Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
This account presents information on all aspects of the biology of Pulmonaria officinalis L. (Common Lungwort) that are relevant to understanding its ecological characteristics and behaviour. The main topics are presented within the framework of the Biological Flora of the British Isles: distribution, habitat, communities, responses to biotic factors, responses to environment, structure and physiology, phenology, floral and seed characteristics, herbivores and disease, history and conservation.
Pulmonaria officinalis is a distylous, perennial rosette hemicryptophyte that is naturalized in Britain. It occurs predominantly in the understory of broadleaved, mixed, open woods rich in hornbeam, beech and oak, but is has also been recorded from hedges, banks alongside streams, (sunken) roadsides, built-up areas, gardens, rubbish tips, yew woodland and, less frequently, coniferous woodland. Its native range stretches from southern Sweden, in the north, to northern Italy, in the south, and from western Germany to Poland and Lithuania. It is most common in central Europe.
Pulmonaria officinalis reproduces both by sexual and vegetative means. The underground parts consist of a slowly creeping rhizome with adventitious roots. Despite the presence of an elaiosome that is attractive to ants that likely facilitates seed dispersal, populations often show a significant fine-scale spatial genetic structure and small neighbourhood sizes.
The leaves and flowering stalks are covered in hairs of varied length and stiffness, and stems and inflorescences are sparsely covered with glandular hairs. Flowers of P. officinalis are heterostylous, with distinct pin and thrum morphs. The corolla varies from purple, violet or blue to shades of pink and red, or sometimes white. The flowers are mainly pollinated by solitary bees and bumblebees.
Pulmonaria officinalis has significantly expanded its distribution in the British Isles since the 18th century, especially in southern England. P. officinalis is currently not regarded as threatened in Europe. Nonetheless, small, isolated populations of the species often show reduced reproductive success and lower genetic diversity, potentially affecting their long-term survival. Preservation of local habitat conditions by regular opening of the forest canopy and restoration of gene flow among populations are required to maintain viable populations of P. officinalis in the long term.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Common Lungwort. Boraginaceae, tribe Boragineae. Pulmonaria officinalis L. is a bristly, semi-evergreen, perennial rosette hemicryptophyte. Rosettes arising from long, thin, horizontal, slowly creeping rhizomes with adventitious roots. Basal rosette leaves 16 × 10 cm, hispidulous, ovate, pointed at the apex, cordate at the base narrowing into winged stalks up to 15 cm; leaf lamina light green with clearly delineated pale green to silvery-white spots or blotches. Flowering stems originating among the leaf axils of rosettes, unbranched, 10–20 (30) cm tall with small clasping leaves. Inflorescence a terminal scorpioid cyme; flowers actinomorphic, bisexual, distylous, 5–15 per stem. Corolla changing during anthesis from red through purple to violet and finally blue, rarely white; petals 5, connate, not pubescent except for the hair ring at the entrance of the corolla tube; lobes 6.3–6.8 mm in width; tube about equalling calyx, 11.6 × 2.0 mm. Calyx pubescent with five triangular teeth, swelling during fruit set. Nectaries in a ring at the bottom of the corolla tube around the ovary base where nectar accumulates. Stamens 5, epipetalous, inserted at the apex of corolla tube in thrum-eyed flowers, at the middle of tubes in pin-eyed flowers. Gynoecium syncarpous, consisting of two carpels divided by false septa into four locules each containing one ovule, a single style arising from a central depression in the superior four-lobed ovary. Long-style stigmas positioned beyond the mouth of the corolla tube above the ring of floral hairs, short style half as long. Fruit a schizocarp of nutlets consisting of four nutlets with a single seed surrounded by a protecting hard pericarp. Nutlets 3.4 × 2.6 mm, c. 7.5 mg, downy, ovoid and acute, dark brown to blackish at maturity, with fleshy white elaiosome of irregular shape, up to 0.93 mm in diameter, surrounded by the raised ring at the base of the nutlet.
The genus Pulmonaria L. contains about 18 species that are native to Europe and Western Asia (Sauer 1975; Bolliger 1982; Hewitt 1994; Bennett 2003), but their taxonomy is subject to debate (Kirchner 2004). Gams (1927) considered that all Central European species of Pulmonaria could be traced back to four Tertiary species including P. officinalis (Fig. 1). Merxmüller & Sauer (1972) disputed this theory because of non-correspondence of chromosome counts of putative hybrids and parental species. Sauer (1975) combined morphological, karyological and biogeographical data to delineate eight groups of pure species and a small ninth group containing hybrids. A recent phylogenetic study of nuclear and chloroplast DNA showed that hybridization in the genus rarely occurs, and variation in combination with convergent evolution in morphological characteristics has led to the misinterpretation of pure species as hybrids (Kirchner 2004).
