma lobes; B, anthers; C, petal; D, sepal; E, sepal tip; F, hypanthium; G, ovary; H, bract. originally identiied by A. Lawalrée as O. biennis and later annotated by Dietrich, Jean and the irst author. It inally bears four names of species that belong to three different series! Diagnostic characters Although similar, the species in subsection Oenothera are (with some experience) easily recognized in the ield. The diagnostic characters, such as hairiness, colour of vegetative parts and size of loral parts (Fig. 2), are strikingly constant. However, herbarium material is frequently more dificult to identify because some characters are lost, especially in poorly prepared material (see before; also Jean 1975). The following characters are particularly useful for an accurate identiication (mostly based on Rostański & Karlsson 2010): Mode of growth. The stem is erect in most species but ascending in a few. The tip of the growing inlorescence is erect in most species but in some, e.g. Oenothera parvilora and related species from series Rugglesia, it is drooping (as beautifully illustrated in Renner 1956). This is an important character that can easily be overlooked, e.g. in late specimens where the axis has straightened. Also, in some species of series Rugglesia, this character is less obvious, for instance in O. parvilora (Linder 1958). Hair types (Fig. 3). There are trichomes of three main types: 1) short, soft, crispate hairs on the vegetative part of the stem and, in some species, on the ovary and fruit; 2) glandular hairs, i.e. patent, glistening, short and thick (sometimes slightly clavate) trichomes, on rachis, hypanthium, sepals, ovary and fruit in many species; and 3) long, stiff, more or less patent, straight or slightly bent hairs on most parts of the plant. In some species the hairs rise from a well-delimited, multicellular base which is often an intense red (presence of anthocyanin pigment). The shape of the thickened hair base varies. In most cases it is conical and about as high as wide; more rarely it is wide and low, as in O. depressa, or distinctly elongated, cylindrical and ± bent, as in O. ersteinensis. Leaf shape. The relative leaf width varies between the species and, to some extent, within the species. Some are always narrow-leaved; others, like O. biennis and O. cambrica are always broad-leaved. The leaves are usually lat, but in some species they are crinkled and sometimes have a slightly twisted apex (e.g. in O. depressa). Leaf hairiness. The leaves are usually sparsely hairy, but in some species, e.g. O. hirsutissima and in series Devriesia (e.g. O. canovirens and O. depressa), they appear grey-green and velvety to the touch, due to a fairly dense indumentum. Leaf colour. The midrib of the leaves is whitish in some species, ± red in others. However, the red colour is sometimes not developed in species normally having red midribs, e.g. in specimens growing in shade. Rachis colour. The main axis (rachis) of the inlorescence can be either green throughout as in O. biennis or an intense red towards the apex as in O. ersteinensis and O. rubricaulis. This character is more stable within species than the colour of leaves and sepals. Sepal colour. The sepals are green in some species, whereas in others, e.g. O. glazioviana, they have red streaks or a red tinge or they may be entirely deep red (e.g. O. rubricalyx). The red colour is, however, not always developed. There are also species with a genetically conditioned variation in this character (e.g. O. oehlkersii). Buds (Fig. 4). The sepal-tips can be entirely separated or pressed to each other in their full length, or together in their proximal part but diverging distally. The length of the sepal tips and the length of the sepals (including the tip, and measured in fully mature buds) differ between some species. Flower size. The length and width of the petals, as well as the length/width ratio, are diagnostic. The length of the hypanthium also differs much between species; it is not correlated with petal size. Care should be taken to study (left) Figure 3. Hair types: A, hair with bulbous base; B, crispate hair; C, glandular hair; D, bristle hair. (right) Figure 4. Flower buds and sepal tips: A, sepal tips appressed throughout; B, sepal tips appressed at base, separated at apex; C, sepal tips separated (V-shaped); D, sepal tips separated (U-shaped). K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 16 Figure 5. Capsule and morphological variation of capsule apex: A, capsule; B, capsule valve; C, teeth acute; D, teeth truncate; E, teeth emarginate. the plant at its peak of lowering. Later in season the lowers become notably smaller, especially in the large-lowered taxa. In some species the lowers are sometimes or even usually cleistogamous (e.g. O. depressa). Position of stigma. The stigma is surrounded by the anthers at anthesis in most species, but distinctly raised above them in e.g. O. glazioviana and O. oehlkersii. Hairiness of capsules. The density of trichomes of different kinds on the capsules is of major diagnostic value, but care must be taken to compare capsules at the same height, because the hairiness varies with position (i.e., between lower, older capsules and upper, younger ones). Generally, glandular hairs are more numerous on the upper, younger capsules. There are species where glandular hairs are absent on the lower (older) capsules, but present on the upper (younger) ones. Capsule valves (Fig 5). The tips of the capsule valves are rounded, truncate or more or less emarginate. Their outline is best seen in capsules that are about to open. Degree of naturalization of Oenothera in Belgium In Belgium Oenothera species are mostly found in environments which have been heavily disturbed by human activity: ports and railway grounds, sand and gravel pits, dredge dumping areas, sand and mineral depots, coal mining spoil heaps, ruderalized sand dunes, etc. The lora of such environments is often rather ephemeral. Although Oenothera often occurs in great numbers, this is, in the most of its habitats, a temporary phenomenon. Moreover, the knowledge of the genus in Belgium is virtually nonexistent and as such only the experience of the second author can serve as a basis. Because of this, the correct estimation of the extent of naturalization for many species is very dificult. Perhaps only a few species are more or less widespread and relatively common. This corresponds with previous indings: Mihulka & Pyšek (2001) indicated that only few species are successful in terms of increasing abundance while the majority of the species remain rare and local. Probably the most widespread species for the moment is O. delexa. It also appears to be the most expansive of all species. Locally, for example in the coastal sand dunes, O. glazioviana is also well established as is O. fallax. The latter is spreading vigorously, for example, in the ports of Ghent and Antwerp. This species has also become one of the most frequently found species in the west of Germany (e.g. Wittig and Tokhtar 2003). At least in the area around Ghent, O. rubricaulis is well established and in expansion. This species was already found in great numbers in 1985 (e.g. E. Robbrecht 2929 in BR) and it is now one of the typical species for this area. The same holds true for O. ersteinensis around Antwerp and in the Kempen. O. biennis is a perplexing case: it is, without doubt, the species with the longest history in Belgium but currently appears to be, to the greater extent, replaced by related species. Although, this species is still regularly collected, this is much less so in comparison with the other species cited before. Locally, in the second half of the 19th century, O. angustissima (syn.: O. muricata auct. belg. non L.) was more or less established, for example, along a railway track in the vicinity of Sint-Truiden (at least between 1860 and 1884). All the remaining Oenothera taxa are either not well known enough, or are only very locally more or less established, or (without doubt in the most cases) strictly ephemeral. Some of the aforementioned species (e.g. O. glazioviana) are naturalized in great number in natural habitats, for instance in coastal dunes. In the sense of Richardson et al. (2000) they can probably be considered as invasive species. However, there are no indications that they replace indigenous vegetation (‘transformers’). As shown elsewhere, species from other (sub-) sections that were introduced relatively recently proved to exhibit a much stronger invasive behaviour (e.g. Frean et al. 1997). However, Tokhtar et al. (2011) demonstrated that (notho-) species from subsection Oenothera can also act as invasive species, especially if one putative parent is of North American and the other of European origin. Finally, an additional dificulty in assessing the degree of naturalization of Oenothera is the morphological plasticity of many populations. It often happens that Oenothera populations change strongly during the course of years: small but apparently more or less stabilized populations of rare species often disappear in a hybridization swarm of commoner species. For example, a nice population of O. victorinii very gradually disappeared from rough ground in the vicinity of Roeselare. The few remaining plants became less and less typical, doubtlessly as a result of introgression with the co-existing populations of O. glazioviana and/or O. delexa, eventually disappearing altogether. The same holds true for a population of O. hirsutissima found on a sand storage area in the canal zone of Bruges: between 2004 and 2007 a few typical plants were collected whilst from 2008 onwards only non-deined hybrids appeared to be present. Analogously, a population of O. wratislaviensis in the Antwerp port area, discovered in 2007, is replaced now by plants that exhibit characters of this species, as well as of the widely K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 17 spread O. ersteinensis. This is a well-known phenomenon in Oenothera: Renner (1950) referred to it as “Der Kampf zwischen der Arten”. Geographical origin of the genus Oenothera The origin of the genus Oenothera is quite uncertain. According to Wagner et al. (2007) its centre of diversity is in southwestern North America and it is now widely distributed in temperate to subtropical areas in North and South America with a few species in Central America as well. A number of species have become naturalized nearly worldwide (Frean et al. 1997), most likely in post-Columbian times. According to the European school, on the contrary, several species are endemic to and arose in Europe. This applies not only for the European mutants and hybrids (that are unknown in America) but probably also for some ‘pure’ species such as O. biennis. Most likely, Oenothera subsection Oenothera underwent its primary differentiation in North America but has a secondary centre of evolution in Europe (Rostański & Karlsson 2010). Quite a few originated by hybridization in Europe and became stabilized due to the extraordinary genetic mechanism mentioned above, and have subsequently become widespread. Species like O. biennis and O. rubricaulis have been present for a long time in Europe. Since exactly matching biotypes are not known in the wild in North America, early immigration and speciation has been proposed (Rostański 1968, 2004), while Dietrich et al. (1997) suggest a hybridogenous origin in Europe from North American taxa introduced in historic time. Yet another possibility (comm. R. Jean, August 2014) is that Oenothera is, indeed, of North American origin but that species of subsection Oenothera already entered the Eurasian Continent, via the Bering land bridge, at the end of the last glacial period. This could explain why, genetically, ‘European’ species of Oenothera differ to such an extent from ‘American’ congeners. Species from subsections Raimannia (originating in North America) and Munzia (originating in South America), on the contrary, were doubtlessly introduced in post-Columbian times in Europe, have a much more limited distribution (mostly in coastal areas, without having penetrated deeper in the Continent) and are identical to plants found in their respective areas of origin. Identiication key Despite being stable, and as such occurring in the absence of their putative parental species, Oenothera paradoxa and O. wratislaviensis have not been included in the identiication key. Both are rare and relatively atypical taxa that are dificult to incorporate into the key. They are briely described with their assumed parental species (O. canovirens, O. depressa and O. subterminalis). Both are hybrids between species of the series Devriesia and Rugglesia. Their capsules display an indumentum that is more or less intermediate, with an indeinite mixture of short glandular and stiff, long and eglandular hairs. From the series Devriesia and Oenothera, the unixed hybrid O. delexa × O. depressa has also been recorded on several occasions. As previously mentioned, it is virtually impossible to identify sporadically occurring hybrids. Plants which either cannot or only with dificulty be keyed-out belong in all probability to hybrid swarms. The current study follows the so-called European school and thus accepts a narrow species concept. Nevertheless, it seemed also useful to always refer to the corresponding name in the taxonomic concept as deined by the so-called American school. For this reason, for every taxon of the section Oenothera subsection Oenothera the alternative name under the latter is indicated between square brackets. 1 Petals yellow when fresh (rarely pale yellow) (but in some species drying pinkish, orangish or reddish!). Capsule cylindrical or fusiform, very rarely clavate and abruptly constricted to base (only in O. fruticosa) .... 3 2 Capsule indehiscent. Seeds usually 3­4. Flowers strongly zygomorphic ........... Oenothera lindheimeri Capsule dehiscent. Seeds usually numerous. Flowers actinomorphic .............................................. O. rosea 3 Flowers opening at sunrise. Capsule clavate, abruptly constricted or cuneate to the base (apparently pedicel­ late), narrowly winged .... O. fruticosa subsp. glauca Flowers opening near sunset. Capsule cylindrical or fusiform, without wings, not constricted to base ...... 4 4 Classiication of Oenothera in Belgium The overview of all the taxa collected in Belgium is divided over the different sections, subsections and series (Table 2). Taxa collected exclusively before 1950 are preceded by an asterisk (*). In the rest of this paper (except in the appendix), no further reference is made to these taxa (see earlier; also: Verloove et al. 2014). This classiication is in accordance with the recent revision by Wagner et al. (2007) with the exception of Oenothera subsection Oenothera. This implies that the genus Gaura L. is included in Oenothera. Petals white or pink when fresh. Capsule clavate, ab­ ruptly constricted or cuneate to the base (apparently pedicellate) .............................................................. 2 Seeds ellipsoid to globose, not prismatic and angled. Capsule fusiform, enlarged towards apex (2­4 mm wide at base). Petals yellow, often with a red spot at base (turning pinkish on drying) .................. O. stricta Seeds prismatic, distinctly angled. Capsule cylindrical, narrowed towards apex (6­8 mm wide at base). Petals yellow, unspotted (exceptionally fading whitish or pink­ ish on drying) ........................................................... 