Pulmonaria officinalis is a variable species that strongly resembles Pulmonaria obscura Dumort., both in morphology and chromosome number. Both species also have a strongly overlapping Central to East European distribution range (Hultén & Fries 1986). As a result, delimitation of the two species is difficult, and they have often been treated as subspecies (Hegi 1927; Hempel 1981; Hultén & Fries 1986; Schmeil & Fitschen 2009). The close relatedness of these two taxa compared with other Pulmonaria species was recently confirmed by a phylogenetic study and supports their delimitation as subspecies (Kirchner 2004):
Pulmonaria officinalis ssp. officinalis. Leaves with clearly delineated white spots, semi-evergreen. Calyx V-shaped completely covered with glandular hairs.
Pulmonaria officinalis ssp. obscura Dumort. (Suffolk Lungwort). Unspotted or rarely with faint, green spots along the veins. Leaves darker than ssp. officinalis and only summer green. Calyx U-shaped without glandular hairs.
In this account, we refer to P. officinalis spp. officinalis, unless otherwise stated.
Occasionally, plants with white flowers and white spots are intermingled in natural populations; these are usually associated with gardens and are therefore believed to be cultivars that have escaped from cultivation (e.g. P. officinalis ‘Sissinghurst White’).
Pulmonaria officinalis is a naturalized herb, with a wide but predominantly southern distribution in the British Isles. It typically occurs in the understory of broadleaved, mixed, open woods rich in hornbeam, beech and oak.
I Geographical and altitudinal distribution
Pulmonaria officinalis is considered naturalized in Great Britain (Hegi 1927; Preston, Pearman & Dines 2002; Hill, Preston & Roy 2004). It has been recorded in 682 (24%) 10-km grid squares in Great Britain, most frequently in southern England, with a sparser distribution in the Scottish highlands and occurring in only three and two 10-km grid squares on the Isle of Man and the Shetland Islands, respectively. P. officinalis is absent from the Inner and Outer Hebrides, Orkney and Channel Islands (Fig. 2). In Ireland, the species is rare and only found in the north-east, in 8 (0.8%) of 985 10-km squares covering Ireland (Fig. 2; Preston, Pearman & Dines 2002). It is mainly found in lowland areas, reaching a maximal altitude of 385 m in Forest-in-Teesdale, Durham (Dahl 1998; Preston, Pearman & Dines 2002; Pearman & Corner 2013). The ssp. obscura is native, although it has been found in only three woodlands that are all located within a single 10-km square in Suffolk (TM07; Birkinshaw & Sanford 1996; Stace 2010).
Pulmonaria officinalis is native to Central and Eastern Europe, reaching north to 55°–56° N in Skåne, southern Sweden, and south to 41°–45° N in Italy (Tuscany) and the Balkans (Fig. 3; Hultén & Fries 1986). Eastwards, the range extends to Romania and Macedonia, and extends as far as western Poland (Merxmüller & Sauer 1972) and Lithuania in the NE (Fig. 3; Hultén & Fries 1986). The western range margin lies in the region of the Upper Rhine, as far as Voeren in eastern Belgium and South-Limburg in the Netherlands (Merxmüller & Sauer 1972; Weeda et al. 1988). In Austria, the species is widespread, whereas in Switzerland, it only occurs by Lake Constance (German border), in the valley of the Thur and Rhone, at the Canton of St. Gallen, in Schaffhausen, Aargau, Neuenburg at the Lake Neuchâtel, Freiburg and Ticino (Hegi 1927). Pulmonaria obscura (or ssp. obscura) has a wider and more northerly European distribution including Finland and Russia.
Outside its native range, in Western Europe, P. officinalis can be locally abundant (Belgium, the Netherlands; Fig. 3). Although the exact history of this species in this part of Europe is unknown (Weeda et al. 1988; Van Landuyt et al. 2006), its occurrence is often linked with the historical presence of a monastery or castle (Weeda et al. 1988) and is therefore considered as an established, ancient (before 1600 A.D.) garden escape (‘stinzenplant’, Bakker & Boeve 1985). According to Preston, Pearman & Dines (2002), P. officinalis is not native in much of Western Europe.
On the European mainland, P. officinalis can reach altitudes above 1000 m a.s.l. (Hegi 1927). In Slovakia, P. officinalis is found up to 450–520 m in the Kremnica Mountains (Tatra) (Schieber 2007) and in the Apuseni Mountains (Carpathians) in north-western Romania (Laviniu 2010). In Bavaria, Germany (Alps), P. officinalis occurs up to 1230 m, in Tirol, Austria up to 1720 m, in Sexten, Italy even up to 1900 m (Hegi 1927).
(A) Climatic and topographical limitations
Pulmonaria officinalis is a shade-tolerant herb and is categorized as a European temperate species by Hill, Preston & Roy (2004). It occurs mainly in deciduous forests and, less frequently, in yew woodland or coniferous woodland, but has also been recorded in more open places, such as along hedges, stream banks and ruderal habitats (roadsides, built-up areas, gardens, rubbish tips; Hegi 1927; Preston, Pearman & Dines 2002; Hill, Preston & Roy 2004). The mean January and July temperatures of the 10-km squares occupied by P. officinalis in Britain are 3.4 and 15.3 °C, respectively, with a mean annual precipitation of 922 mm (Hill, Preston & Roy 2004). The Ellenberg continentality value of 5 indicates that P. officinalis occurs in an intermediary climate between the Atlantic coast climate and the continental climate of inner Asia (Ellenberg et al. 1991).
Pulmonaria officinalis occurs generally on slightly damp to wet, base-rich loamy soils (Ellenberg pH value R =8). It grows abundantly on basaltic soils and limestone in the understory of broadleaved, mixed, open woods rich in hornbeam, beech and oak (Hegi 1927; Runge 1980; Ellenberg 1988; Hill, Preston & Roy 2004). In southern Sweden, P. officinalis is found on soils with a pH range between 4.5 and 7.0, but it prefers soils with a pH above 5.0 (Falkengren-Grerup 1986). Pulmonaria officinalis is a moist-site indicator and prefers intermediate to highly fertile soils (Ellenberg value for moisture F =5, for nitrogen N =6).
In Britain, soils where P. officinalis occurs are often derived from calcareous parent material such as sedimentary limestones, shales, clays and superficial glacial drift deposits (Rodwell 1991). The species is further found on a broad range of soils between rendzinas and brown calcareous earths, on the one hand, and brown podzolic and true podzols on the other (Rodwell 1991). It is absent from saline sites (Hill, Preston & Roy 2004).
The communities in which P. officinalis is found in the British Isles are not well known. Based on the British plant communities of Rodwell (1991), it can be assigned to woodland communities with a fertile soil that are moderately to deeply shaded. These communities include Fraxinus excelsior–Acer campestre–Mercurialis perennis (W8), a community found on calcareous mull soils on limestone in the relatively warm and dry lowlands of southern Britain (Rodwell 1991). Pulmonaria officinalis also occurs as a component of semi-natural oakwoods that were formerly used for coppice such as Quercus robur–Pteridium aquilinum–Rubus fruticosus woodland (W10) and that are typically distributed over the lowlands of England (Keith Kirby, pers. comm.). The species may also occur in woodlands where Fagus sylvatica has attained dominance on free-draining, base-rich and calcareous soils, that is, in Fagus sylvatica–Mercurialis perennis communities (W12). These communities mainly occur in the south-eastern lowlands of Britain, especially in Kent, East and West Sussex, and Hampshire (Rodwell 1991), and may explain the high abundance of P. officinalis in this region.
In Central Europe, P. officinalis was assigned by Ellenberg (1988) to the Lamiastrum ecological group of plants growing under broadleaved woodland, comprising also Lamiastrum galeobdolon, Carex sylvatica, Geum urbanum, Primula elatior, Vinca minor and Paris quadrifolia. This ecological grouping prefers base-rich soils ranging from fairly dry to slightly damp. Pulmonaria officinalis is mainly found in mixed woods rich in hornbeam (Carpinion) and in beech and mixed beech woodlands (Fagion) of Central Europe where it often occurs with Primula elatior, Anemone nemorosa, Ficaria verna, Scilla bifolia, Stellaria holostea and Symphytum tuberosum in the understorey of ancient woodlands (Hegi 1927; Ellenberg 1988; Hermy et al. 1999). In Fagion woodlands, P. officinalis occurs in average to rich brown-mull beech woods such as slightly moist limestone beech woods (Galio odorati-Fagetum), melick-beech woods (Melico-Fagetum) and sycamore- or high-altitude beech woods (Aceri-Fagetum) (Runge 1980; Ellenberg 1988; Oberdorfer 1992). Ellenberg (1988) denoted P. officinalis as a ‘negative’ differentiating species together with Arum maculatum, Mercurialis perennis, Sanicula europaea and Stachys sylvatica that only survive in soils with higher lime content, for dividing the poorer brown-earth mull where these species are lacking from the limestone beech woods. Pulmonaria officinalis is also a widely distributed species in mixed oak-hornbeam forests (Carpinion) such as wood bedstraw–oak–hornbeam (Galio-Carpinetum), stitchwort–oak–hornbeam (Stellario-Carpinetum) and Wild arum–oak–hornbeam woods (Querco-Carpinetum aretosum) (Runge 1980; Ellenberg 1988; Oberdorfer 1992).