5 5 Stigma distinctly elevated above anthers at anthesis. Petals always large, 30­55 mm long ........................ 6 Stigma surrounded by anthers at anthesis or only slightly exceeding anthers. Petals small or large ..... 9 6 Sepals entirely green or greenish yellow. Stem with­ out hairs with red bulbous base [O. biennis × glazioviana?] ..................................................... O. oehlkersii K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 18 7 8 9 10 11 12 13 14 Sepals red­striped, suffused red or uniformly deep red. Stem with hairs with distinct red bulbous base ........ 7 Leaf blade nearly plane, subentire at margin, up to 25 mm wide, the upper surface grayish green (in­ dumentum of dense, short hairs). Hypanthium (loral tube) without or with scattered glandular hairs. Ovary and young capsule densely white hairy, without or with few glandular hairs [O. elata] ........... O. hirsutissima Leaf blade distinctly wavy, irregularly dentate at mar­ gin, up to 40 mm wide, the upper surface green and nearly glabrous. Hypanthium densely glandular hairy. Ovary and young capsule predominantly glandular hairy, without or with few long, white and eglandular hairs [O. glazioviana] ............................................... 8 Sepals uniformly deep red ................... O. rubricalyx Sepals red­striped or suffused red ..... O. glazioviana Ovary and young capsules densely covered with long white hairs, without or with few glandular hairs. Cauline leaves usually lanceolate, upper surface (at least when young) grayish green with indumentum of dense, short hairs, velvety to the touch. Capsule teeth emarginate or truncate at apex [O. villosa] ............................... 10 Ovary and young capsules not densely white hairy, with a mixture of (predominantly) glandular and (scat­ tered) patent eglandular hairs. Cauline leaves elliptic to lanceolate, glabrous to slightly hairy. Capsule teeth rounded or cuneate at apex, rarely slightly emarginate ................................................................................ 12 Stem with numerous hairs with red bulbous base. Bracts and upper cauline leaves with distinctly wavy margins and twisted at apex. Flowers often cleisto­ gamous .................................................. O. depressa Stem usually without or (rarely) with sparse hairs with red bulbous base. Bracts and upper cauline leaves with lat margins. Flowers chasmogamous, fully open­ ing ........................................................................... 11 Capsule teeth truncate at apex .................. O. villosa Capsule teeth emarginate at apex ...... O. canovirens Inlorescence axis slightly curved before lowering (straightening with time). Cauline leaves narrowly lan­ ceolate, always with red midrib. Sepal tips not adher­ ing at base, with U or V shaped sinus1. Petals small, at most 20 mm long (series Rugglesia) ..................... 13 Inlorescence axis straight from the beginning. Cauline leaves elliptic to lanceolate, with white or red midrib. Sepal tips in bud adherent at base (series Oenothera) ................................................................................ 15 Sepal tips in bud 2­3 mm long, with U shaped sinus. Petals 6­12 mm long [O. parvilora] ...... O. parvilora Sepal tips in bud 2­9 mm long, with V shaped sinus. Petals 10­20 mm long ............................................ 14 Stem dark red, at least in lower half. Sepal tips in bud 4­9 mm long [O. parvilora] ............ O. subterminalis Stem green or slightly suffused red in lower half. Sepal tips in bud at most 3 mm long [O. biennis × oakesiana] ............................................................. O. issleri The shape of the sepal tips in bud is a useful diagnostic feature, especially to distinguish between series Oenothera and Rugglesia. However, this character should preferably be assessed on fresh material. See ig. 4. 1 15 Inlorescence axis and upper half of stem without hairs with red bulbous base. Buds, sepals and inlorescence axis uniformly green ............................................... 16 Inlorescence axis and upper half of stem with numer­ ous hairs with red bulbous base. Buds, sepals and in­ lorescence axis either green, red, or green suffused with red .................................................................. 20 16 Inlorescence axis nearly glabrous (at most with very scattered, long eglandular hairs) [O. biennis] ............. ....................................................................... O. nuda Inlorescence axis densely hairy, in part with short, glandular hairs ....................................................... 17 17 Petals 20­30 mm long [O. biennis] ......................... 18 Petals 10­20 mm long ............................................ 19 18 Cauline leaves lanceolate. Petals about as long as wide. Lowermost (oldest) capsules without glandular hairs ............................ O. cambrica var. impunctata Cauline leaves wider, elliptic to elliptic­lanceolate. Pet­ als obviously wider than long. All capsules glandular hairy .......................................................... O. biennis 19 Petals 9­12 mm long. Hypanthium up to 30 mm long. Capsule up to 40 mm long. Sepal tips 3­4 mm long [O. parvilora] .................................................. O. delexa Petals 10­20 mm long. Hypanthium up to 40 mm long. Capsule 30­50 mm long. Sepal tips 5­7 mm long [O. biennis] ................................................... O. victorinii 20 Cauline leaves with wavy margins. Sepals always red­ striped. Petals 25­30 mm long and slightly wider [O. biennis × glazioviana] .................................. O. fallax Cauline leaves plane. Sepals green or suffused with red (red­striped in O. ersteinensis only). Petals often smaller, 10­28 mm long ......................................... 21 21 Inlorescence axis at the beginning of lowering red (stem red or green) ................................................ 22 Inlorescence axis at the beginning of lowering green (stem red or green) [O. biennis] ............................. 24 22 Pustulate hair base cylindrical, much longer than wide and often slightly curved downwards. Sepals mostly red­striped (rarely green). Stem usually entirely red [O. biennis] ............................................. O. ersteinensis Pustulate hair base conical, ca. as long as wide. Se­ pals green. Stem green, at least in lower half ........ 23 23 Hypanthium 15­25 mm long. Petals 10­20 mm long and 9­18 mm wide [O. biennis] ........... O. rubricaulis Hypanthium 25­35 mm long. Petals 20­28 mm long and about as wide [O. biennis?] ... O. rubricauloides 24 Petals 6­12 mm long ............................. O. royfraseri Petals 15­30 mm long ............................................ 25 25 Cauline leaves lanceolate, with red midrib. Petals 15­25 mm long. Hypanthium 30­40 mm long. Inlor­ escence lax, more or less pyramidal. Stem nearly al­ ways red, at least in lower half .......... O. pycnocarpa Cauline leaves wider, elliptic­lanceolate, with white or red midrib. Petals 20­30 mm long. Hypanthium 25­35 mm long. Inlorescence dense, oblong. Stem usually green throughout ........... O. cambrica var. cambrica K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 19 Table 2. Classiication of the Oenothera taxa collected in Belgium. Taxa collected exclusively before 1950 are preceded by an asterisk. Based on Wagner et al. (2007). For subsection Oenothera the names in the second column are those of the so­called American school. Section Gaura (L.) W.L. Wagner & Hoch * Oenothera curtilora W.L. Wagner & Hoch (syn.: Gaura parvilora Douglas ex Lehm.) Oenothera lindheimeri (Engelmann & A. Gray) W.L. Wagner & Hoch (syn.: Gaura lindheimeri Engelmann & A. Gray) Section Hartmannia (Spach) W.L. Wagner & Hoch Oenothera rosea L’Hérit. ex Ait. * Oenothera speciosa Nuttall Section Kneifia (Spach) Straley Oenothera fruticosa L. subsp. glauca (Michaux) Straley * Oenothera perennis L. Section Oenothera Subsection Munzia Dietrich Series Allochroa (Fischer & Meyer) Dietrich * Oenothera afinis Cambess. in St.­Hilaire * Oenothera indecora Cambess. in St.­Hilaire Oenothera stricta Ledeb. ex Link Subsection Oenothera Series Devriesia Rostański Oenothera canovirens Steele Oenothera depressa E. Greene Oenothera paradoxa Hudziok Oenothera villosa Thunb. Oenothera wratislaviensis Rostański ex Rostański O. villosa O. villosa O. biennis O. villosa ? Series Linderia Rostański Oenothera hirsutissima (A. Gray ex S. Watson) de Vries O. elata Series Oenothera Oenothera biennis L. Oenothera cambrica Rostański Oenothera delexa Gates Oenothera ersteinensis Linder & Jean Oenothera fallax Renner emend. Rostański Oenothera glazioviana Micheli Oenothera nuda Renner ex Rostański O. biennis O. biennis O. parvilora O. biennis O. biennis × O. glazioviana O. glazioviana O. biennis Oenothera oehlkersii Kappus ex Rostański Oenothera pycnocarpa Atkinson & Bartlett in Bartlett Oenothera royfraseri Gates Oenothera rubricalyx Gates Oenothera rubricaulis Klebahn Oenothera rubricauloides Rostański Oenothera victorinii Gates & Catches. O. biennis × O. glazioviana? O. biennis O. biennis O. glazioviana O. biennis O. biennis? O. biennis Series Rugglesia Rostański * Oenothera angustissima Gates * Oenothera cruciata Nuttall ex G. Don Oenothera issleri Renner ex Rostański * Oenothera oakesiana (A. Gray) J.W. Robbins ex S. Watson & Coulter Oenothera parvilora L. Oenothera subterminalis Gates O. parvilora O. parvilora O. biennis × O. oakesiana O. oakesiana O. parvilora O. parvilora Subsection Raimannia (Rose) Dietrich * Oenothera laciniata Hill Notes on the species recorded in Belgium after 1950 For each of the species included in the key, additional information is provided with respect to its identiication and differentiation from related or similar species, current and/or historical distribution in Belgium, degree of naturalization, etc. The most important synonyms have also been added. This is useful since several species were initially recognized under other names in Europe. And, inally, the probable area of origin of each species is also provided (see before). K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 20 Figure 6. Oenothera biennis. Brugge, E-side of Boudewijnkanaal, sand storage area, August 2007. Characteristic features are the entirely green stem and sepals and the medium-sized petals. After having compared specimens from America and Europe, Rostański (1985) came to the conclusion that ten Oenothera species described from Europe were in fact already previously described from North America. However, part of this synonymy was rejected subsequently (see for instance under O. cambrica and O. ersteinensis). It may well be that none of the species described from Europe exactly match and that they are best referred to using their ‘European’ names (comm. R. Jean, August 2014). • Oenothera biennis L., Sp. Pl. 1: 346. 1753. Figure 6. Probable area of origin: Eurasia? Oenothera biennis doubtlessly was the irst introduced species of the genus in Belgium. Several sources conirm its presence at the end of the 18th century when the species was already locally abundant according to Crépin (1860): “(…) dès 1792, Roucel la signalait déjà comme abondante aux environs de Bruxelles, Gand et Termonde.” Yet, in the last few decades, it has been less often observed and it is even uncertain if it can really be considered as naturalized at present. From the recent distribution map of Flanders (Verloove 2006b) it does appear to be slightly expanding but it is doubtful if this relects reality. Oenothera biennis is easily recognizable by the medium sized petals that are wider than long, the total absence of hairs with red bulbous base, the uniformly green-colored sepals and lower buds, etc. Oenothera delexa and O. victorinii are more or less similar but have clearly smaller petals. Most similar probably is O. cambrica (especially var. impunctata). The latter has slightly narrower leaves, petals that are as long as wide and at least the lower capsules are eglandular. At the coal mining spoil heap of Grande Machine à Feu near Dour existed, for years, a population with pale yellow petals (last conirmed in 2007). Such plants belong to f. sulphurea de Vries. At the same site, it was also observed that this taxon formed hybrids with the coexisting population of Oenothera fallax. At various locations where O. biennis and O. delexa occur sympatrically occasional hybrids between these species have also been observed. • Oenothera cambrica Rostański, Fragm. Florist. Geobot. 23(3-4): 285. 1977. Figure 7. Probable area of origin: Europe? Oenothera cambrica is a rare but possibly overlooked species (see before for the differentiation with the related O. biennis). For a long time it was known only from the British Isles (from where it was initially described; Rostański 1977) but more recently it has also been observed in the northwest of France (comm. D. Mercier). The typical variety (var. cambrica) has a red punctuate stem whilst var. impunctata Rostański has a uniformly green stem. In the last few years both varieties have been collected, in Roeselare (2001) and in Zwijnaarde (2007, 2014). For some taxa which were described in Europe, it inally turned out that these were documented before in America. Rostański (1985) cited ten such cases, including K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 21 = O. velutinifolia Hudziok, Verh. Bot. Vereins Prov. Brandenburg 105: 103. 1968. Probable area of origin: North America. Oenothera canovirens is a very rare species in Belgium. It was last collected in 1959 from a sand raised site in Antwerpen-Linkeroever. Along with O. depressa, it belongs to the series Devriesia. Typically, these species have dense, stiff white hairs (with or without a few, barely visible glandular hairs) on the young fruits and leaves are densely hairy and velvety to the touch. Moreover, the upper leaves are often conspicuously falcate. Confusingly, they easily form hybrids with species of the series Oenothera; these hybrids display a much less typical assemblage of characters. The putative hybrid of O. canovirens and O. subterminalis (= O. wratislaviensis), a taxon that is also found in areas where one or both parental species are absent, has recently been collected on several occasions, including, for example, the port of Antwerp, in the Baai van Heist and at an abandoned coal mining site in Hensies. This hybridogenous species differs from O. canovirens by the clearly red striped lower buds, the red midrib of the leaves and by petals being broader than long. O. canovirens is best differentiated from O. depressa by the lattened or slightly V-shaped bracts and leaves (without wrinkled margins), the chasmogamous lowers and by the complete or almost complete absence of hairs with a red bulbous base. Figure 7. Oenothera cambrica (var. cambrica). Zwijnaarde (Gent), roadside, August 2014. This taxon resembles O. biennis but the stem is clearly red-punctate. Oenothera cambrica. From this, the correct name for O. cambrica appeared to be O. novae-scotiae Gates. However, according to Dietrich (1991) the type of this species is certainly not identical with O. cambrica, nor with O. biennis and probably more closely related to O. parvilora. Whichever taxonomical concept is used, the name O. novae-scotiae cannot be used for O. cambrica. • Oenothera canovirens Steele, Contr. U.S. Natl. Herb. 13(10): 365-366. 