Besides its occurrence in mixed beech and hornbeam woodlands, P. officinalis is found in mixed woods of more damp sites near springs and along associated small streams rich in ash and alder (Alno—Ulmion). The Carici remotae—Fraxinetum and Equiseto telmateiae—Fraxinetum associations are similar in that both communities are often narrow, associated with streams and characterized by high abundance of Equisetum telmateia in the herb layer. These communities often occur on soils that are saturated with highly oxygenated, eutrophic water, where water stagnation does not occur (Runge 1980; Oberdorfer 1992; Cornelis et al. 2009). In addition to Fraxinus excelsior as dominant tree species, Carici remotae—Fraxinetum contains Alnus glutinosa as well, but only in the Equiseto telmateiae-Fraxinetum, where their ranges overlap can P. officinalis and P. obscura occur together (Oberdorfer 1992). In Belgium, these two communities are confined to the few regions where P. officinalis is typically found (Cornelis et al. 2009). A third association (Pruno-Fraxinetum) is broader than the previous associations, mostly occurring in the more or less flat parts of rivulet valleys, where the water table reaches 20–70 cm below ground level. Although this habitat type is regularly to occasionally inundated, water slowly seeps into the soil so that peat formation rarely occurs (Oberdorfer 1992; Cornelis et al. 2009). A fourth major type of ash–alder (Alno—Ulmion) woods, Querco-Ulmetum with which P. officinalis is often associated, mainly occurs along larger streams/rivers (e.g. Rhine, Danube) where groundwater fluctuations are common, and therefore, fewer species of moist soils occur. Oberdorfer (1992) also documented P. officinalis in floodplain forests as a component of Salicetum albae and in pine-forests of Erico—Pinetum sylvestris where it usually occurs at low frequency.
IV Response to biotic factors
The roots (but not the leaves) of P. officinalis contain pyrrolizidine alkaloids, which are toxins produced by the plant as a defence against herbivory (Dobler et al. 2000). The flea beetle Longitarsus lateripunctatus, whose larvae feed on the roots (the adults eat the leaves), is a specialized feeder on P. officinalis and is capable of sequestering the pyrrolizidine alkaloids present in the roots of its host plant to circumvent the plant's defence mechanism (Dobler et al. 2000). The hairy foliage of P. officinalis presumably reduces damage by slugs, snails, deer and rodents (Bennett 2003).
V Response to environment
Pulmonaria officinalis can be found as scattered individuals, as linear populations alongside streams and rivers and, if conditions (e.g. light) are favourable, it can develop dense, conspicuous populations that dominate the local vegetation, with rosette densities up to 13 m−2 (Fig. 4a; S. Meeus, pers. observ.). In a dense Belgian population, the position and morph type of each flowering individual were recorded within a plot of 20 × 8 m (Fig. 4a). Local densities varied between 0 and 8, and 0 and 10 m−2 for pin and thrum morphs, respectively. Plants also showed significant spatial clustering, particularly plants of the pin morph (< 5 m; Fig. 4b). This pattern is best explained by the weak self-incompatibility system of this morph and asymmetrical pollen deposition (see 'Floral and seed characters' A, C).
(B) Performance in various habitats
Pulmonaria officinalis is a semi-shade plant with an Ellenberg value of 5, indicating that the species prefers places with more than 10% relative illumination when trees are in leaf (Hill, Preston & Roy 2004). Although the species is shade-tolerant, flowering is likely to be reduced in heavy shade (Bennett 2003). The lack of coppicing in Belgian populations has been considered responsible for a decline in population size and density, resulting in bottlenecks and leading to the observed skewed morph frequencies in small populations in northern Belgium (Van Landuyt et al. 2006; Brys, Jacquemyn & Beeckman 2008b).
(C) Effect of frost, drought, etc
Pulmonarias in general are very hardy and P. officinalis can tolerate minimum winter temperatures down to −29 °C, which makes them attractive as garden ornamentals in western Europe and states such as Oregon and northern California in the USA, where the climate is not too extreme (Bennett 2003). In a semi-evergreen species such as P. officinalis, however, rosette leaves formed during summer often wilt during hard winters. Papp et al. (2011) investigated variation in histological features between windy/shady habitats and sunny habitats in P. officinalis. Trichomes were found to be significantly longer in windy/shady habitats serving as a higher level of protection against strong wind exposure.