1911. = O. renneri H. Scholz, Wiss. Z. Pädagog. Hochschule Potsdam, Math.-Naturwiss. Reihe 2: 206. 1956. • Oenothera delexa Gates, Philos. Trans., Ser. B 226: 332. 1936. = O. lipsiensis Rostański & Gutte, Ber. Arbeitsgem. Sächs. Bot. n.s. 9: 69. 1971. Figure 8. Probable area of origin: North America. Oenothera delexa is, at present, perhaps the commonest species in Belgium. The irst record dates from 1915 (Verloove 2006a) but it has only been spreading vigourously in recent decades (see current distribution map for Flanders in Verloove 2006c). This species was for a long time confused with O. parvilora, a species which does not resemble it at all, except that they both have small petals. O. delexa resembles, for the most, O. biennis and O. victorinii (all with green sepals and lower buds, absence of hairs with red bulbous base, etc.). It differs from O. biennis by its much smaller petals that are not broader than long. O. victorinii, in turn, has petals, hypanthium and capsules that are slightly larger. Gutte & Rostański (1971) described Oenothera lipsiensis from Germany as a new species. Later, it became evident that Gates had already earlier described the same species on the basis of American plant material (as O. delexa) (Rostański 1985), the latter binomial evidently having priority. Although this species is currently, without doubt, the most commonly found species in Belgium, it is elsewhere in Europe much less common. Rostański (2006) concluded that it was more or less widespread in Germany, K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 22 leaves that are velvety to the touch make it a striking and easy to recognize plant. In 2007, it was observed at an abandoned coal mining site in Hensies. An earlier Belgian record for Brussels (Jean 1975) now appears to be referable to O. oakesiana. Just as with Oenothera canovirens, O. depressa easily forms hybrids with species from the series Oenothera and Rugglesia. At a coal mining spoil heap in Beringen-Mijn Figure 8. Oenothera delexa. Zwijnaarde (Gent), roadside, August 2014. This species is recognized by its small-sized petals and the absence of hairs with a red bulbous base. whereas in Sweden its occurrence was considered as very rare. According to ield observations carried out by the second author, it is also to be found in the Netherlands and France, close to the Belgian frontiers. At a sand storage area in the canal zone of Bruges, a hybrid with Oenothera fallax has also been found growing among the parental species. Likewise, in the port of Bruges and at several other locations, putative (unstabilized) hybrids of this species with O. biennis have also been collected. • Oenothera depressa E. Greene, Pittonia 2: 216. 1891. = O. hungarica (Borbás) Borbás, Magyar Bot. Lapok 2: 246. 1903. = O. salicifolia Desf. ex G. Don, Gen. Hist. 2: 685. 1832, nom. inval. Figure 9. Probable area of origin: North America. This species belongs, just as Oenothera canovirens, to the characteristic series Devriesia (see earlier). It differs mostly from the latter by the striking wavy edges of the leaves and bracts, the presence of numerous hairs with a red-bulbous base and the usually cleistogamous lowers. This combination of characteristics as well as the numerous white, eglandular hairs on the young fruits and the Figure 9. Oenothera depressa. Hensies (Les Sartis), abandoned coal mining site, July 2007. This species is softly grey-hairy throughout, its leaves have wavy margins and lowers are often cleistogamous. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 23 a reasonably large population of O. paradoxa is found, at least since 2001. This is a probable (stable) hybrid of O. depressa and O. subterminalis (or O. parvilora). It has striking lower buds that are only suffused with red for the uppermost third but for the remainder it is atypical and dificult to insert in the identiication key. It is moreover unclear if (at least part of) the current concept of O. paradoxa is in agreement with the original concept of this species. [Compare for example Jäger & Werner (2005): “Stg. und Bstandachse nicht getupft” with Hudziok’s original description (Hudziok 1968): “Caulis … rubropunctulatus …”] In addition to O. paradoxa, a probable hybrid of O. delexa and O. depressa has recently been collected in Zeebrugge and Zwijnaarde. In large parts of Central and Eastern Europe, Oenothera depressa is one of the most common species (e.g. Jehlík & Rostański 1979, Rostański et al. 1994, Zlatković et al. 1998, Rostański 2006). It is found especially in dried out riverbeds (e.g. Deschâtres 1954; sub O. strigosa) and similar habitats that are, for the most part, not found in Belgium. It is possible, that it has been overlooked along the River Maas, where it could ind a suitable habitat. However, O. depressa obviously becomes much rarer towards Western Europe (compare also with Rostański 1982). This species is often referred to as Oenothera salicifolia Desf. ex G. Don (1832). This is an invalid name since the epithet was already given to a different species by Lehmann (1824). The oldest valid name for this taxon appears to be O. depressa. • Oenothera ersteinensis R. Linder & R. Jean, Bull. Soc. Bot. France 116: 523. 1970. ?= O. perangusta Gates, Canad. Field-Naturalist 64: 142. 1950. Figure 10. Probable area of origin: North America. Oenothera ersteinensis is one of the relatively easy-toidentify species in Belgium. It is unmistakable due to its striking stem hairs: the thickened red hair base is not conical-shaped but cylindrical (much longer than broad) and often slightly curved. It is further characterized by its (usually) red striped sepals and lower buds, the relatively small petals and the (mostly) deep-red stem. Since 1971, it has, on two occasions, been collected in the vicinity of Liège (Ougrée and Chenée). However, it is most well known in the port of Antwerp where, since 2001, it has been found on a regular basis. At present (2014) it has become one of the most widely spread species in the port area. Also in the Kempen it is not rare at all (comm. R. Barendse). Linder & Jean (1969) described Oenothera ersteinensis as a new species from northeastern France. Rostański Figure 10. Oenothera ersteinensis. Balen (Wezel), August 2010 (lowers and stem) and Grobbendonk, Albertkanaal, July 2013 (hairs). This is a characteristic species with a deep-red inlorescence axis, red striped sepals, medium-sized petals and hairs with a red bulbous base that is longer than wide. In the similar O. rubricaulis, sepals are green and the hair base is conical. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 24 (1985) commented that the species had already been earlier described by Gates on the basis of American plant material (as O. perangusta). Noteworthy, is that Dietrich et al. (1997) considered O. perangusta as a synonym for O. oakesiana, whilst O. ersteinensis was considered as belonging to O. biennis. Both are indeed probably distinct, the Belgian plants belonging to O. ersteinensis, not to O. perangusta. Also, in Rostański et al. (2010) both were eventually accepted as separate species. • Oenothera fallax Renner emend. Rostański, Fragm. Flor. Geobot. 11: 507. 1965. Figure 11. Probable area of origin: Europe. This stable hybrid of Oenothera biennis (♂) and O. glazioviana (♀) parentage is one of the strongest spreading species in Belgium (see also Wittig & Tokhtar 2003). It was already recorded at the end of the 19th century (valley of river Bocq) but its genuine naturalization probably started in the irst half of the 20th century when it began to occur in coastal dunes where O. glazioviana and O. biennis were already naturalized. It is now locally common, especially at sites in the port of Antwerp and Ghent but also in numerous other localities throughout the country. After O. biennis, O. fallax is the most numerous species in the Rhine valley in Germany (Wittig et al. 1999). Oenothera fallax is readily recognizable by its showy red-striped lower buds and its medium sized petals and cannot be confused with any other species. It is mostly found in populations with both parental species but is also found without them. Moreover, this hybridogenous species is also cultivated in gardens. At a sand depot in the canal zone of Bruges, a (nonstabilized) hybrid has been found with Oenothera delexa. • Oenothera fruticosa L., Sp. Pl. 1: 346–347. 1753. subsp. glauca (Michx.) Straley, Ann. Missouri Bot. Gard. 64(3): 403. 1977 [1978]. = O. tetragona Roth, nom. inval. Probable area of origin: North America. This ornamental was recorded only once, in 1998, near a waste recycling factory in Ghent. According to Dietrich (1999) this is one of the most frequently cultivated species but this does not seem to be the case in Belgium. It is a very attractive garden plant with its club shaped fruits, day lowering and for being a perennial (not biennial, the most typical life cycle found in the other taxa occurring in Belgium). For a long time, O. fruticosa was accommodated in a distinct genus (Kneifia Spach). This taxon has long been referred to as Oenothera tetragona, an invalid name. The correct name, according to Wagner (2014), is O. fruticosa subsp. glauca. Figure 11. Oenothera fallax. Grobbendonk, Albertkanaal, July 2013. Typical features of this species are the red striped sepals, the medium-sized petals and the presence of numerous hairs with red bulbous base on the stems. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 25 Figure 12. Oenothera glazioviana. Kortrijk, August 2014. Very characteristic species with large petals, sepals that are suffused with red and stems that are red punctate. Compared with O. hirsutissima, leaves are wider, less hairy and with distinctly wavy margins. • Oenothera glazioviana Micheli, Fl. Bras. 13(2): 178179. 1875. = O. erythrosepala Borbás, Magyar Bot. Lapok 2: 245. 1903. Figure 12. Probable area of origin: Europe. Oenothera glazioviana is one of the few genuinely established species in Belgium. Initially published as O. lamarckiana (misapplied), subsequently as O. erythrosepala (synonym), this species was already well-established in the irst half of the 20th century according to Goffart (1945), especially in the coastal dunes: “se propage au littoral où il abonde dans les dunes (…)”. In 1920, it was already found in the vicinity of the coastal towns of Koksijde and Nieuwpoort. It appears that it is still spreading, as shown on the distribution map for Flanders in Verloove (2006d), with a concentration of its expansion clearly located on the Belgian coast. In Belgium it can only be confused with O. hirsutissima (see earlier), O. rubricalyx (see further) and O. oehlkersii, all with large petals and with stigma branches elevated high above the anthers. From the latter it is best differentiated by the presence of hairs with red bulbous base on the stems and the sepals that are suffused with red. Oenothera glazioviana probably arose in Europe as a result of hybridization or mutation. It is unknown in North America, at least as a wild species. • Oenothera hirsutissima (A. Gray ex S. Watson) de Vries, Mutationstheorie 1: 327. 1901. ≡ O. elata Kunth subsp. hirsutissima (A. Gray ex S. Watson) W. Dietrich. Figure 13. Probable area of origin: North America, Mexico. Oenothera hirsutissima is a rare and ephemeral species in Belgium. Up to the present it has always been found on sand: in 1961 in the sand dunes near to Nieuwpoort (Belgian coast) and, more recently, at a sand storage area alongside the canal zone of Bruges (between 2004 and 2007) and at a sand raised site located at the Kluizendok in the Ghent port area (2007). This species most resembles O. glazioviana (large petals, stigma branches elevated high above the anthers, presence of hairs with red bulbous base, etc.). It is nevertheless differentiated by the typical appressed whitish hairs (absence of glandular hairs) on young leaves and capsules and the narrower, nearly entire leaves that are velvety to the touch. At the site in Bruges it has been observed that O. hirsutissima gradually dimin- K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 26 Figure 13. Oenothera hirsutissima. Brugge, E-side of Boudewijnkanaal, sand storage area, July 2007. This species is much reminiscent of O. glazioviana but it has softly hairy upper leaf surfaces. Its petals often turn orangish. ished, apparently as a result of introgression or hybridization with the accompanying species (especially O. biennis, O. delexa, O. fallax and O. glazioviana). Oenothera hirsutissima is possibly overlooked in Europe. This species is also cultivated but is probably not fully hardy in Western Europe (Dietrich 1999). Also in Germany, it was recently found in the wild (Rostański & Meierott 2006). The systematic position of O. elata, O. hirsutissima and O. hookeri Torr. & A. Gray, the latter two considered as subspecies of the former by Dietrich et al. (1997), is very unclear. Rostański (1985) assigned O. elata s.str. to the series Devriesia, whilst he included both other taxa in the series Linderia. From this complex O. hirsutissima is the most widely spread taxon (Dietrich et al. 1997). • Oenothera issleri Renner ex Rostański, Fragm. Florist. Geobot. 11: 514. 1965. Figure 14. Probable area of origin: Europe. Oenothera issleri is a very rare and ephemeral species in Belgium and has been collected only twice so far. It was seen along a railway track in Jambes (Namur) in 1955 and more recently, in 2014, on bare sandy ground in an Figure 14. Oenothera issleri. Beveren (Verrebroek), bare sandy area, August 2014. Typical features of this species are its oblique stem and lower buds with sepal tips with V-shaped sinus. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 27 Figure 15. Oenothera lindheimeri. Vlissingen (Netherlands), dry grassland, September 2009. Unmistakable by its zygomorphic lowers and previously accommodated in a separate genus (Gaura). industrial area in Verrebroek (Beveren). It is a stable hybrid of O. biennis and O. oakesiana parentage (series Oenothera and Rugglesia respectively). Oenothera issleri is also known from a single population on a coal mining spoil heap in northwestern France, relatively close to the Belgian frontiers (Jean & Delay 2008). Figure 16. Oenothera nuda. Blaton, sand pit, July 2007. Fairly characteristic in being virtually glabrous throughout. • Oenothera lindheimeri (Engelm. & A. Gray) W.L. Wagner & Hoch, Syst. Bot. Monogr. 83: 213. 2007. ≡ Gaura lindheimeri Engelm. & A. Gray, Boston J. Nat. Hist. 5: 217. 1845. Figure 15. Probable area of origin: North America. Oenothera lindheimeri is at present often cultivated as an ornamental in Belgium (‘prachtkaars’). It is unmistakable with its numerous white (or rose) lowers borne on an inlorescence, which is long and drawn out. In recent years it has, now and then, been found on waste ground or levelled soil, often in places where garden waste has been disposed. In Vlissingen (the Netherlands) it has been recorded in dry grassland in the port area, close to the railway station where it persists in small number since its initial discovery in 2009. This species was until recently included in the genus Gaura but molecular phylogenetic analysis has demonstrated that this genus is better included in Oenothera (Wagner et al. 2007). • Oenothera nuda Renner ex Rostański, Nordic J. Bot. 27(2): 138 (-139). 