VI Structure and physiology
Pulmonaria officinalis is a rosette hemicryptophyte (Fig. 5). The ‘summer’ or rosette leaves are covered with trichomes that reach high densities on the adaxial side of the leaf and lower densities on the abaxial side. Both the stem with the ‘spring’ leaves and inflorescences are covered in hispid hairs and sparse, minute, non-glutinous glandular hairs (Hegi 1927). Gaberščik et al. (2001) reported an average specific leaf weight of 0.441 g dm−2, a leaf trichome density of 2.39–2.42 mm−2 and trichome length of 573–709 μm. Often more than three flowering stems emerge from the overwintering rosette in spring, although the stems elongate and flowers mature at different rates so that one individual may continue to flower for almost the entire flowering season. The stem is characterized by brown scales at the base where it joins the underground parts that consist of thin, branched rhizomes with numerous adventitious roots (Fig. 5a,b).
Harley & Harley (1986) reported that about half of all species of the Boraginaceae were examined for mycorrhizas and two-thirds of these can form arbuscular mycorrhizas. Pulmonaria officinalis is, however, only occasionally associated with symbiotic fungi (Fitter & Peat 1994). One of the three sources quoted by Harley & Harley (1986) that recorded mycorrhizas in P. officinalis reported a double infection, that is, with a second fungal endophyte (in most cases assigned to the genus Rhizoctonia). Two other sources reported absence of mycorrhiza in P. officinalis.
(C) Perennation: reproduction
Pulmonaria officinalis is a polycarpic perennial. It is able to reproduce clonally through woody rhizomes that form daughter rosettes near the mother plant. However, P. officinalis is reported not to spread clonally in Britain (Hill, Preston & Roy 2004). Meeus, Honnay & Jacquemyn (2012b) found clonal diversities (G/N, where G is the number of genets and N the number of ramets) varying between 0.87 in some Belgian populations to 1 in German populations. These results indicate that despite its capacity for extensive clonal growth, sexual reproduction is the dominant mode of reproduction in this species under natural conditions.
The high variation in chromosome number in the genus Pulmonaria has been used to delineate species, disentangle the complex phylogeny of the genus and to identify putative hybrids (Tarnavschi 1935; Merxmüller & Grau 1969; Merxmüller & Sauer 1972; Sauer 1975). Pulmonaria officinalis has been reported to have 16 chromosomes (Merxmüller & Sauer 1972; Sauer 1975; Moore 1982; Dobeš, Hahn & Morawetz 1997; Vosa & Pitolesi 2004), whereas other authors have reported 14 chromosomes (Clapham, Tutin & Warburg 1952; Góralski, Lubczyńska & Joachimiak 2009). Zonneveld, Leitch & Bennett (2005) measured a 2C DNA amount of 3.2 pg for P. officinalis.
The species has been considered (hyper-)diploid (Tarnavschi 1935; Kirchner 2004). However, recent population genetic research on the species in Belgium and Germany using nuclear microsatellite markers indicated that P. officinalis is tetraploid (Molecular Ecology Resources Primer Development Consortium et al. 2011). Flow cytometric measurements of similar DNA amounts in 300 individuals (mean = 2.9 pg/2C) from two large Belgian populations suggested that all individuals in those P. officinalis populations had the same ploidy level (S. Meeus, unpubl. results). Because the microsatellite profiles showed no differentiation of allele sets and allele combinations were completely random (no fixed heterozygosity), it has also been suggested that P. officinalis is an auto- rather than an allotetraploid (Meeus et al. 2012a).
(E) Physiological data
Growth starts with the development of a flowering stem and small spring leaves before full expansion of the tree canopy and P. officinalis is therefore exposed to full sunlight during its reproductive phase, during which plants are more vulnerable to UV-B (Gaberščik et al. 2001). Variegation of leaves was initially thought to protect against photo-damage (Esteban et al. 2008). The light green blotches on the leaves of P. officinalis result from loosely arranged palisade parenchyma cells, in contrast to the tightly packed cells of the darker green areas. This provides the different leaf portions with different optical properties. Despite the higher reflectance of the blotches for all wavelengths, the light green leaf areas are not more photo-protected than the other leaf portions (Esteban et al. 2008). In both leaf areas, the same amount of UV-B absorbing compounds was stored under elevated UV-B radiation (Gaberščik et al. 2001).
The species uses the C3 metabolic pathway of carbon fixation. This is the common pathway for temperate plant species that typically occur in areas where sunlight flux density is moderate and ambient temperature intermediate (Fitter & Peat 1994). Bauer & Martha (1981) measured a CO2 compensation point (minimum CO2 concentration inside the leaves) of 30.8 ± 0.1 μL CO2 L−1 in P. officinalis under standardized conditions (temperature of 20 °C, ambient O2 concentration and light saturation). The change in leaf type from spring/stem leaves to summer leaves in combination with the gradual closure of the tree canopy induces a change in photosynthetic capacity and carbon assimilation of the plant with a peak in photosynthesis during spring (Schimper & Von Faber 1935; Larcher 2003).