2009. Figure 16. Probable area of origin: Europe. Oenothera nuda is distinctive with its stem, loral tube, capsule and inlorescence that are virtually glabrous. For the rest, it is particularly reminiscent of O. biennis. This species has been found earlier on a few occasions in Wallonia (Houx and Huy). More recently, O. nuda has also been collected from a sand pit near to Blaton and at an abandoned coal mining site in Hensies. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 28 • Oenothera parvilora L., Syst. Nat. (ed. 10) 2: 998. 1759. Probable area of origin: North America. In Belgium, Oenothera parvilora is rare and declining. Nearly all references cited in the literature are in fact referable to O. delexa (see earlier). O. parvilora belongs (along with O. subterminalis) to the series Rugglesia. This group is characterized by a slight to clearly curved inlorescence just before lowering and the sepal points which are separated from the base. Unfortunately, both characteristics are only really clear in fresh material. Moreover, this species is also differentiated from O. delexa by the frequent presence of hairs with a red bulbous base. O. parvilora has, since 1884, been collected on 15 occasions, especially around Brussels. The majority of the inds date from the 1950s and 1960s. Yet more recently, it has also been collected in Halle (near to Brussels) (1978), Heist-aan-Zee (Belgian coast) (2000) and Jemeppe-sur Meuse (1984). Figure 17. Oenothera oehlkersii. Brugge, rough ground, August 2014. This species is much reminiscent of O. glazioviana but it has green sepals and stems. • Oenothera oehlkersii Kappus ex Rostański, Feddes Repert. 96 : 9. 1985. Figure 17. Probable area of origin: Europe. Oenothera oehlkersii is a striking hybridogenous species which cannot be confused with any other species in Belgium. It resembles for the most O. glazioviana (large petals, stigma branches elevated high above the anthers, etc.). It differs from the latter by its uniformly green sepals and lower buds and by the total absence of hairs with a red bulbous base. Oenothera oehlkersii was already collected in Liège in 1884 but all of the remaining inds are (very) recent. Since 2001, this species has been found fairly often (e.g. in Balegem, Bruges, Ghent, leper, Mouscron, Roeselare, Zonhoven, Zwijnaarde), sometimes in large amounts (for example, at an old dumping site alongside the E40 motorway in the vicinity of Zwijnaarde). Currently, it is a fairly regularly found species, partially due to the fact that it is easy to identify. O. oehlkersii has been claimed as a hybrid of O. glazioviana and O. suaveolens Desf. parentage. However, the presence of a long style is a recessive character that rather suggests a mutant (translocation) of O. glazioviana, not a hybrid (comm. R. Jean, January 2009). • Oenothera pycnocarpa Atkinson & Bartlett in Bartlett, Rhodora 15: 83. 1913. = O. chicaginensis de Vries ex Renner & Cleland, Z. Indukt. Abstammungs- Vererbungsl. 66: 275. 1933. Figure 18. Probable area of origin: North America. This species is a recent newcomer to Belgium (the oldest inding dates from 1992 in Queue-du-Bois). It was regularly found, especially since 2007, at several different sites in the port of Antwerp as at a demolition site in the port of Bruges. The largest population (at least 1000 plants) was, also in 2007, discovered at an excavated coal mining spoil heap in Gilly near to Charleroi. At this site it appeared that Oenothera pycnocarpa was more or less established; however, three years after its discovery it completely disappeared, despite its habitat remaining unchanged (comm. M. Lannoy, August 2014). This species bears some resemblance to O. biennis (medium sized petals, green sepals) but differs from it by the presence of hairs with a red bulbous base, the stem which in the lower half is colored red (also often peach-colored) and in the upper half green, at least at the beginning of lowering. Moreover, O. pycnocarpa lowers rather late, from the second half of July to the beginning of August. In the west of Germany it is the 3rd most numerous species (e.g. Wittig et al. 1999, Wittig & Tokhtar 2003). Bearing this in mind, it is possible that it has been overlooked in the east of Belgium. • Oenothera rosea L’Hér. ex Ait., Hort. Kew. 2: 3. 1789. Probable area of origin: Central America. This unmistakable species, which at irst sight resembles more an Epilobium (on behalf of its relatively small, rose coloured petals), was collected on several occasions in the 19th century in Wallonia. Around 2000, it was also observed at an abandoned coal mining site near to Saint Ghislain, where it did not persist (comm. P. Dupriez). K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 29 Figure 18. Oenothera pycnocarpa. Brugge, industrial area, demolition site, August 2007. This species is related to O. biennis but differs from it by its red-punctate stem that is always reddish in the lower half. Figure 19. Oenothera rubricalyx. Rekkem, LAR-transportzone, grassland, July 2014. This species differs from O. glazioviana by its entirely blood red sepals. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 30 In the extreme Southwest of France, Oenothera rosea is commonly naturalized. • Oenothera royfraseri Gates, Philos. Trans., Ser. B 226: 285, ig. 30, 31. 1936. = O. turoviensis Rostański, Fragm. Florist. Geobot. 11: 514. 1965. Probable area of origin: North America. In common with other species, including Oenothera delexa and O. parvilora, this species also has small petals. It can be differentiated from O. delexa by the presence of hairs with a red bulbous base on the stem. By the absence of a curved inlorescence as well as the basally contiguous sepal tips it can be differentiated from O. parvilora. Oenothera royfraseri was found in large numbers at a demolition site in the port of Antwerp in 2007 but not conirmed subsequently. • Oenothera rubricalyx Gates, Rep. (Annual) Missouri Bot. Gard. 20: 133. 1909. ≡ O. glazioviana f. rubricalyx (Gates) Lambinon, Nouv. Fl. Belg., Grand-Duché Luxemb., Nord France, 5th ed.: 1048. 2004. Figure 19. Probable area of origin: North America. A beautiful population of this species was discovered in dry grassland (former sand raised site) in the Lauwe/Aalbeke/Rekkem transport zone (LAR) in 2014. The species occurs in fairly large amounts and looks more or less established. Oenothera rubricalyx is a mutant of O. glazioviana with entirely deep red sepals. It is striking and garden worthy. It seems to be rare in the wild in Europe and was previously only reported from coastal dunes near Etaples in northwestern France (Rostański et al. 1994, Lambinon & Verloove 2012). • Oenothera rubricaulis Klebahn, Jahrb. Hamburg. Wiss. Anst. 31: 12. 1914. Figure 20. Probable area of origin: Europe. Although this species was already collected in 1887 (Antwerp) it has only become more or less widespread in recent years. Especially around the city of Ghent (e.g. in the port area, where it has been observed at least since 1985 and where it is irmly established now) Oenothera rubricaulis has become a reasonably common species. Recently, it has also been seen more often around Antwerp. It differs from most other species by its small to medium sized petals and the deep red inlorescence axis. It is similar to O. ersteinensis; however, the latter has red-bulbous hairs with a cylindrical (not conical) base and mostly redstriped sepals. However, O. rubricaulis most resembles O. rubricauloides (see under). Elsewhere in Europe, Oenothera rubricaulis is also a rapidly expanding species. It is even the second most numerous Oenothera species in Scandinavia (Rostański 2006) and in Eastern Europe, it is considered – along with O. biennis and O. depressa – the species with the highest invasion rate (Tokhtar & Groshenko 2014). Figure 20. Oenothera rubricaulis. Zeebrugge port area (near Baai van Heist), ruderalized sand dunes, July 2007. Characteristic features of this species, compared with the similar O. ersteinensis, are the entirely green sepals and the stem hairs that have a conical base. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 31 • Oenothera rubricauloides Rostański, Ann. Bot. Fenn. 44: 395. 2007. = O. rubricaulis var. longistylis Gutte & Rostański, Ber. Arbeitsgem. Sächs. Bot. 11: 187. 1981. Probable area of origin: Europe. This hybridogenous species was collected in 2007 on two occasions around Ghent, each time at locations where (among others) Oenothera glazioviana as well as O. rubricaulis were found. Rostański (2007), who recently described this taxon, did not cover its possible relationship. A cross between the two mentioned species appears quite possible: O. rubricauloides is in all loral characteristics more or less intermediate between both and the stigma branches are somewhat elevated above the anthers (hence the variety name if considered an infraspeciic taxon of O. rubricaulis). Alternatively, Rostański & Karlsson (2010) suggested a hybrid formula O. muricata × O. biennis. In Scandinavia this taxon is reasonably widespread and for this reason it is considered as an independent species. • Oenothera stricta Ledeb. ex Link, Enum. Hort. Berol. Alt. 1: 377. 1821. Probable area of origin: South America. At the end of the 19th century (at least between 1878 and 1900), Oenothera stricta was more or less established at the old city walls of Antwerp (“station abondante”). Now and then it was also collected elsewhere (e.g. De Panne, Fleurus, Wilsele, etc.) but perhaps always as an ephemeral alien. In the 20th century only two collections are forthcoming: Bellecourt (around 1950) and Angleur (1984). Elsewhere in Europe, O. stricta is, in many places, established, mostly in coastal sand dunes (e.g. France, Great Britain, Portugal, etc.). It obviously is conined to the climatologically milder parts of Europe and may perhaps be not fully hardy in Belgium. Typical for this species are its nearly hemispherical seeds, the capsules which are gradually narrowing towards base and the petals which are turning reddish or pinkish after lowering. • Oenothera subterminalis Gates, Philos. Trans., Ser. B 226: 278, Fig. 26, 27. 1936. = O. silesiaca Renner, Ber. Deutsch. Bot. Ges. 9: 455. 1942. Figure 21. Probable area of origin: North America. Just like Oenothera parvilora, this species belongs to the series Rugglesia and thus combines the drooping tip of the inlorescence (just before lowering) with the sepal tips which are separated from the very base. The petals are mostly larger, the stem has a deeper red color and sepal tips are longer (see key). Oenothera subterminalis is one of the characteristic species found in the coal mining basins of northwest France. Strangely, it is virtually absent at identical habitats found just over the border in Belgium (Borinage) (see Figure 21. Oenothera subterminalis. Hénin-Beaumont (France), coal mining spoil heap, July 2009. This species has narrowly lanceolate leaves, a deep red stem and relatively small petals. Before lowering its inlorescence axis is slightly curved. also Jean 1975). In 1979, it was collected on a single occasion at a spoil heap in Châtelineau. In 2004, it was also collected near the Zeebrugge port area (Baai van Heist) together with O. wratislaviensis, its possible hybrid with O. canovirens (see earlier). This species was described by Renner on the basis of European material (as Oenothera silesiaca). As in similar cases, later it appeared that it was already earlier described in America (as O. subterminalis) (Rostański 1985), although both may be distinct as well (comm. R. Jean, August 2014). • Oenothera victorinii Gates & Catches., J. Linn. Soc., Bot. 49: 182. 1933. = O. nissensis Rostański, Fragm. Florist. Geobot. 11: 508. 1965. = O. rostanskii Jehlík, Folia Geobot. Phytotax. 20(4): 439. 1985. Figure 22. Probable area of origin: North America. Oenothera victorinii mostly resembles O. biennis and O. delexa (see earlier). This species was only collected in Belgium for the irst time in 1992 (Lasne). In the last few years it has certainly been observed more often, for K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 32 Figure 22. Oenothera victorinii. Osłowo, Podlaskie region, Siemiatycze commune (Poland), July 2011. This species much resembles O. delexa but it has slightly larger petals. Figure 23. Oenothera villosa. Brugge, E-side of Boudewijnkanaal, sand storage area, July 2014. This species shares the soft greyish pubescence with O. depressa but its stem is hardly (or not at all) red-punctate, leaf margins are not crispate and lowers are chasmogamous. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 33 instance in Gentbrugge (1998-2004) and Roeselare. At both sites it persisted over the course of several years but eventually disappeared. It was also recently collected at several locations around Antwerp. Just as with the previous species, it turned out only recently that this species was already earlier known in North America and thus was needlessly described from Europe (Rostański 1985). • Oenothera villosa Thunb., Prodr. Pl. Cap. 1: 75. 1794. Figure 23. Probable area of origin: North America. Oenothera villosa was discovered in 2007 in a sand storage area alongside the Boudewijnkanaal in Bruges and regularly conirmed subsequently (ca. 5-10 plants still present in 2014). This species from section Devriesia is much reminiscent of Oenothera canovirens (both with a velvety leaf indumentum, lat leaf margins, hairs with red bulbous bases absent or scarce on the stem, etc.). It is distinguished from the latter by its capsule teeth that are truncate to nearly round at apex instead of emarginate. Acknowledgements. – Raymond Jean (France) and Adam Rostański (Poland) are thanked for their general contribution to this paper. Kevin Balkwill (South Africa), Werner Dietrich (Germany) and Warren Wagner (U.S.A.) kindly provided relevant literature. Michel Lannoy and Rutger Barendse provided chorological data for particular species. Sven Bellanger (Meise) prepared the line drawings. Finally, Jacques Lambinon (Liège) kindly assisted with data from the LG herbarium, and Steven Janssens (Meise) provided information on genetics. References Crépin F. (1860) – Manuel de la lore de Belgique. Bruxelles, Librairie Agricole d’Emile Tarlier. De Langhe J.E., Delvosalle L., Duvigneaud J., Lambinon J., Lawalrée A., Mullenders W. & Vanden Berghen C. (1967) – Flore de la Belgique, du Nord de la France et des régions voisines. Liège, Desoer. Delipavlov D. (1998) – New taxa and chorological data for the lora of Bulgaria. Thaiszia 8: 121-138. Deschâtres R. (1954) – Une espèce adventice nouvelle des vallées de l’Allier et de la Loire: Oenothera strigosa (Rydberg) Mackensie et Bush. Bull. Soc. Bot. Fr. 101(7-9): 348350. Deschâtres R., Rostański K. & Jean R. (2013) – Oenothera ligerica, espèce nouvelle du centre et du sud de la France. Journ. Bot. 61: 17-24. Dietrich W. (1991) – The status of Oenothera cambrica Rostański and O. novae-scotiae Gates (Onagraceae). Watsonia 18: 407408. Dietrich W., Wagner W.L. & Raven P.H. (1997) – Systematics of Oenothera section Oenothera subsection Oenothera (Onagraceae). Syst. Bot. Monogr. 50: 1-234. Durand T. (1899) – Phanérogames. In: De Wildeman E. & Durand T. (eds.), Prodrome de la lore belge, tome III. Bruxelles, A. Castaigne. Frean M., Balkwill K., Goid C. & Burt S. (1997) – The expanding distributions and invasiveness of Oenothera in southern Africa. S. Afr. J. Bot. 63(6): 449-458. Gates R.R. (1933) – Some phylogenetic considerations on the genus Oenothera, with descriptions of two new species. J. Linn. Soc., Bot. 49: 173-197. Gates R.R. (1958) – Taxonomy and genetics of Oenothera. The Hague, W. Junk. Goffart J. (1945) – Nouveau manuel de la lore de Belgique et des régions limitrophes (3e éd.). Liège, Desoer. Gutte P. (1998) – Zur Kenntnis der sächsischen Oenothera-Arten. Abhandlungen und Berichte Naturkundemuseums Görlitz 70(2): 81-94. Gutte P. & Rostański K. (1971) – Die Oenothera-Arten Sachsens. Ber. Arbeitsgem. Sächs. Bot. N.F. 9: 63-88. Haeupler H. & Mür T. (2000) – Bildatlas der Farn- und Blütenplanzen Deutschlands. Stuttgart, Verlag Eugen Ulmer. Hudziok G. (1968) – Die Oenothera-Arten der südlichen Mittelmark und des angrenzenden Flämings. Verh. Bot. Ver. Prov. Brandenburg 105: 73-107. Jäger E.J. & Werner K. (eds.) (2005) – Rothmaler Exkursionslora von Deutschland. Band 4. Gefässplanzen: Kritischer Band. Berlin, Springer Verlag. Jean R. (1975) – Les Oenothera de Belgique dans l’herbier du Jardin Botanique National. Dumortiera 3: 19-25. Jean R. (1990) – Contributions loristiques régionales 19881989. Contributions de Raymond Jean. Bull. Soc. Bot. N. Fr. 43: 18. Jean R. & Delay J. (2008) – L’Oenothère de issler, Oenothera issleri dans le département du Nord. Bull. Soc. Bot. Nord Fr. 61: 37-42. Jehlík V. (1997) – Oenothera. In: Slavík B. (ed.), Kvĕtena České Republiky, vol. 5: 68-94. Praha, Academia. Jehlík V. & Rostański K. (1979) – Beitrag zur Taxonomie, Ökologie und Chorologie der Oenothera-Arten in der Tschechoslowakei. Folia Geobot. Phytotax. 14: 377-429. Jehlík V. & Rostański K. (1995) – Notes on the genus Oenothera subsect. Oenothera (Onagraceae) in the Czech Republic. Folia Geobot. Phytotax. 30: 435-444. Krakos K.N., Reece J.S. & Raven P.H. (2014) – Molecular Phylogenetics and Reproductive Biology of Oenothera Section Kneifia (Onagraceae). Syst. Bot. 39(2): 523-532. Lambinon J., De Langhe J.E., Delvosalle L. & Duvigneaud J. (1998) – Flora van België, het Groothertogdom Luxemburg, Noord-Frankrijk en de aangrenzende gebieden. Derde druk. Meise, Nationale Plantentuin van België. Dietrich W. (1999) – Die Nachtkerzen unserer Gärten. Gartenpraxis 8: 21-29. Lambinon J., Delvosalle L., Duvigneaud J. (2004) – Nouvelle Flore de la Belgique, du Grand-Duché de Luxembourg, du Nord de la France et des Régions voisines. Cinquième édition. Meise, Jardin botanique national de Belgique. Dietrich W. (2000) – Oenothera. In: Castroviejo S. et al. (eds.), Flora Iberica, vol. 8: 90-100. Madrid, Real Jardín Botánico. Lambinon J. & Verloove F. (2012) –Nouvelle Flore de la Belgique, du Grand-Duché de Luxembourg, du Nord de la K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 34 France et des Régions voisines. Sixième édition. Meise, Jardin botanique national de Belgique. Lang W. (2001) – Zum derzeitigen Kenntnisstand der Nachtkerzen-Sippen (Oenothera L.) in der Pfalz und deren angrenzenden Gebieten. Flora Fauna Rheinland-Pfalz 9(3): 751812. Lehmann J.G.C. (1824) – Semina in Horto Botanico Hamburgensi. Hamburg. Linder R. (1958) – Une clé pour la détermination des oenothères. Bull. Assoc. Philom. Alsace Lorr. 10(3): 61-64. Linder R. & Jean R. (1969) – Oenothera ersteinensis, espèce nouvelle. Bull. Soc. Bot. Fr. 116: 523-529. Mihulka S. & Pyšek P. (2001) – Invasion history of Oenothera congeners in Europe: a comparative study of spreading rates in the last 200 years. Journal of Biogeography 28(5): 597609. Otto H.-W. (1970) – Die Nachtkerzen (Oenothera) der Oberlausitz. Abhandlungen und Berichte des Naturkundemuseums Görlitz 45(8): 1-18. Rakaj M. & Rostański K. (2009) – New species of the genus Oenothera L. from coasts of the Albania and Montenegro. Natura Montenegrina 8(3): 163-171. Renner O. (1917) – Versuche über die gametische Konstitution der Oenotheren. Zeitschrift für induktive Abstammungs- und Vererbungslehre 18:121-294. Renner O. (1942) – Europäische Wildarten von Oenothera. Ber. Deutsch. Bot. Ges. 60: 448-466. Renner O. (1950) – Europäische Wildarten von Oenothera. II. Ber. Deutsch. Bot. Ges. 63: 129-138. Renner O. (1956) – Europäische Wildarten von Oenothera. III. Planta 47: 219-254. Richardson D.M., Pyšek P., Rejmánek M., Barbour M.G., Panetta F.D. & West C.J. (2000) – Naturalization and invasion of alien plants: concepts and deinitions. Diversity and Distributions 6: 93-107. Rostański K. (1968) – Neophytism of species of the genus Oenothera L. occurring in Europe. Materialy zakhadu itosocjologii Stosowanej U. W. 25: 67-79. Rostański K. (1977) – Some New Taxa in the Genus Oenothera L. Subgenus Oenothera. Part III. Fragm. Florist. Geobot. 23(3-4): 285-293. Rostański K. (1982) – The species of Oenothera L. in Britain. Watsonia 14: 1-34. Rostański K. (1985) – Zur Gliederung der Subsektion Oenothera (Sektion Oenothera, Oenothera L., Onagraceae). Feddes Repert. 96(1-2): 3-14. Rostański K. (1991) – The representatives of the genus Oenothera L in Portugal. Boletim Da Sociedade Broteriana 64(2): 5-33. Rostański K. (2004) – Kontrowersje wokół pochodzenia dwuletniego (Oenothera biennis: Onagraceae) w Europie. Fragm. Flor. Geobot. Polonica, Suppl. 6: 53-57. Rostański K. (2006) – The occurring of Oenothera species in Scandinavia. Biodiv. Res. Conserv. 1-2: 64-68. Rostański K. (2007) – Three new species of Oenothera from the Nordic countries. Flora Nordica Notes 32. Ann. Bot. Fennici 44: 393-396. Rostański K., Deschâtres R., Dutartre G., Sornicle R. & Jean R. (1994) – Floristique du genre Oenothera L., section Oe- nothera, sous-section Euoenothera, (Onagraceae) en France. Acta Bot. Gallica 141(6/7): 1-13. Rostański K., Dzhus M., Gudžinskas Z., Rostański A., Shevera M., Šulcs V. & Tokhtar V. (2004) – The genus Oenothera L. in Eastern Europe. Kraków, Polish Academy of Sciences. Rostański K. & Forstner W. (1982) – Die Gattung Oenothera (Onagraceae) in Österreich. Phyton 22(1): 87-113. Rostański K. & Karlsson T. (2010) – Oenothera. In: Jonsell B. (ed.), Flora Nordica, vol. 6: 132-148. Stockholm, The Swedish Museum of Natural History. Rostański K. & Meierott L. (2006) – Zur Gattung Oenothera L. in Franken – mit besonderer Berücksichtigung von Oenothera stucchii Soldano (neu für Deutschland). Forum Geobotanicum 2: 19-23. Rostański K. & Ramst U. (2001) – The distribution of Oenothera L. in Estonia. Esti Loodsmuuseumi Töid 4: 44-57. Rostański K., Rostański A., Gerold-Śmietańska I. & Wąsowicz P. (2010) – Evening-Primroses (Oenothera) occurring in Europe. Kraków, Polish Academy of Sciences, W. Szafer Institute of Botany. Rostański K. & Schnedler W. (1991) – Zur derzeitigen Kenntnis der Oenothera-Sippen (Nachtkerzen) in Hessen. Oberhessische Naturwissenschaftliche Zeitschrift 53: 43-117. Roucel F.A. (1792) – Traité des plantes les moins fréquentes qui croissent naturellement dans les environs des villes de Gand, d’Alost, de Termonde et Bruxelles, etc. Paris/Bruxelles, Lemaire/ Bossange. Sell P. & Murrell G. (2009) – Flora of Great Britain and Ireland. Vol. 3 Mimosaceae – Lentibulariaceae. Cambridge, Cambridge Univ. Press. Soldano A. (1980) – Per una migliora conoscenza di Oenothera L., subgenere Oenothera in Italia I. Le specie presenti nel Vercellese. Atti Inst. Bot. Lab. Critt. Univ. Pavia 13: 145-158. Soldano A. (1983a) – Per una migliore conoscenza del genere « Oenothera » L., subgenere « Oenothera » in Italia. II. Descrizione di due nuove specie presenti nella pianura Padana occidentale (Dycotiledonae, Onagraceae). Riv. Piem. St. Nat. 4: 127-135. Soldano A. (1983b) – Per una migliore conoscenza di Oenothera L., subgenere Oenothera (Onagraceae), in Italia. III. Le specie presenti in Versilia e nella Piana Massese (Toscana nordoccid.). Arch. Bot. Biogeogr. Ital. 58 (3-4): 174-189. Soldano A. (1993) – Il genere Oenothera L. subsect. Oenothera, in Italia (Onagraceae). Natura Bresciana, Ann. Mus. Civ. Sc. Nat. Brescia 28: 85-116. Tison J.-M. & de Foucault B. (2014) – Flora Gallica. Flore complète de la France. Mèze, Editions Biotope. Tokhtar V. & Groshenko S. (2014) – Differentiation of the Climatic Niches of the Invasive Oenothera L. (Subsect. Oenothera, Onagraceae) Species in the Eastern Europe. Adv. Environ. Biol. 8(10): 529-531. Tokhtar V., Vinogradova Y.K. & Groshenko A.S. (2011) – Microevolution and Invasiveness of Oenothera L. species (Subsect. Oenothera, Onagraceae) in Europe. Russian Journal of Biological Invasions 2(4): 273-280. Tokhtar V.K. & Wittig R. (2003) – Variability and correlative structure of morphological loral characters in European Oenothera populations. Ukr. Botan. Journ. 60(6): 698-704. van der Meijden R. (2005a) – Boekbespreking: K. Rostański, M. Dzhus, Z. Gudžinskas, A. Rostański, M. Shevera, V. Šulcs K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 35 & V. Tokhtar., The genus Oenothera L. in Eastern Europe. Gorteria 31: 97. van der Meijden R. (2005b) – Heukels’ Flora van Nederland, 23e druk. Groningen, Wolters-Noordhoff. Verloove F. (2002) – Ingeburgerde plantensoorten in Vlaanderen. Brussel, INBO. [Mededeling van het Instituut voor Natuurbehoud 20] Verloove F. (2006a) – Catalogue of neophytes in Belgium (1800-2005). Brussels, National Botanic Garden of Belgium. [Scripta Botanica Belgica 39] Verloove F. (2006b) – Oenothera biennis. In: Van Landuyt W. et al., Atlas van de lora van Vlaanderen en het Brussels gewest: 622. Brussel/Meise, INBO, Nationale Plantentuin van België en Flo.Wer. Verloove F. (2006c) – Oenothera delexa. In: Van Landuyt W. et al., Atlas van de lora van Vlaanderen en het Brussels gewest: 622. Brussel/Meise, INBO, Nationale Plantentuin van België en Flo.Wer. Verloove F. (2006d) – Oenothera glazioviana. In: Van Landuyt W. et al., Atlas van de lora van Vlaanderen en het Brussels gewest: 623. Brussel/Meise, INBO, Nationale Plantentuin van België en Flo.Wer. Verloove F., Lambinon J. & Hoste I. (2014) – Casuals omitted from the sixth edition of Nouvelle Flore de la Belgique. Dumortiera 104: 74-82. Wagner W.L. (2014) – The correct name for a subspecies of Oenothera fruticosa L. (Onagraceae). PhytoKeys 34: 15-17. Wagner W.L., Hoch P.C. & Raven P.H. (2007) – Revised classiication of the Onagraceae. Syst. Bot. Monogr. 83: 1-240. Wisskirchen R. & Haeupler H. (1998) – Standardliste der Farnund Blütenplanzen Deutschlands. Stuttgart, Verlag Eugen Ulmer. Wittig R., Lenker K.-H. & Tokhtar V. (1999) – Zur Soziologie von Arten der Gattung Oenothera L. im Rheintal von Arnheim (NL) bis Mulhouse (F). Tuexenia 19: 447-467. Wittig R. & Tokhtar V.K. (2003) – Die Häuigkeit von Oenothera-Arten im westlichen Mitteleuropa. Feddes Repert. 114(5-6): 372-379. Zlatković B., Anačkov G., Boža P. & Adamović D. (1998) – Distribution of species of the genus Oenothera L. (Myrtales, Oenotheraceae) in Serbia. Thaiszia 8: 43-51. Appendix: overview of specimens examined In this overview all species of Oenothera ever recorded in Belgium (including those recorded exclusively prior to 1950) are alphabetically listed. Records are provided chronologically per province. Records from the current provinces Flemish Brabant and Walloon Brabant are grouped together. Oenothera afinis • Liège: Vovegnez, graviers, 09.1909, P. Halin s.n. (LG); Vovegn­ ez (Wegnez), berge de la Vesdre, 12.09.1909, M. Halin s.n. (BR). Oenothera angustissima = O. rubricuspis Renner ex Rostański • Brabant: Héverlé, talus, 08.1864, O. De Dieudonné s.n. (BR); Tirlemont, talus du chemin de fer, 08.08.1864, H. Vandenborn s.n. (LG); Dieghem, vias ferreas, 1877, C. Baguet s.n. (BR); entre Groenendael et La Hulpe, ligne du chemin de fer, 08.1877, O. Hecking s.n. (BR); • Limburg: St­Trond, talus du chemin de fer, 30.07.1860, H. Vandenborn s.n. (LG); St­Trond, bord de chemin, 24.07.1861, A. Thielens s.n. (BR); St­Trond, 07.1862, A. Thielens s.n. (LG); St­ Trond, talus du chemin de fer, 05.08.1862, H. Vandenborn s.n. (BR); St­Trond, 1864, Vandenborn s.n. (BR); St­Trond, talus du chemin de fer, 08.08.1864, Vandenborn s.n. (BR, LG); St­Trond, talus du chemin de fer, 22.07.1865, H. Vandenborn s.n. (LG); St­Trond, talus du chemin de fer, 22.07.1865, H. van Heurck & A. Martinis s.n. (BR, GENT); St­Trond, talus du chemin de fer, 08.1867, H. Vandenborn s.n. (BR); St­Trond, bords des chemins, cette plante n’existe en Belgique que sur les talus d’un chemin de fer à St­Trond, où elle tend à se naturaliser, 05.08.1867, A. Thielens & A. Devos s.n. (BR, GENT, LG); St­Trond, talus du chemin de fer, 1869, J.­E. Bommer s.n. (BR); St­Trond, 10.1870, P. de Chestret 154 (BR); St­Trond, fortiications, 07.1884, P.J. Delrez s.n. (LG). Oenothera biennis • Antwerp: Anvers, endroits frais, bassin de natation, 06.1855, Tosquinet s.n. (BR); Anvers, Kiel, digue de l’Escaut, 08.1860, A. Vanheurck s.n. (LG); Tongerlo, sous les murs de l’abbaye, 07.1862, Carnoy s.n. (BR); Tongerloo, 07.1863, A. Thielens s.n. (BR); Lierre, bords des chemins, 11.07.1874, L. Piré s.n. (BR); Anvers, fortiications près de la porte de Wilrijk, 23.06.1878, H. Vandenbroeck s.n. (BR); Calmpthout, terrain inculte, 08.07.1882, J. Hennen s.n. (BR); Brasschaet, au Peerdsbosch, station de Vieux Dieu, 07.1896, J. Spas s.n. (BR); Anvers, fortiications à Berchem, 07.1900, R. Godding s.n. (BR); Wilryck, fortise 6, 09.1903, C. Picquet s.n. (BR); Emblehem­sas, helling van den aquaduc, 06.07.1913, E. van Rompaey GIII/1436.1 (BR); He­ renthals, colline, 25.07.1920, V. Lambert s.n. (BR); St. Anna, chemin (naturalisé), 15.08.1934, G. De Gottal s.n. (BR); Ant­ werpen, Zuidervesting, grazige plaats, 17.08.1942, Frison s.n. (GENT); Malines, Coloma, talus du chemin de fer, 15.07.1949, N. Cnops 49.118 (BR); Deurne, op droge grond, 19.07.1951, H. Mer­ vielde s.n. (BR); Brecht, le long de voie ferrée, 07.1953, J. Plancke 104 (BR); Kruisschans, zandgrond, 12.07.1953, E. Jacques 341 (BR) ; Rijmenam (IFBL D5.31.24), zandig terrein, 13.07.1956, A. Jans 127/56 (BR); Postel, Moeren, rudéral, bord de route, 02.09.1956, A. Lawalrée 8078 (BR); Postel, chemin sablonneux près de la tourbière de Postel, 21.09.1956, J. Lambinon (LG); Antwerpen­West, zandgrond, 11.07.1957, Léothade De Ruyver s.n. (BR); Sint­Anna (IFBL C4.26.14), opgespoten gronden, 28.06.1959, E. van Rompaey GIII/1436.1 (BR); Postel, digue d’un étang, 03.08.1965, A. Lawalrée 13185 (BR); Mol­Postel, langs waterkant, 03.08.1965, S. Peeters 123 (BR); Mol­Maat, langs het kanaal aan sluis, 08.1967, H.G. Rabijns 1465 (BR); Turnhout, weg naar Baarle­Hertog, grazige ruigte, niet afgemaaide plaats, 04.08.1974, J. Aerts 74/57 (GENT); Turnhout, Kempenlaan, tus­ sen Gierledreef en Antwerpsesteenweg, opgehoogde braak­ liggende grond langs de weg, 09.07.1976, J. Aerts 76/38 (GENT); • Brabant: Kessel­Loo, chemin de fer (talus), 07.1855, C. Baguet s.n. (BR); près de Bruxelles, naturalisé à Groenendaal, 08.1856, Martinis s.n. (BR); Groenendael, 18.07.1861, L. Piré s.n. (BR); Uccle, sur les hauteurs, 09.1862, L. Van Den Borren s.n. (BR); Betekom, bords du Demer, 20.07.1863, C. Baguet s.n. (BR); Louvain, parc, talus du chemin de fer, 07.1865, C. Baguet s.n. (BR); Forest, talus sablonneux, 06.07.1866, L. & V. Coomans s.n. (BR); Auderghem, 1868 (BR); Forest, champ sablonneux, L. & V. Coomans s.n. (BR); Forest, lieux incultes, 25.07.1875, A. Gravis s.n. (LG); St. Josse­ten­Noode, talus du chemin de fer, 18.06.1883, L. Guelton s.n. (BR); Ottignies, 05.1885, H. Vindevogel s.n. (BR); Louvain, talus du chemin de fer, 02.06.1887, A. Busschodts s.n. (BR); Parck, talus du chemin de fer, près de l’abbaye, 06.1888, C. Piquet s.n. (BR); Blauwput, 29.08.1894, E. Michel s.n. (BR); Wilsele, lieux incultes, 07.1896, C. Baguet s.n. (BR); Uccle (Crab­ begat), 06.09.1918, F. Stockmans s.n. (BR); Forest, derrière le Parc Duden, carrières de sable, 10.09.1919, F. Stockmans s.n. K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 36 (BR); entre Vilvorde et Haeren, talus du chemin de fer, <1920, Vermoesen & Guns s.n. (BR); Louvain, talus du chemin de fer en­ tre les 2 lignes de Malines et d’Aerschot, 30.08.1920, Vermoesen s.n. (BR); Louvain, Blauwput, voie du chemin de fer vers Malines, 30.08.1920, Vermoesen s.n. (BR); Etterbeek­Cinquentenaire, cendrée de la station, 14.07.1923, E. Michel s.n. (BR); Haeren, talus le long des voies du chemin de fer de l’état, 26.09.1923, V. Lambert s.n. (BR); Uccle, <1929, J. Henry s.n. (BR); Vilvorde, remblai du chemin de fer, 02.07.1930, De Vits s.n. (BR); Vilvorde, remblai du chemin de fer, 28.07.1930, V. Lambert s.n. (BR, LG); Louvain, canal, 07.07.1933, R. Mosseray & F. Vandevelde s.n. (BR); Aarschot, terrain inculte le long du chemin de fer de cein­ ture Anvers­Diest, 29.07.1937, E. Michiels s.n. (BR); Parc (lez Louvain), talus de la tranchée de la voie ferrée, 07.07.1941, A. Lawalrée 720 (BR); Genval, décombres, 29.06.1946, J.