(F) Biochemical data
Fischer et al. (2008) compared the chemical composition and nutritional quality of P. officinalis seeds and elaiosomes and found higher accumulations of easily digestible low molecular weight compounds (e.g. amino acids) in the elaiosomes compared with the accumulation of high molecular weight compounds (e.g. proteins) in the seeds (Table 1). In particular, the high oleic acid (C18:1n9c), trehalose, nitrogen-rich amino acid (glutamine, histidine, arginine) and proline concentrations provide P. officinalis’ elaiosomes with a high nutritional value for ants and their larvae (Table 1; Fischer et al. 2008).
Table 1. Biochemical composition of seeds compared with that of their elaiosomes for Pulmonaria officinalis (reproduced from Fischer et al. 2008). (a) Average protein and amino acid content (mg g−1 DM) and the percentage of amino acids in the total amino acid composition, (b) average lipid content (mg g−1 DM) and fatty acid composition (%) and (c) starch and soluble low molecular weight carbohydrates (mg g−1 DM) and carbohydrate composition (%) for seeds and elaiosomes, separately (only reported if constituted more than 5% of total amino acids/fatty acids/carbohydrates content of elaiosomes/seeds of at least one species of the 15 central European plants studied by Fischer et al. 2008)
The aerial parts of P. officinalis contain alkaloids, saponins, phenolic acids, tannins, flavonoids, vitamin C and γ-linolenic acid (Brantner & Kartnig 1994). Silica and calcium carbonate are also found in the cell walls of the bristles of Pulmonaria (Kirchner 2004). Because of the high water-soluble antioxidant capacity [2.02 ± 0.14 mM Trolox equivalent antioxidant capacity (TEAC)] and phenol content (673.39 ± 9.92 μM) of P. officinalis leaves, they are still applied in Bulgarian phytotherapy, most often as an antitussive and expectorant (Ivanova et al. 2005).
Pulmonaria officinalis is a (semi-)evergreen forest herb that flowers mainly from the end of March/beginning of April until May. As such, flowering starts before complete closure of the tree canopy (Schimper & Von Faber 1935). The rosette slowly starts to form during spring by the appearance of young summer leaves in the middle of the rosette and expands after the flowering season during summer.
Development of the flowering stem and flower buds may begin in February, and flowering starts at the end of February at the earliest. The terminal cymes contain four to eight open flowers and flowers last approximately for 6–8 days (Oberrath, Zanke & Böhning-Gaese 1995). From phenological observations since 1953, van Vliet (2008) found an average flowering period for P. officinalis of 80 days which was shortened by high mean temperatures in January. Schieber (2007) measured an average duration between first flowering (10% shoots flowering) and full flowering (50% shoots flowering) of 8 days. Pulmonaria officinalis shows high yearly variation in the onset of flowering (SD = 7.2 days), but there was no significant correlation between temperature and flowering date (Schieber 2007).
Seeds ripen from May until June and change colour from green to black, at the same time as the formation of a new rosette of shade-adapted summer leaves. Next-generation seedlings can reach maturity and start to flower in their first year of growth only if they germinated soon after seed production of the previous flowering season (September–October) (S. Meeus, pers. observ.).
VIII Floral and seed characters
(A) Floral biology
Pulmonaria is one of eight heterostylous genera within the Boraginaceae and contains only distylous species (Merxmüller & Sauer 1972; Olesen 1979). Two mating groups can be distinguished morphologically in distylous species according to the position of styles and stamens in the flower. Hildebrand (1865) reported that the pistil of pin-eyed flowers of P. officinalis (c. 10 mm) was about twice the length of the stamens (c. 5 mm), and a corresponding but reversed arrangement of styles (5–6 mm) and stamens (10–12 mm) was found in thrum-eyed flowers. The observations regarding the dimorphic nature of this species together with the results of a concise pollination experiment were included and evaluated in Darwin's work (1877) where he postulated that the reciprocal arrangement of stamens and styles in the floral morphs functions as a mechanical device to promote insect-mediated cross-pollination between opposite morphs. Recently, Brys et al. (2008a) confirmed distyly in P. officinalis by measuring anther and stigma heights and calculating reciprocity indices (R) cf. Richards & Koptur (1993). For both long (R =0.016) and short (R =0.003) levels, reciprocity indices were close to zero, indicating high reciprocity among the morphs (Fig. 6a; Brys et al. 2008a). However, despite the strong reciprocal placement of the stigmas and anthers, pollen transfer was found to be asymmetric in this species with more pollen grains deposited on long (pin) styles of which only 35% originated from thrums (Brys et al. 2008a) (Fig. 6b). Asymmetrical pollen deposition in Belgian P. officinalis populations has been attributed to the short proboscises of the principal pollinators and the occurrence of floral hairs at the corolla tube entrance of the pin corolla (Brys et al. 2008a).