­E. Rogis­ ter s.n. (BR); Schaarbeek, gare de formation, 07.1947, E. Michel s.n. (BR); Bruxelles, jonction nord­midi, décombres, 08.1947, E. Michel s.n. (BR); Woluwé, bois, 06.07.1950, M. Couteaux s.n. (BR); entre Oud­Heverlee (gare) et Corbeek­Dijle, décombres au bord du chemin, 01.07.1951, A. Lawalrée 3619 (BR); Kessel­Lo, Blauwput, puinen, 12.07.1951, J. Pelgrims s.n. (BR); Anderlecht, décombres, 08.1954, s.c. (BR); Heverlee, wegkant, 01.11.1954, M.­T. De Roo s.n. (BR); Auderghem, coin du Boulevard du Souve­ rain et de l’Avenue Hermann­Debroux, terre remuée cette année pour les travaux du viaduc­autoroute, 23.09.1973, A. Lawalrée 18221 (BR); • East Flanders: Aeltere (Aalter), Craenepoel, s.d., s.c. (GENT); Aeltere, sapinières, 1851, Coemans s.n. (BR) + Gand, chemin de remblais dans les prés aux moines (semences probable­ ment apportées avec les décombres), 1851, Coemans s.n. (BR); Gand St.­Pierre, 08.1854, Scheidweiler s.n. (BR); Tronchiennes (Drongen), 07.1855, s.c. (GENT); Gand, 1871, s.c. (BR); Bellem, station, 07.1873, E. Vandermeersch s.n. (GENT); Bellem, bois (coupes), naturalisé, 19.07.1883, L. Magnel s.n. (BR); Waasmun­ ster, Uilenkasteel, 21.07.1903, Guns s.n. (BR); Assels, oostzijde spoorweg, 15.09.1931, s.c. (GENT); Sint­Amandsberg, spoor­ berm, 23.07.1966, J. Van Den Haute 164 (BR); Gent, even voor­ bij Meulestedebrug, driehoekig terrein tegenover Café “Les Rout­ iers”, kruidenrijke vegetatie met Malva sylvestris, 06.07.1974, E. Robbrecht 638 (BR); Gent, RUG, Plantentuin Maton, K.L. Lede­ ganckstraat 35, braakliggend terrein, zonnig, droog, opgehoopt, 05.08.1976, H. Viane 62 (GENT); Sint­Denijs­Westrem, station, in verwaarloosd terrein, vroeger opslagplaats van goederen en kolen, 03.08.1978, D. Duytschaever 60­61 (GENT); Zelzate, weg Gent­Zelzate via Linkeroever, graafwerken ten N van ex­ ecutieoord Rieme (IFBL C3.33.34), ruigte, 17.09.1982, E. Rob­ brecht 2530 (BR); Appels (IFBL D3.28.24), gestorte grond rand weg, 16.09.1992, K. Thielemans s.n. (BR); Harbour of Ghent, Singel near Farmanstraat (IFBL D3.13.11), ruderal gravelly road verge, 24.06.2000, F. Verloove 4409 (BR); Gent, Hundelgem­ sesteenweg near river Schelde (IFBL D3.23.33), waste ground, 27.06.2004, F. Verloove 5714 (BR); Belzele (N­Gent), N of Ring­ vaart near junction with Brugse Vaart (IFBL D3.11.21), stony bor­ der of canal, common, 24.06.2007, F. Verloove 6777 (BR); Kerk­ brugge (N­Gent) (IFBL C3.52.24 + 53.13), former railway yard, 24.06.2007, F. Verloove 6778 (BR); • Hainaut: Braine­le­Comte, tunnel, 05.07.1949, S. Depasse 88/22 (BR); Quivrain, railway yard, gravelly soil, 04.07.2004, F. Verloove 5726 (BR); Frameries, bas du terril du PASS, 08.2006, D. Geerinck­Coutrez 9701 (BR); Dour, La Grande Machine à Feu (terril) (IFBL G3.53.12­21), coal mine heap, 07.07.2007, F. Verloove 6872 (BR); Mons, Mont Ostène, terril du Pass (IFBL G3.45.41), terril, zone en combustion, 25.07.2007, F. Verloove 6866 (BR); • Liège: Rochefort, s.d., F. Crépin s.n. (BR); Tilff, 06.1871, H. Donckier s.n. (LG); Fraipont, rocailles au bord de la Vesdre, 1872, s.c. (GENT); entre Trooz et Fraipont, lieu rocailleux, 1872, Strail s.n. (BR); Ougrée, bord de la Meuse, 30.06.1872, J.­L. Wathelet s.n. (BR); entre Limbourg et [...], le long de la […], 08.1874, O. De Dieudonné s.n. (BR); environs de Visé, 1876, A. Hardy s.n. (BR); Flaire, 09.09.1883, P.J. Delrez 450 (LG; mixed collection with O. cruciata); Soiron, […], 09.09.1883, P. Delrez s.n. (LG); Modave, jardin, 04.08.1884, G. Evrard s.n. (BR); Esneux, 1888, C. Sladden s.n. (BR); Val­Dieu (Charneux), bord d’un chemin, 07.1890, M. Halin s.n. (BR); Tilff, bords de l’Ourthe (il n’y en avait que 2 ou 3 pieds, j’ai remarqué qu’il était plus abondant près d’Esneux), 26.06.1892, C. Sladden s.n. (BR); Comblain­au­ Pont, 29.07.1902, H. de Luesemans s.n. (BR, LG); Lambermont, 1905, P. Doubleman s.n. (LG); Esneux, sous la Roche aux Fau­ cons, sentier broussailleux, 10.07.1906, A. Maréchal s.n. (LG); Vovegnez (Wegnez), berge de la Vesdre, 07.1907, M. Halin s.n. (BR); Seraing, décombres, 09.08.1907, H. Henin s.n. (LG); Val­ Benoit, terrain vague, 13.07.1908, M. Halin s.n. (BR); Theux, 13.07.1908, s.c. (BR); Liège, 08.1908, P. Doubleman s.n. (LG); Vierset­Barse, gare de Régissa, 08.07.1919, A. Charlet s.n. (LG); Vesdre, 12.09.1920, A. Visé s.n. (LG); Ensival, Vesdre, 07.1940, A. Visé s.n. (LG); Fond des Cris (Chaudfontaine), 09.1946, L. Re­ nard s.n. (LG); Tilff, quai Ste. Anne, sentier longeant la rive droite de l’Ourthe, 18.08.1962, A. Lawalrée 11711 (BR); • Limburg: Hasselt, derrière le cimetière, 1877, O. Geraets s.n. (BR); Wychmael (Beverloo), 16.07.1911, M. Guns s.n. (BR); Curange­Hasselt, voies du chemin de fer, aux environs des at­ eliers de réparation, assez répandu, 07.08.1920, Vermoessen s.n. (BR); Leopoldsburg, sables, 07.1956, J. Lebeau s.n. (BR); Genk (IFBL D7.51.22), port, 26.08.1956, A. Lawalrée 8032, 8034 (BR); Genk­Waterschei, déblais d’usine, 21.07.1958, A. Lawal­ rée 10159 (BR); Bree (‘t Hasselt), langs brug, oever Zuid­Wil­ lemsvaart, 17.07.1962, J. Pelgrims 1454 (BR); Leopoldsburg, langs een weg, 07.1967, H.G. Rabijns 1462 (BR); Zolder, <1998, L. Janssen s.n. (BR); • Luxembourg: Orval, abbaye, 26.07.1856, Gravet s.n. (BR); Forêt d’Orval, coteau, 08.1882, E. Chapuis s.n. (LG); Virton, à la grande route vers Rabais, 08.08.1949, A. Lawalrée 2417 (BR); Lahage, halte, près de la halte de la voie ferrée, 04.07.1963, A. Lawalrée 12499 (BR); limite des communes Mirwart, Arville et Smuid, bord de la voie ferrée, 09.09.1964, A. Lawalrée 12864 (BR); Buzenol, gare, 27.07.1968, M. Lambert s.n. (LG); entre Or­ val et […], route, 24.07.1974, D. Geerinck 782 (BR); • Namur: Anseremme, bords de la Lesse, 07.1867, C.L. Guilmot s.n. (BR); Han, bois à l’entrée de la grotte, 06.1882, P. de Ches­ tret s.n. (BR); Eprave, abondant à la lisière d’un bois, 07.1886, F. Pietquin s.n. (BR); Belvaux, 1889, H. Verheggen s.n. (BR); La Plante, naturalisé, 06.1890, Leclerc s.n. (BR); Belvaux, coteau près du Gouffre de Belvaux, 09.1936, O. Gras s.n. (BR); Eprave, route de Han, 08.1938, O. Gras s.n. (BR); Anseremme, décom­ bres, 06.09.1942, O. Gras s.n. (BR); Houyet, talus du chemin de fer vers Hout, 07.1943, J. Legrain s.n. (BR); Durnal, Pipeti, Vallée du Bocq, rive droite, talus dans une chênaie silicicole, 17.07.1977, E. Sérusiaux 77/B/434 (LG); Namur, chemin du Pont de Bricques à Jambes (IFBL G5.47.11), 10.07.1986, D. Geerinck 3728 (BR); environs de Sosoye, long de l’ancienne voie ferrée, ballast de la voie ferrée, 13.10.1990, D. Geerinck 1569 (BR); vallée de la Moli­ gnée, près de l’abbaye de Maredsous, ballast de la voie ferrée, 13.10.1990, G. Bruynseels 1662 (BR); • West Flanders: Nieuport­Bains, lieux incultes, 23.07.1892, C. Baguet s.n. (BR); La Panne, lieux incultes, 27.07.1892, C. Baguet s.n. (BR); Duinbergen, dunes, 07.1929, E. Dewildeman s.n. (BR); Wenduine, 07.1946, L. Renard s.n. (LG); Comines, bord de la Lys, 28.06.1947, J. Baily s.n. (BR); Adinkerke, sable au bord d’une avenue, 03.09.1954, J.­L. De Sloover 2581 (BR); De Panne, dunes, 08.1956, J. Lebeau s.n. (BR); Avekapelle (IFBL D1.11.11), braakliggende akker, 06.10.1997, H. Ruysseveldt 2007 (BR); Gits (IFBL D1.38.13), ruigte bij het station, 27.08.1987, F. Verloove 6 (BR); Nieuwpoort (IFBL C1.41.11), open plek geixeerd duin, 07.08.1995, H. Ruysseveldt s.n. (BR); Waregem (Eertbrugge) (IFBL E2.26.13), grazige vegetatie berm, brug over autosnel­ weg, 24.06.2003, H. Ruysseveldt 3459 (BR); Brugge, E­side of Boudewijnkanaal (IFBL C2.11.24), sand deposit, 08.08.2007, F. Verloove 6853 (BR). Oenothera cambrica var. cambrica • East Flanders: Zwijnaarde (Gent), Ringvaart (W­side), close to E40 motorway (IFBL D3.32.21), worked­up roadside, 17.08.2014, F. Verloove 10953 (BR). K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 37 • West Flanders: Roeselare, Bruggesteenweg x Leopold III­laan (IFBL D1.48.31), ruderal area, 21.07.2001, F. Verloove 4993 (priv. herb. F. Verloove). • Oenothera cambrica var. impunctata • East Flanders: Zwijnaarde (Z­Gent) (IFBL D3.32.24), road verge, along river Schelde, one specimen, 10.07.2007, F. Ver­ loove 6809 (priv. herb. F. Verloove). Oenothera canovirens • Antwerp: St.­Anna (IFBL C4.26.14), opgespoten gronden, 28.06.1959, E. Van Rompaey 1436­2 (BR); • Liège: Entre Ensival et Pepinster, graviers de la Vesdre, 09.1901, P. Halin s.n. (LG). Oenothera cruciata • Brabant: Forest, 06.1870, Coomans s.n. (BR); • Liège: Flère, Gomélèvai, 09.09.1883, P. Delrez 450 (LG; mixed collection with O. biennis). Oenothera delexa • Antwerp: Anvers­Ouest (rive gauche), 18.08.1955, L. De Ruyver s.n. (BR); Koersel, bord d’un sentier, 23.09.1962, E. Hostie s.n. (GENT); Balen­Nete, route de Postel, bord de la route, 03.08.1965, A. Lawalrée 13157 (BR); Antwerpen­Noord (IFBL C4.16.33), opgespoten gronden ten W van Oosterweel, 04.08.1965, E. Van Rompaey 1436­2 (BR); Hoboken, langs weg, 05.10.1965, M. Van den Wijngaert s.n. (BR); Turnhout, Dombergstraat, tussen Wa­ tertappingstraat en pannenfabriek Tuca, inz. noordelijke boord van de weg, 22.09.1969, J. Aerts s.n. (GENT); Turnhout, kanaal, noordelijke oever nabij de grens Turnhout­Beerse, 25.07.1974, J. Aerts 74/48 (GENT); Mechelen­Battel, oude Gentse Steenweg nabij Heffen, Heembeemd (IFBL D4.27.13), verwilderd langwer­ pig perceel, droog, spoorwegsintels, 14.08.1974, L. Vanhecke 4260 (BR); Lillo (IFBL B4.54.21), opgespoten gebied langs de Galgeschoor, droge open zandgrond, pioniersbegroeiing, 11.09.1974, L. Vanhecke 4318 (BR); Turnhout, Everdongenlaan, enkele meters ten zuiden van de brug over de Aa, op de rand van de weg, 09.07.1976, J. Aerts 76/37 (GENT); Turnhout, kanaal, noordelijke dijk, nabij de grens met Beerse, op de oever van het kanaal, 22.07.1976, J. Aerts 76/48 (BR, GENT); Turnhout, werf De Clercq (IFBL B5.48.11), tussen steenslag, 1 eks., 22.07.1979, A. Vermeijen 79/75 (BR); Mechelen, Mechels Veld, terrain vague près Du Pont de Nemours, 06.07.1980, R. Wechuysen 1023 (BR); Turnhout, Nieuwe Kaai (IFBL B5.47.42), wegenbouw Van Gorp, aardhopen grondwerken, 05.09.1981, J. Aerts 81/79 (GENT); Burcht (comm. actuelle Zwijndrecht) (IFBL C4.25.44), terrain vague sablo­argileux près de l’Escaut, 24.09.1990, J. Lambinon 90/B/504 (BR, LG); Harbour of Antwerp, between 4e and 5e Ha­ vendok (IFBL C4.16.32), road verge, along railway track, 1 ex., 29.07.2007, F. Verloove 6870 (BR); • Brabant: Anderlecht, talus de la voie ferrée à la Petite Ile, 09.1938, G. André s.n. (BR); Evere, terrains vagues, 17.09.1944, E. Hostie s.n. (GENT); Evere, Kerkebeek, puinen van verwoeste huizen bij de kerk, 23.10.1948, E. Michiels s.n. (BR); Jette­Diegem, woest terrein bij den Paelboschwegel, 23.07.1949, E. Michiels s.n. (BR); Kessel­Lo, Blauwput, stort van puinen, 28.08.1951, J. Pelgrims s.n. (BR); Anderlecht, remblai chemin de fer, 09.1951, A. Lefebvre s.n. (BR); Nodebais, décombres, 27.09.1954, L. Muyl­ dermans 858 (BR); Leuven, Keizersberg, wegkant, 15.10.1954, M.­T. De Roo s.n. (BR); Kessel­Lo, Martelarenlaan, braakliggend terrein langs de spoorweg, 25.07.1955, J. Pelgrims 109 (BR, GENT, LG); Hofstade, prairie boissée près grand lac, sol sablon­ neux, 24.09.1955, R. Wilczek 1532 (BR); Auderghem, Parc des Princes, Avenue Leemans, terrain vague, 22.07.1960, A. Lawal­ rée 11030, 11032 (BR); Kessel­Loo, terrain vague, 31.07.1962, J. Bouharmont 1383 (BR); Ixelles (Boondael), Chaussée de Boitsfort, le long du chemin de fer, plante très abondante partout, 08.1965, L. Dubois 1647 (BR); Auderghem, à l’angle de l’Avenue Hermann­Debroux et du Boulevard du Souverain, terre remuée il y a quelques mois à peine lors de la construction de l’autoroute en viaduc, une colonie abondante et uniforme, 23.09.1973, A. Lawal­ rée 18225 (Soc. Ech. Pl. Vasc. Eur. Bass. Médit.: BR, GENT, LG); Auderghem, Rue de la Vignette (IFBL E4.36.24), terrain rudéral­ isé sur sable, 08.1977, D. Geerinck­Coutrez 1686 (BR); Kessel­ Lo, Koning Albertlaan, straatberm, 08.08.1978, s.c. (BR); Haacht, station Wespelaar­Tildonk, rond « La Corbeille » (IFBL D5.42.41), 20.07.1980, R. Wechuysen 1060 (BR); Uccle, Rue de Bourdon, terrain sableux pauvre, pelouse ouverte et rase, 01.08.1980, M. Tanghe s.n. (BR); Anderlecht, Rue Bollinckx, sur terre de remblai, limoneuse, mêlée, 23.07.1981, M. Tanghe s.n. (BR); Strombeek­ Bever, chantier du Ring de Bruxelles, croisement avec le chemin Meise­Heysel, buttes de sol amené, groupement riche en an­ nuelles, 23.09.1981, E. Robbrecht 1592 (Soc. Ech. Pl. Vasc. Eur. Bass. Médit.: BR, GENT); Boitsfort, Avenue de la Forestière (In­ ternational School), pelouse à saules, 04.10.1982, M. Tanghe s.n. (BR); Watermael­Boitsfort, terrain rudéralisé dans le quartier des Pêcheries, 09.1984, D. Geerinck­Coutrez 3380 (BR); Woluwe­ Saint­Lambert (IFBL E4.27.31), prairie près du Cora, 08.1987, D. Geerinck­Coutrez 4372 (BR); Diest (IFBL D6.31.44), oud stort langs de Demer, 07.07.1990, E. Jacques 17572 (BR); Waterma­ el­Boitsfort (IFBL E4.46.42), terrain rudéralisé de la Foresterie, 08.1992, D. Geerinck­Coutrez 6716 (BR, LG); Wemmel, Allée des Peupliers, fasciatie, 1995, Verheyden s.n. (BR); • East Flanders: Langerbrugge, 1932, […] (GENT); Denderleeuw, Grote Steenweg, op uiterwaarden, zandgrond, 22.08.1971, J. Van den Haute 605 (BR); Eke (IFBL D3.41.14), 14.07.1973, D. Duytschaever 7 (GENT); Oudegem, spoorwegbrug over de Dender, berm van de spoorwegbrug (grazig en kruidenrijk), 18.07.1976, E. Robbrecht 1055 (BR, GENT); Denderleeuw (IFBL E3.28.21), recent opgehoogd terrein, 07.08.1976, H. Ruysseveldt 1504 (BR); Welle­Wellemeersen (IFBL E3.18.23), braakliggende akker, 11.09.1980, H. Ruysseveldt 1503 (BR); Gent, westzijde Handelsdok, bij douane (IFBL D3.12.44), grazige braakliggen­ de terreinen (opgevuld voormalig dok), 10.1987, E. Robbrecht 3539 (BR); Appels (IFBL D3.28.24), grazige wegkant op dijk, 15.07.1992, A. Van Den Bergh s.n. (BR); Asbeek (IFBL E4.12.14), braakliggende akker, 22.07.1992, W. De Schrijver s.n. (BR); Denderleeuw (IFBL E3.28.21), verruigd hooiland, 14.07.1993, H. Ruysseveldt s.n. (BR); Bornem (Mariekerke) (IFBL D4.12.44), bij een klein kanaal naar de Schelde, 06.09.1998, D. Geerinck­ Coutrez 9319 (BR); Harbour of Ghent, Singel near Farmanstraat (IFBL D3.13.11), ruderal gravelly road verge, 24.06.2000, F. Ver­ loove 4410 (BR); Gent, border of river Schelde at Gentbrugge­ brug (IFBL D3.23.13), demolition area of former steel­factory, abundant, 08.07.2002, F. Verloove 5120 (priv. herb. F. Ver­ loove); Gent, river Schelde near Rijmstraat, crack in pavement, 27.06.2004, F. Verloove 5715 (BR); Gent, Voorhavenlaan (IFBL D3.12.24), ruderal area, 27.06.2004, F. Verloove 5716 (BR); Zwi­ jnaarde (Z­Gent) (IFBL D3.32.24), former dump site near river Schelde and E40 motorway, 10.07.2007, F. Verloove 6873 (BR); Zwijnaarde (Z­Gent), border of river Schelde (IFBL D3.32.24), ruderal road verge, 10.07.2007, F. Verloove 6874 (BR); Harbour of Ghent, S­side of Kluizendok (IFBL C3.43), artiicial sandscape, 10.07.2007, F. Verloove 6875 (BR); Zwijnaarde (Z­Gent), border of river Schelde (IFBL D3.32.24), road verge, 10.07.2007, F. Ver­ loove 6876 (BR); • Hainaut: Marchiennes­au­Pont, cendrées le long de la voie fer­ rée, 08.1949, J. Duvigneaud s.n. (BR); Mons, bord d’un chemin, 10.1949, F. Buxant s.n. (BR); Jamioulx, talus chemin de fer, 08.1954, J. Lebeau s.n. (BR); Viesville, terrain vague, 29.07.1956, P. Demalsy s.n. (BR); Soignies, carrière de Perlonjour, chantier, 19.07.1969, A. Lawalrée 15859 (BR); Chapelle­lez­Herlaimont, terril n° 7, pente schisteuse, M. Tanghe s.n., 13.08.1970 (BR); Cuesmes, terrain rudéral, 03.10.1976, C. Vanden Berghen s.n. (BR); Baudour, 24.07.1983, S. Depasse 88/102 (BR); Soignies (IFBL F3.58.32), terrain rudéralisé sur la pierre blanc de la carrière de Hainaut, 14.09.1985, D. Geerinck­Coutrez 3618 (BR); Boussu, terril Saint­Antoine, coalmine heap, 04.07.2004, F. Verloove 5719 (BR); Quivrain, railway yard, gravelly area, 04.07.2004, F. Ver­ loove 5727 (BR); Blaton, réserve naturelle « La Grande Bruyère » (IFBL G3.21.41), ancienne sablière, très commun, 25.07.2007, K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 38 F. Verloove 6869 (BR); Quiévrain, gare (IFBL G3.41.44), along railway track, 12.07.2007, F. Verloove 6877 (BR); Hensies (Les Sartis) (IFBL G3.31.44), coal mine heap, 12.07.2007, F. Verloove 6879 (BR); Hornu, terril du Sept (IFBL G3.44.31), coal mine heap, 25.07.2007, F. Verloove 6867 (BR); • Liège: Dison, terrain vague, 06.1915, P. Doubleman s.n. (LG); Vierset­Barse, gare de Régissa, 08.07.1919, A. Charlet s.n. (BR); Ensival, Vesdre, 07.1940, A. Visé (BR); Chaudfontaine, carrières, 16.08.1943, A. Isaäcson s.n. (BR); Fond de Cris (Chaudfontaine), terrain vague, 16.08.1943, A. Visé s.n. (LG); Vallée de la Vesdre, bord de la route Liège­Verviers, 26.09.1960, N. Cnops 60.163 (BR); Angleur, rive gauche de l’Ourthe, en aval de l’île Rous­ seau, terres remaniées, 29.08.1973, J. Duvigneaud 73/B/1131; Ougrée, ancien crassier des usines métallurgiques, bois Saint­ Laurent, 22.07.1974, J. Duvigneaud 74/B/953 (LG); Angleur, gra­ viers déversés dans la plaine alluviale de l’Ourthe, 30.07.1974, J. Duvigneaud 74/B/1013 (LG); Amay, graviers le long de la Meuse, 30.09.1975, W. Bellotte 75/707 (LG); Angleur, rive gauche de l’Ourthe, 20.10.1975, Hechtermans & Monfort 75/752 (LG); Her­ stal (IFBL F7.24.33), décombres d’une vieille usine au nord de la gare, 09.08.1980, R. Wechuysen 1118 (BR); Jemeppe­sur­ Meuse, terril houiller du Bas Laveu, friche en bordure d’une zone en combustion, 25.08.1984, J. Lambinon 84/B/640 (LG); Hamoir (IFBL G7.42.32), au S de l’agglomération, le long de la vallée de l’Ourthe, rive gauche, site rudéral, 13.08.1985, A. Lawalrée 25745 (BR); Angleur (IFBL F7.43.22), pied du remblais du chemin de fer près de l’ancienne gare, 27.07.1986, J. Lambinon 86/B/208 (LG); Angleur (IFBL F7.43.21), pied de mur près du canal de l’Ourthe, à proximité du Pont Marcotty, 03.08.1988, J. Lambinon 88/B/265 (LG); Queue­du­Bois, parking usine Aldi, 04.09.1992, J. Beau­ jean 92/86 (LG); Chênée, décombres au Piedrou, 30.09.1992, J. Beaujean 92/131 (LG); • Limburg: Genk, Waterschei, Horenszee, bord de route, 26.08.1956, A. Lawalrée 8010 (BR); Hasselt, rudéral, 07.1960, L. Delvosalle s.n. (BR); Stokrooi, langs kanaal, 08.07.1960, H. Vannerom s.n. (BR); Genk, aan de spoorweg van de Kolenhaven, 21.07.1967, J. Rammeloo 309 (GENT); Genk­Langerloo, zan­ dige wegberm, 22.07.1967, L. Vanhecke 67/4 (BR); Genk, terrain vague, 02.08.1979, P. Claes s.n. (LG); • Luxembourg: entre Chantemelle et Châtillon, dans une sablon­ nière, 13.08.1968, V. d’Ansembourg 3737 (BR); Forrières, bord de la voie ferrée vers Jemelle, association rudérale, plusieurs petites colonies, au total +/­ 50 pieds, 04.09.1998, A. Lawalrée 26229 (BR); • Namur: Namur, terrain vague, 22.09.1954, J.­L. De Sloover 2581 (BR); Jambes, bord de la voie ferrée près des usines Ma­ terne, 09.1955, J. Lambinon s.n. (LG); Lustin, bord de la voie fer­ rée près de la gare, 04.09.1956, J. Lambinon s.n. (LG); Godinne (îles), colonisatrice de pierres, 06.08.1972, E. Sérusiaux 72/B/38 (LG); Dorinne, rive gauche du Bocq (IFBL H5.28.14), ancienne carrière de Famennien, 20.08.1978, J. Duvigneaud 78B603 (BR); Rochefort, carrière abandonnée, 28.08.1978, B. Bastin s.n. (BR); Namèche (IFBL G6.31.13), friche, 09.08.1983, J. Saintenoy­ Simon s.n. (BR); Noville­sur­Méhaigne (IFBL F5.47.14), gare désaffectée, 10.08.1983, J. Saintenoy­Simon s.n. (BR); Grands Malades (IFBL G5.37.14), friche, 21.08.1983, J. Saintenoy­Simon s.n. (BR); • West Flanders: Brugge, Sint­Michiels, spoorwegberm, 29.07.1973, R. Viane 93 (GENT); bos van Houthulst, ruigte, 11.09.1977, L. Vanhercke s.n. (BR); Haven Roeselare, Beurt­ kaai (IFBL D1.48.43), wegberm, achter een bermmarkering, 04.07.1992, F. Verloove 383 (BR); Rekkem (IFBL E1.58.22), wand greppel langs maïsveld, 29.07.2000, H. Ruysseveldt 2643 (BR); Roeselare, near the railway station (IFBL D1.48.34), rail­ roadbank, several specimens, established, 12.07.2001, F. Ver­ loove 4992 (priv. herb. F. Verloove); Zedelgem, ongebruikte spoorweg, 11.09.2001, H. Ruysseveldt 3009 (BR); Ingelmunster, railway track, 03.07.2004, F. Verloove 5717 (BR); Kortemark, railway yard, gravelly area, common, 08.07.2004, F. Verloove 5721 (BR); Rumbeke­Roeselare (IFBL D1.58.32), former clay­pit Ostyn, dump area, 18.07.2004, F. Verloove 5731 (BR); Brugge, port (IFBL C2.22.11), ruderal area, gravelly soil, 28.06.2007, F. Verloove 6878 (BR, LG) (hybrid); Zeebrugge (port­area), SE of Distrigas­plant, A. Ronsestraat (IFBL B2.42), sandy ruderal area, +/­ 10 ex., 18.07.2007, F. Verloove 6859 (BR); Heist, Baai van Heist (IFBL B2.32.33), seadunes, +/­ ruderalized, 18.07.2007, F. Verloove 6860 (BR); Brugge, Pathoekeweg, W of Nijverheidsdok (IFBL C2.21.22), ruderal area (demolition site), small population of uniform plants, 08.08.2007, F. Verloove 6856 (BR). Oenothera depressa • Hainaut: Hensies, Les Sartis (IFBL G3.31.44), coal mine heap, scattered specimens, 12.07.2007, F. Verloove 6801 (priv. herb. F. Verloove, BR); • Liège: Horto Leod., 1828, R. Courtois s.n. (LG). Oenothera ersteinensis • Antwerp: Harbour of Antwerp, Eerste Havendok (IFBL C4.16.42), sandy, worked­up road verge, nice population, >100 specimens, 22.07.2001, F. Verloove 4991 (priv. herb. F. Ver­ loove, BR, LG); Harbour of Antwerp, SW­side of Hansadok (IFBL C4.15.23), along railway track, 2 ex., 29.07.2007, F. Verloove 6831 (priv. herb. F. Verloove); Retie (IFBL C6.13.21), omgewerkte berm, 08.07.2009, R. Barendse s.n. (BR); Brecht, Groot Schiet­ veld (Brechtse Baan) (IFBL B4.48.24), zandige braakgrond bij in­ gang schietstand, talrijk, 11.07.2010, F. Verloove 8131 (BR); Ant­ werp (Luchtbal) (IFBL C4.17.13), ground heaps, by railway track, 03.08.2014, F. Verloove 10951 (BR); • Liège: Ougrée­Sart Tilman, domaine universitaire, en bordure de la route près de l’Institut de Botanique, 13.08.1971, P. Auquier 1566 (BR); Chenée, décombres au […], 30.09.1992, J. Beaujean 92/131 (BR); • Limburg: Lommel­Balendijk (IFBL C6.25.24), in redelijke ho­ eveelheid (met O. glazioviana, O. biennis), 07.08.2010, R. Barendse s.n. (BR). Oenothera fallax • Antwerp: Harbour of Antwerp, SW­side of Delwaidedok (IFBL B4.45.33), sandy border of a railway­track, numerous specimens, 22.07.2000, F. Verloove 4627 (priv.herb. F. Verloove); • Brabant: Bruxelles, chantiers de la jonction Nord­Midi, 09.1951, A. Lawalrée 3790 (BR); Bruxelles, chantier de la jonction Nord­ Midi, association rudérale, 04.08.1956, A. Lawalrée 7775 (Soc. Fr. Ech. Pl. Vasc. n° 2982; BR); Auderghem, Avenue Chaudron, terrain vague, remblai, 11.07.1958, A. Lawalrée 10037 (BR); Rijmenam (IFBL D5.31.22), langs de baan aan ‘t Zwart Water, 16.08.1967, N. Cnops 67.567 (BR); Sint­Jans­Molenbeek, braak­ liggend terrein, 22.06.1971, J. Loots 105 (BR; mixed with O. glazioviana); Haeren, près de Bruxelles, gare de triage, talus ru­ déralisé, 29.08.1987, G. Bruynseels 1257 (BR); • East Flanders: Haven van Gent, Gent­Zelzate via Linkeroever, graafwerken ten N van executieoord Rieme (IFBL C3.33.34), ruig­ te, 17.09.1982, E. Robbrecht 2529 (BR); Port of Ghent, Farman­ straat near Kennedylaan (IFBL D3.13.13), roadsides, railways and rough places, ca. 15 individuals in old railway bed, 19.08.1986, E. Robbrecht 3122 (BR); Zwijnaarde (Z­Gent), langs Ringvaart (IFBL D3.32.24), voormalige zandopspuiting, enkele planten met O. biennis, 23.06.2001, F. Verloove 4822 (BR); Sint­Denijs­Wes­ trem, oprit E40 richting Oostende, zandig braakliggend terrein, talrijk, 21.06.2004, F. Verloove 5655 (BR); • Hainaut: Dour, terril Grande Machine à Feu, coalmine heap, 04.07.2004, F. Verloove 5720 (BR); Hensies (Les Sartis) (IFBL G3.31.44), coalmine heap, 04.07.2004, F. Verloove 5718, 5722, 5723 & 5725 (BR); • Liège: Neupré, Rotheux­Rimière, jonction des routes de Marche et d’Esneux, bord rudéralisé de la route, 20.07.1979, P. Auquier 5069 (LG); • Limburg: Waterschei, ancien […], près de […], 09.1956, J. Lambinon s.n. (BR); • Namur: vallée du Bocq, bords des eaux, 08.1889, Dens s.n. (BR); between Freyr and Waulsort along the river Meuse, 06.09.1991, R. Govaerts 33 (BR); K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 39 • West Flanders: Middelkerke, dune, 25.07.1934, P. [...] s.n. (BR); Le Coq, wegkanten, 14.08.1939, L. Standaert s.n. (BR); Koksijde, dans les dunes, très répandue partout, 07.1966, L. Dubois 1646 (BR); Koksijde, près du moulin, 20.08.1970, R. Boutique 808 (BR); St­Idesbald, dune rudéralisée, 28.07.1974, B. Bastin s.n. (BR); Knokke, Duinbergen, restje duin tussen villa’s langs Kon. baan, 01.11.1978, R. Viane 697 (GENT); Knokke­Heist, dunes rudérali­ sées, 07.1982, D. Geerinck­Coutrez 688 (BR); Nieuwpoort (IFBL C1.41.11), open plekjes geixeerd duin, 07.08.1995, H. Ruyssev­ eldt s.n. (BR); Wervik (IFBL E1.46.24), tussen resten afgebroken fabrieksgebouw, 01.08.2000, H. Ruysseveldt 2647 (BR); Brugge, Pathoekeweg, W of Nijverheidsdok (IFBL C2.21.22), ruderal area (demolition site), 1 ex., 08.08.2007, F. Verloove 6855 (BR). Oenothera fruticosa subsp. glauca • East Flanders: Gent­West, Vinderhoute, Vinderhoutse Meersen (IFBL D3.11.22), ruderale inrijlaan van een afvalverwerkend bedri­ jf langs de R4, samen met andere tuinvluchtelingen, 05.07.1998, F. Verloove 3381 (BR). Oenothera glazioviana • Antwerp: Hoboken, braakland aan de staalgieterij, 02.07.1937, E. Frison s.n. (GENT); Wildert, barre grond, zand, 01.08.1953, E. Jacques 354 (BR); Essen, ’t Hoefken, adventief in Arnosere­ to­Scleranthetum, 09.10.1956, J. Traets 520 (BR); Duffel (IFBL D4.18.13), op spoorwegdam, 02.09.1957, N. Cnops 57.313 (BR); Antwerpen (IFBL C4.36.23), afgegraven vestingwal, ruig­ te, 18.08.1967, E. Van Rompaey GIII/1435­1 (BR); Antwerpen (IFBL C4.36.23), afgegraven vestingwal, ruigte, 21.08.1967, E. Van Rompaey GIII/1435­1 (GENT); Duffel, bord de la route Lierre­Malines, à l’ouest de la voie ferrée Anvers­Malines (IFBL D4.18.13), talus, 02.07.1970, A. Lawalrée 16128 (BR); Turnhout, Smalvoortstraat (IFBL B5.47.32), bouwwerf, 21.07.1979, J. Aerts 79/25 (GENT); ‘s Gravenwezel, oude baan (IFBL C4.28.22), aan de voet van de oprit van de brug over het Albertkanaal, zandige wegkant, 21.07.1986, E. Jacques 15852 (BR); Haven van Ant­ werpen, N­Delwaidedok (IFBL B4.45.13), spoorwegemplace­ ment, 08.10.2001, E. Molenaar s.n. (BR); • Brabant: Park­Heverlee, spoorweg naar Waver, 01.08.1909, E. Michiels s.n. (BR); Keerbergen, marécage, 12.08.1920, E. Hostie s.n. (GENT); Wilsele, talus du chemin de fer, 19.07.1923, J. Lebrun s.n. (BR); Uccle, sablière, 07.1928, J. Lebrun 2680 (BR); Aarschot, woest terrein achter het station, 30.07.1937, E. Michiels s.n. (BR); Louvain, Mont­César, pelouses, 29.09.1940, A. Lawalrée 525 (BR); Aarschot, woest terrein achter het sta­ tion, 02.07.1941, E. Michiels s.n. (BR); Uccle, talus de la voie ferrée, 18.07.1951, G. André s.n. (BR); Bruxelles, chantier jonc­ tion Nord­Midi, rudéral, 04.08.1956, A. Lawalrée 7779 (BR); Kessel­Loo, terrain vague, 31.07.1962, J. Bouharmont 1382 (BR); Kessel­Lo, spoorwegberm, 18.06.1964, J. Pelgrims 1822 (BR); Sint­Jans­Molenbeek, braakliggend terrein, 22.06.1971, J. Loots 105 (BR; mixed with O. fallax); Auderghem, Jardin Massart, grande parcelle expérimentale, 03.08.1976, M. Tanghe s.n. (BR); Auderghem, Chaussée de Wavre (IFBL E4.37), bord de route, 07.1977, D. Geerinck­Coutrez 1591 (BR); près de Haren, gare de formation, terrain rudéralisé, 01.09.1987, G. Bruynseels 1247, 1552 (BR); Ramsberg (ten E van Tienen) (IFBL E6.21.21), stortp­ laats tussen maïsakkers en laagstamboomgaarden, 02.10.2002, L. Vanhecke s.n. (BR); • East Flanders: Haven van Gent, Gent­Zelzate via Linkeroever, graafwerken ten N van executieoord Rieme (IFBL C3.33.34), ruigte, 17.09.1982, E. Robbrecht 2528 (BR); Haven van Gent, Gent­Zelzate via Linkeroever (km 7.0), bij executieoord (IFBL C3.33.44), grote zandhopen, 16.08.1984, E. Robbrecht 2829 (BR); Gent, Dampoort, railway yard, now demolished, 10.07.2004, F. Verloove 5728 (BR); • Hainaut: Montigny­sur­Sambre, colonise certains terrils de charbonnages, 05.08.1930, Culot s.n. (BR); Jamioulx, le long du chemin de fer, terrain vague, 07.1954, J. Lebeau s.n. (BR) ; Beau­ vechain, friche, 01.10.1954, L. Muyldermans 841 (BR); • Liège: Visé, bord de la Meuse, 08.1868, E. Marchal s.n. (BR); Vierset, bois de Faqueval, 08.1889, A. Charlet s.n. (LG); Lam­ bermont, décombres, 07.1908, P. Halin s.n. (LG); Theux (Hod­ bomont), 09.08.1948, J.­M. Warlet 538­2584 (BR); Angleur, do­ maine universitaire du Sart Tilman, le long de la route axiale, bord de la route, accotements remués, 06.07.1971, A. Lawalrée 16675 (BR); Ougrée, Sart Tilman, domaine universitaire, en bordure de la route près de l’Institut Botanique, 13.08.1971, P. Auquier 1566 (LG); Angleur, talus en bord de route, 04.07.1973, J. Beaujean 167 (LG); Tilff, route aux abords du château de Colonster, bord de route, 09.08.1973, J. Duvigneaud s.n. (LG); Jupille­sur­Meuse, bord de route, 14.09.1979, J. Beaujean 79/146 (LG); Ben­Ahin (IFBL G6.24.14), milieu rudéralisé, 01.09.1983, J. Rousselle s.n. (LG); Gué (?) [Ben­Ahin], IFBL G6.24.14, terrain vague, 01.09.1983, J. Saintenoy­Simon s.n. (BR); • Limburg: Bokrijk, deux pieds dans un lieu cultivé, 28.07.1889, H. Dupont s.n. (LG); Berlingen, bij ‘t Kasteel van Rullingen, gras­ boord, 02.08.1934, E. Michiels s.n. (BR); Genk (IFBL D7.51.22), port, 26.08.1956, A. Lawalrée 8033 (BR); Waterschei, ancien crassier près d’un charbonnage, 09.1956, J. Lambinon s.n. (LG); Wellen, wegboord, 18.07.1958, E. Michiels s.n. (BR); Beringen, vaartkant, 27.08.1959, E. Jacques 3939 (BR) ; Beringen, langs een weg aan de mijnterril, 07.1967, H.G. Rabijns 1464 (BR); Genk, aan de spoorweg van de kolenhaven, 21.07.1967, J. Ram­ meloo 310 (GENT); Genk, Kolenhaven, langs de spoorlijnen, abondant in droge zandige vegetatie, 22.07.1967, H. Stieperaere s.n. (GENT); Genk, terrains vagues, 17.08.1980, P. Claes s.n. (LG); Lanaken (Neerharen), Hochter Bampd (IFBL E7.15.11), grindbank langs de Maas, 02.10.2010, F. Verloove 8289 (BR) ; • Luxembourg: Poncelle (Tintigny), prairie aride (sable), assez abondant, 20.08.1945, V. d’Ansembourg 599 (BR); Buzenol, jardin près de la voie ferrée, 08.1949, J. Duvigneaud s.n. (BR) ; Poncelle, terrain sec, 13.09.1946, J. Legrain s.n. (BR); Torgny, 20.09.1959, A. Louette s.n. (LG); • Namur: Lustin, bord de la voie ferrée, 04.09.1956, J. Lambinon s.n. (BR); Andenne, dans des décombres, 1962, F. Siebertz s.n. (LG); • West Flanders: Nieuport, ruines, 1920, A. Navez s.n. (BR); Houthulst, forêt, 01.07.1920, Guns s.n. (BR); Coxyde, dunes, 10.08.1920, L. Magnel s.n. (BR); Clemskerke, pâtur­ ages, 31.08.1923, A. Isaacson s.n. (BR); Westende, bord de la route, 07.09.1924, J. Lebrun s.n. (BR); Koksijde, dunes ixées, 20.08.1928, A. Maréchal s.n. (LG); Duinbergen, 08.1931, E. Dewildeman s.n. (BR); Middelkerke, 07.1946, J. Duvigneaud s.n. (BR); Wenduine, 07.1946, L. Renard s.n. (LG); Ostende, partout, 06.07.1947, A. Visé s.n. (BR); St­Idesbald, dunes vois­ inages des villas, 02.09.1947, V. d’Ansembourg 1025 (BR); Oostduinkerke, 08.1948, L. Renard s.n. (LG); Groenendijk, dunes herbeuses, 12.08.1949, A. Maréchal s.n. (LG); Koksijde, duinen, 14.07.1952, J. Pelgrims s.n. (BR); St­Idesbald, dunes ru­ déralisées, 15.08.1953, J. Duvigneaud s.n. (BR); Koksijde, ten W van de Normandie, rand weg, 17.07.1957, E. Jacques 3080 (BR) ; Coxyde (Saint­Idesbald), dunes, 06.07.1958, A. Lawalrée 9991 (BR); Coxyde­Bains, près du moulin, dunes rudéralisées, 16.07.1960, A. Lawalrée 11078 (BR); Nieuport, dunes, 10.1961, P. Auquier 462 (LG); Le Coq, dunes, 05.08.1962, J.