Additional dimorphisms in corolla tube size, anther size, pollen grains and stigmatic papillae were found. Corolla tubes are, generally, deeper (pin: 11.9 mm, thrum: 11.3 mm) and wider (pin: 2.1 mm, thrum: 1.8 mm) in pins than in thrums, although the tube of thrum flowers is somewhat dilated at the level of the short stamens (Hildebrand 1865; Darwin 1877; Brys et al. 2008a). Short stamens of pin-eyed flowers tend to have larger anthers (pin: 1.9 mm, thrum: 1.8 mm) compared with the long stamens of thrum-eyed flowers and produce nearly twice as much pollen (pin: 26 810 pollen grains per flower) than the latter (thrum: 14 360 pollen grains per flower). These pollen grains are, however, smaller in size than those produced by the anthers of thrums (average length: 22.8 (±0.09) and 29.6 (±0.11) μm, respectively). Although the crown area of the stigmatic papillae of pins (42.09 ± 0.50 μm2) is generally smaller than that of thrum plants (98.03 ± 1.10 μm2), the stigmatic papillae themselves are longer and the interpapillar distances larger in pins than in thrums (Brys et al. 2008a). The number of flowers per plant varies from 11 to 105, but there appears to be no significant difference in flower production between pin and thrum plants (Brys et al. 2008a).
During anthesis, flowers of Pulmonaria change the colour of their corolla from red to blue. Oberrath, Zanke & Böhning-Gaese (1995) reported a gradual decrease in nectar content during colour transition and a clear preference of pollinators for young, red flowers (Müller 1883; Brys et al. 2008a). This colour change is not triggered by insect visits, pollination or fertilization, but is solely due to the ageing of the flowers (Süssenguth 1936; Oberrath, Zanke & Böhning-Gaese 1995). The retention of old blue flowers intends to enlarge floral display for attracting insects from long distances (Oberrath & Böhning-Gaese 1999). Once pollinators reach the plant, the possession of a large floral display is disadvantageous because of the risk of geitonogamous pollination. Colour change poses a solution for this dilemma by directing approaching pollinators only to the reproductive and rewarding red flowers, focusing insect visits per plant (Oberrath & Böhning-Gaese 1999).
Pulmonaria officinalis flowers produce copious nectar (c. 1.5 μL nectar per flower, but this is highly variable) which mainly attracts Hymenopteran pollinators (Table 2) with an early flight period such as Bombus queens that have survived the winter and the solitary bee species Anthophora plumipes (March–May) (Knuth 1909). Anthophora plumipes is widespread in Europe and common (similarly to P. officinalis) in central to southern England and Wales, but it is completely absent from Ireland and Scotland (NBN gateway 2013). From the multiple recordings in the literature (Table 2) and the efficiency with which A. plumipes transfers pollen between P. officinalis flowers as opposed to bumblebees (Brys et al. 2008a), this species is considered as the principal pollinator of P. officinalis. However, personal observations in Belgium and Germany indicate that although A. plumipes is observed more frequently in German populations compared with Belgian ones, bumblebees account for the majority of all visits. Unlike the bumblebee species (Bombus terrestris, B. pratorum, B. pascuorum), however, A. plumipes succeeds in transporting thrum and pin pollen segregated on its body due to its long proboscis (14.3 mm) (Fig. 6b; Brys et al. 2008a).