­E. Rogister s.n. (BR); Nieuport, sables +/­ rudéralisés, 07.1963, L. Delvosalle s.n. (BR); Sint­Idesbald, dunes près de la localité balnéaire, 08.07.1965, A. Lawalrée 13112 (BR); Koksijde, dans les dunes, très répandue partout, 07.1966, L. Dubois 1648 (BR); De Panne, droge wegkant aan het standbeeld Leopold I, 29.07.1967, E. Robbrecht 67.13 (BR); Saint­Idesbald (sud du village), panne rudéralisée, 09.07.1968, P. Auquier 1142 (LG); Koksijde, Doorn­ panne, begroeide duinpanne, 10.07.1973, M. Lejeune 84 (BR) ; Hoeke, Steenoven, wegberm in zware kleigrond, 15.07.1973, R. Viane 94 (GENT); Oostduinkerke, duinen, 07.07.1974, J. Loots 571 (BR); De Panne, verkaveling “Westhoek”, zuidelijke weg (IFBL C0.56.41), verruigde Salix repens­Rosa pimpinell.­zoom aan de rand van Ligustrum­struweel met Euonymus europaeus, 03.09.1977, P. Goetghebeur 2790 (GENT); De Panne (IFBL C0.56.44), dunes, 01.11.1978, P. Sotiaux s.n. (BR); Coxyde, dunes rudéralisées, 20.09.1979, W. Debaisieux s.n. (LG); Kok­ sijde, wegkant, 23.07.1980, R. Vanquathem 110 (BR); Oost­ K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 40 duinkerke, Doornpanne, dépression dunaire, 14.10.1982, M. Tanghe s.n. (BR); Coxyde, dune rudéralisée, 27.07.1983, J. Beaujean 83/10 (LG); Heist­Duinbergen, Zeegrasstraat (IFBL B2.32.43), in verruigde duinen, braakland, 21.08.1983, J. Van Den Haute 1455 (BR); Nieuwpoort­Bad, duinen, 08.09.1984, J. Loots 1066 (BR); Oostende, Fort Napoleon (C1.14.42), talus sa­ blonneux en contrebas de dunes rudéralisées, 09.1991, R. Fabri 1524 (Soc. Ech. Pl. Vasc. Eur. Occ. Bassin. Médit. n°15322: BR, LG); Diksmuide, industrieterrein (IFBL D1.23.32), droge grond, 21.06.1992, F. Verloove 286 (BR); Kortemark (IFBL D1.26.42), spoorwegemplacement, op grintpuin, 24.06.1992, F. Verloove 311 (BR); N382 tussen Roeselare en Izegem, ter hoogte van kanaal naar Leie te Kachtem (IFBL D2.51.13), wegberm, onder de vangrail, 18.07.1998, F. Verloove 3184 (BR); Adinkerke, road towards Bray­Dunes (IFBL D0.16.11), seadunes, near railway­ track, locally abundant, 26.06.2000, F. Verloove 4624 (priv.herb. F. Verloove); N382 Roeselare­Izegem, near drive­in E403 (IFBL D2.51.11), ruderal area, levelled soil (former dumping place), 09.07.2000, F. Verloove 4626 (priv.herb. F. Verloove). Oenothera hirsutissima • East Flanders: Harbour of Ghent, S­side of Kluizendok (IFBL C3.43), artiicial sandscape, few specimens, 24.06.2007, F. Ver­ loove 6769 (priv. herb. F. Verloove); • West Flanders: Nieuport, dunes, 10.1961, P. Auquier 462 (BR); Brugge, E­side of Boudewijnkanaal, sand depot, 17.07.2004, F. Verloove 5734, 5735 & 5736 (priv.herb. F. Verloove, BR); Brugge, E­side of Boudewijnkanaal (IFBL C2.11.24), sandy ruderal area, few specimens, known since several years (+ hybrids with several other taxa), 28.06.2007, F. Verloove 6768 (priv. herb. F. Verloove); Brugge, E­side of Boudewijnkanaal (IFBL C2.11.24), sand depos­ it, 18.07.2007, F. Verloove 6863 (BR) (hybrid); Brugge, E­side of Boudewijnkanaal (IFBL C2.11.24), sand deposit, 18.07.2007, F. Verloove 6865 (BR) (hybrid). • Liège: Val­Benoit, décombres, 12.07.1923, A. Maréchal s.n. (LG); • Limburg: St­Trond, bords de chemin, 05.08.1867, A. Thielens (BR). Oenothera oehlkersii • East Flanders: Gent, Sneppebrug (IFBL D3.22.13), road verge, near railway track (disturbed soil), 24.06.2007, F. Verloove 6774 (BR); Zwijnaarde (Z­Gent) (IFBL D3.32.24), voormalige stortp­ laats langs de E40, 10.07.2007, F. Verloove 6810 (BR, LG); • Liège: Angleur, fortiications, 07.1884, P.J. Delrez s.n. (BR); • West Flanders: Roeselare, Bruggesteenweg (IFBL D1.48.13), demolition area of former Coca­Cola­factory, several specimens, 21.07.2001, F. Verloove 4994 (priv. herb. F. Verloove, BR); Rum­ beke­Roeselare, former clay­pit Ostyn, dump, 18.07.2004, F. Ver­ loove 5732 (BR); Ieper, railway yard, gravelly area, 27.07.2004, F. Verloove 5730 (BR). Oenothera paradoxa • Limburg: Beringen­Mijn (IFBL D6.15.13), S­side of coal mine heap, common, 08.07.2001, F. Verloove 4848 (priv. herb. F. Ver­ loove); Beringen­Mijn (IFBL D6.15.31), coal mine heap, common, 01.07.2007, F. Verloove 6812 (priv. herb. F. Verloove). Oenothera parvilora • East Flanders: Vrasene, Aven Ackers (industrial area) (IFBL C4.13.31), bare sandy area, 31.08.2014, F. Verloove 11203 (BR). • Namur: Jambes, Materne, bord de la voie ferrée, 09.1955, J. Lambinon s.n. (BR). • Brabant: Dieghem, talus du chemin de fer, 1884, A. Douret s.n. (BR); Bruxelles­Nord, chantier de la jonction Nord­Midi, secteur Botanique, 21.06.1947, A. Lawalrée 1627 (BR); Bruxelles, plaine de l’Exposition de 1935, à 100 m du Planetarium, 08.08.1948, E. Michel s.n. (BR); Bruxelles, chantier jonction, 19.08.1958, M. Coûteaux 742 (BR); Bruxelles, terrain vague, 19.09.1972, A. La­ walrée 17972 (BR); Halle, gare, 02.09.1978, S. Depasse 88/85 (BR); • Hainaut: Viesville, près du canal, rudéralia, 10.08.1955, A. La­ walrée 6870 (BR); Braine­le­Comte, talus SNCB, 22.07.1957, S. Depasse 88/26 (BR); Tertre, 05.08.1967, S. Depasse 88/69 (BR); • Liège: Jemeppe­sur­Meuse, terril houillers du Bas Laveu, friche en bordure d’une zone en combustion, 25.08.1984, J. Lambinon 84/B/640 (BR); • Luxembourg: Halte de Rossart, chantier de bois, 30.08.1965, V. d’Ansembourg 3372 (BR); • West Flanders: Baai van Heist (O­Zeebrugge) (IFBL B2.32.33), ruderal sandy area, near the sea, abundant, 08.09.2000, F. Ver­ loove 4623 (priv.herb. F. Verloove). Oenothera laciniata Oenothera perennis • Brabant: Wilsele, lieux incultes, 07.1893, C. Baguet s.n. (BR); Wilsele, dépendances de l’usine Bodart, 07.1894, C. Baguet s.n. (BR, GENT, LG); Schaarbeek, environs de la gare, 1907, A. Isaacson s.n. (BR); Anderlecht, immondices, 05.07.1920, E. Michel s.n. (BR); • Liège: Juslenville, voie ferrée, 27.09.1903, M. Halin s.n. (BR); Vovegnez, graviers, 12.09.1909, P. Halin s.n. (LG); Vovegnez, berge de la Vesdre, 12.09.1909, M. Halin s.n. (BR); Liège, subs­ pontané dans le jardin du collecteur, 1924, A. Maréchal s.n. (BR); • West Flanders: Coxyde, entre les ils de fer barbelés au pied du Hoogenblikker près du nouveau chemin empierré vers Oost­ duinkerke, un seul pied, 28.09.1919, L. Magnel s.n. (BR). • Antwerp: Capellenbosch, zandgrond, 20.06.1943, L. Standaert s.n. (BR). Oenothera indecora • Brabant: Corbeek­Loo, champ de trèles, 08.1895, C. Baguet s.n. (BR). Oenothera issleri Oenothera pycnocarpa • Hainaut: Hensies, Les Sartis (IFBL G3.31.44), coal mine heap, 12.07.2007, F. Verloove 6880 (BR) (hybrid); Blaton, sablière le long du canal (IFBL G3.21.42), sur sable, 25.07.2007, F. Verloove 6868 (priv. herb. F. Verloove); • Liège: Huy, terrain vague, 16.08.1956, A. Lawalrée 7784 (BR); • Namur: Houx, taillis, 01.10.1933, R. Mosseray s.n. (BR). • Antwerp: Harbour of Antwerp, S­side of 5th Havendok (IFBL C4.16.33), road verge, talus, few specimens, 29.07.2007, F. Verloove 6834 (BR); Harbour of Antwerp, SW­side of Hansadok (IFBL C4.15.23), along railway track, 1 ex., 29.07.2007, F. Ver­ loove 6835 (BR); Waaslandhaven, Kallo, Sint­Jansweg (IFBL C4.14.12), ruderal sandy area, in front of Sagrex sand storage, very numerous, 27.07.2014, F. Verloove 10925 (BR); • Hainaut: Gilly, au NO du cimetière (IFBL G4.48.11), le long d’une route, creusée dans les lancs d’un ancien terril, peuple­ ment très important (1000 pieds au moins), 07.2007, M. Lannoy s.n. (priv. herb. F. Verloove 6854, BR, LG); • Liège: Queue­du­Bois, parking usine Aldi, 04.09.1992, J. Beau­ jean 92/86 (BR); • West Flanders: Brugge, Pathoekeweg, W of Nijverheidsdok (IFBL C2.21.22), waste ground (demolition area), locally abun­ dant, 08.08.2007, F. Verloove 6852 (priv. herb. F. Verloove, BR). Oenothera oakesiana Oenothera rosea Oenothera nuda • Brabant: Bruxelles, travaux de 26.06.1945, A. Lawalrée 1303 (BR); la jonction Nord­Midi, • Hainaut: Ath (Tournai), s.d., Ronday s.n. (BR); Ath, 1874, Ron­ day s.n. (BR); K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 41 • Liège: Vervian, Leodis, sponte in oleraciis, s.d., R. Courtois s.n. (LG). Oenothera royfraseri • Antwerp: Harbour of Antwerp, angle of Amerika­ and Albertdok (IFBL C4.16.41), ruderal area, bare and gravelly, very common, 29.07.2007, F. Verloove 6833 (priv. herb. F. Verloove). Oenothera rubricalyx • West Flanders: Rekkem, LAR­transportzone (IFBL E2.41.32), grassland, several populations, 11.07.2014, F. Verloove 10892 (BR). Oenothera rubricaulis • Antwerp: […], bords de chemin, 07.1887, P. Halin s.n. (BR); Harbour of Antwerp, S­side of Delwaidedok, railway track, 15.08.2004, F. Verloove 5748 (BR); • Brabant: Aarschot, woest terrein achter het station, 06.07.1939, E. Michiels s.n. (BR); • East Flanders: Haven van Gent, Meulestedekaai, aan Meulest­ edebrug (IFBL D3.12.22), kanaalberm met naakte min of meer dichtgeslagen zandbodem, massale populatie, 05.07.1985, E. Robbrecht 2929 (BR); Gent, spoorwegberm tussen Ringvaart en Watersportbaan (IFBL D3.21.24 + D3.22.13), grindrijk terrein, zeer talrijk, 09.10.2005, F. Verloove 6230 (BR); Gent­West, Ring­ vaart x railway track (IFBL D3.21.24), railway track, formerly a massive population, now partly demolished after infrastructural works, 24.06.2007, F. Verloove 6775 (BR); Harbour of Ghent, Singel (E of Grootdok) (IFBL D3.13.11), ruderal area, abundant, 24.06.2007, F. Verloove 6779 (BR); • West Flanders: Heist, Baai van Heist (IFBL B2.32.33), sea­ dunes, +/­ ruderalized, 18.07.2007, F. Verloove 6861 (BR). Oenothera victorinii • Antwerp: Harbour of Antwerp, right bank of river Schelde, SW of 5th Havendok (IFBL C4.15.43), worked up road verge, several specimens, 29.07.2007, F. Verloove 6830 (BR); • Brabant: Lasne, rue du Double Ecot, carrière abandonnée, 12.07.1992, F. Billiet & B. Jadin 2331 (BR); • East Flanders: Gentbrugge, Hundelgemsesteenweg (IFBL D3.23.33), braakgrond, nabij de oevers van de Schelde, 02.08.1998, F. Verloove 3357 (BR); Gent, Hundelgemsesteenweg near river Schelde (IFBL D3.23.33), ruderal area, 10.07.2004, F. Verloove 5729 (BR); • West Flanders: N382 Roeselare­Izegem, near drive­in E403 (IFBL D2.51.11), ruderal area, levelled soil (former dumping place), several specimens, along with O. glazioviana, 09.07.2000, F. Verloove 4625 (priv. herb. F. Verloove); Roeselare, N382 at drive­in E403 (A17) (IFBL D2.51.11), levelled soil, persistent, irst recorded in 2000, 13.07.2001, F. Verloove 4996 (LG). Oenothera villosa • West Flanders: Brugge, E­side of Boudewijnkanaal (IFBL C2.11.42), sand deposit Hanson, one specimen, 18.07.2007, F. Verloove 6816 (priv. herb. F. Verloove); Brugge, E­side of Boudewijnkanaal (IFBL C2.11.42), sand storage, +/­ 5 specimens, 21.07.2014, F. Verloove 10917 (priv. herb. F. Verloove, BR). Oenothera wratislaviensis • Antwerp: Harbour of Antwerp, S­side of 5th Havendok (IFBL C4.16.33), road verge, talus, abundant, 29.07.2007, F. Verloove 6832 (priv. herb. F. Verloove); • Hainaut: Hensies, Les Sartis (IFBL G3.31.44), coalmine heap, 04.07.2004, F. Verloove 5724 (priv. herb. F. Verloove); • West Flanders: Heist, Baai van Heist, sandy grassland near the sea, 17.07.2004, F. Verloove 5733 (priv. herb. F. Verloove, BR). Oenothera rubricauloides • East Flanders: Gent­West, Ringvaart x railway track (IFBL D3.21.24), railway track, 24.06.2007, F. Verloove 6776 (BR); Zwi­ jnaarde (Z­Gent) (IFBL D3.32.24), former dump site, 10.07.2007, F. Verloove 6811 (priv. herb. F. Verloove). Oenothera speciosa • Hainaut: Saint­Ghislain, s.d., Van Bastelaer s.n. (BR). Oenothera stricta • Antwerp: Anvers, fortiications près de la porte de Wilrijk, 23.05 & 23.06.1878, H. Vandenbrouck s.n. (BR); Anvers, fortiications, station abondante, 23.06.1878, H. Vandenbrouck s.n. (BR); An­ vers, fortiications, 08.1882, H. Vandenbrouck s.n. (BR); Anvers, remparts, 31.08.1882, J. Hennen s.n. (BR); Anvers, fortiications, 07.1884, P.J. Delrez 451 (LG); Anvers, remparts, 29.07.1884, J. Hennen s.n. (BR); Anvers, remparts, 07.1887, J. Hennen s.n. (BR, LG); Anvers, fortiications à Berchem, 07.1900, R. Godding s.n. (BR); • Brabant: Wilsele, lieux incultes, 07.1893, C. Baguet s.n. (BR); • Hainaut: Saint­Amand­lez­Fleurus (IFBL G5.21), 06.1883, Grosse s.n. (BR); Bellecourt (IFBL G4.23), sur les talus du chemin de fer, plante naturalisée, <1951, A. Briart s.n. (BR); • Liège: Angleur, friche en bordure d’une zone en combustion, 25.08.1984, J. Lambinon 84/B/640 (LG); • West Flanders: La Panne, 05.06.1881, W. Rouseau s.n. (BR). Oenothera subterminalis • Hainaut: Châtelineau, Châtelet, terril, face sud SW, 11.06.1979, J­C Moniquet s.n. (BR); • West Flanders: Heist, Baai van Heist, seadunes, +/­ ruderal­ ized, 02.08.2004, F. Verloove 5738 (BR). Non-stabilized hybrids Cf. Oenothera biennis × O. delexa • Antwerp: Harbour of Antwerp, Vosseschijnstraat, E of 2e Ha­ vendok (IFBL C4.16.42), railway track, 29.07.2007, F. Verloove 6871 (BR); • East Flanders: Gent, Leon Tertzweillaan (IFBL D3.23.13), omgewoelde, eroderende, zandige, grintrijke spoorwegberm, pio­ niervegetatie met veel adventieven, 06.09.1977, P. Goetghebeur 2802 (GENT); Handzame (IFBL D1.26.33), hoop gestorte grond langs wegkant, 28.08.1999, H. Ruysseveldt 2389 (BR); • West Flanders: Brugge, harbour, E of Groot Handelsdok (IFBL C2.22.11), former railway track + ruderal area, 18.07.2007, F. Ver­ loove 6864 (BR). Oenothera biennis × O. glazioviana • Luxembourg: Buzenol, coupe dans le bois de la Ferme de Buzenol, le long du ruisseau, 08.1949, J. Duvigneaud s.n. (BR). Cf. Oenothera delexa × O. fallax • West Flanders: Brugge, E­side of Boudewijnkanaal (IFBL C2.11.24), sand deposit, 08.08.2007, F. Verloove 6857 (BR). Cf. Oenothera depressa × O. delexa • East Flanders: Zwijnaarde (Z­Gent) (IFBL D3.32.24), entrance of former dump area, one specimen, 10.07.2007, F. Verloove 6808 (priv. herb. F. Verloove); • West Flanders: Zeebrugge (port­area), N of Verbindings­ dok (IFBL B2.41.42), artiicial sandscape near Tropicana­plant, 18.07.2007, F. Verloove 6858 (BR); Zeebrugge­Strand, W­side of port (IFBL B2.41.24), sandy ruderal soil, near railway station, 2 ex., 08.08.2007, F. Verloove 6851 (BR). K. Rostański and F. Verloove, The genus Oenothera in Belgium [Dumortiera 106/2015: 12-42] 42