Table 2. Pollinators recorded for Pulmonaria officinalis
In two Belgian populations, total stigmatic pollen loads were highly variable, ranging from 10 to 332 pollen grains per stigma (mean = 152; median = 148) (Brys et al. 2008a). Stigmas of pin flowers received significantly more pollen grains than those of thrum flowers (on average 173.8 ± 12.5 and 130.6 ± 8.6, respectively), whereas the proportional number of legitimate deposited pollen (= pollen of the opposite morph) was significantly larger on thrum stigmas than on pin stigmas. Stigmas of thrum flowers received on average 58.1% legitimate pollen, while the total stigmatic pollen load of stigmas of pins consisted of only 35.1% legitimate thrum pollen. Hence, on a per pollen basis, pollen grains of thrum plants had a proficiency of legitimate pollen transfer that is more than one and a half times as large as the proficiency of legitimate pollen transfer of pin pollen. The probability of thrum pollen participating in legitimate pollen transfer is also larger than that of participating in illegitimate mating. The difference in flight pattern and proboscis length between bumblebees and the more efficient pollinator A. plumipes (Fig. 6b), which is less frequently observed in Belgian P. officinalis populations, might account for asymmetrical pollen flow between the morphs in natural populations, but also for the stronger spatial genetic structure (using the ‘Sp’ statistic as a measure for quantifying spatial genetic structure, Vekemans & Hardy 2004) within these Belgian populations (Sp = 0.0053) compared with core populations (Sp = 0.0030) in Germany (S. Meeus, O. Honnay, H. Jacquemyn, unpubl. data). Because bumblebees tend to travel smaller distances and forage longer on the same plant, gene dispersal is spatially more restricted in Belgian populations compared with German populations where A. plumipes is observed more frequently (Oberrath, Zanke & Böhning-Gaese 1995; Oberrath & Böhning-Gaese 1999; Brys et al. 2008a).
Bees and bumblebees are rewarded with copious amounts of nectar, which is secreted at the bottom of the corolla tube. Although floral nectar is believed to be initially sterile, recent investigations have shown that the nectar of P. officinalis is often inhabited by micro-organisms, most often yeasts and bacteria (Jacquemyn et al. 2013). Detailed investigations of microbial communities in the floral nectar of ten P. officinalis populations revealed several ascomycote and basidiomycote yeasts, including Metschnikowia reukaufii, Candida bombi, Sporobolomyces roseus and several Cryptococcus species. Additionally, a large number of bacteria was recovered, which belonged to three major phyla (Actinobacteria, Firmicutes and Proteobacteria) (Jacquemyn et al. 2013). The most common bacteria were species from the genera Rhodococcus, Microbacterium and Methylobacterium. Other species that were identified (> 97.5% sequence homology with GenBank sequence) included members from the genera Arthrobacter, Bacillus, Brachybacterium, Brevibacterium, Devosia, Erwinia, Enhydrobacter, Flexivirga, Gordonia, Janibacter, Luteipulveratus, Micrococcus, Moraxella, Nocardioides, Okibacterium, Plantibacter, Ponticoccus, Pseudomonas, Rhodanobacter, Saxeibacter, Staphylococcus and Streptomyces (Jacquemyn et al. 2013). Although two recent studies have shown that the presence of either bacteria or yeasts can have a profound impact on nectar chemistry and on pollination success and seed set (Herrera, Pozo & Medrano 2013; Vannette, Gauthier & Fukami 2013), the significance of the observed microbes for P. officinalis and its pollination needs to be investigated in more detail.
Although morphological intermediates have been frequently observed at locations containing two Pulmonaria species, the extent to which Pulmonaria species hybridize is still under debate. Given the close resemblance of the species, some authors have suggested that hybridization in Pulmonaria should be common and widespread (Fig. 1; Gams 1927), whereas others have stated that hybridization is unlikely due to unequal chromosome numbers of Pulmonaria species resulting in sterile offspring if the four tertiary species were to cross freely (Merxmüller & Sauer 1972). Hegi (1927) reported detailed locations of several P. officinalis hybrids in Central Europe that originated from crosses with the other three tertiary species. The hybrid P. angustifolia × P. officinalis = P. hybrida has been found, for example, in Lower Austria and southern Tirol. Another hybrid, P. montana × P. officinalis = P. digenea, has been observed in Bierbaum (Steiermark) and in Hungary (Fig. 1). However, because detailed morphological and molecular data underpinning hybridization were lacking, the existence of these hybrids still needs to be verified.
In Britain, on the other hand, Darwin (1877) reported failure of seed set after pollinating Pulmonaria longifolia (P. angustifolia) from the Isle of Wight legitimately and illegitimately with pollen from P. officinalis. Sauer (1975), however, reported the incidence of a hybrid from both species in Farnham, Surrey, but without further details. Although the species co-occur in Dorset, New Forest and the Isle of Wight, and show an overlap in their flowering period, they have different ecological requirements (Hill, Preston & Roy 2004) which could pose a barrier for gene exchange (‘pre-mating barrier’) and, thus, hybridization. P. longifolia only tolerates slightly shaded habitats (L =6), whereas P. officinalis can be found in more heavily shaded places. Pulmonaria longifolia also prefers drier places (M =4), less basic soils (R =6) and less nitrogen in the soil (N =5) than P. officinalis.
(C) Seed production and dispersal
Like most other distylous species, P. officinalis is characterized by a heteromorphic incompatibility system and theoretically only pollination between pins and thrums results in seed set (Barrett 1990). Hildebrand (1865) was the first to study seed production in P. officinalis after experimental self-pollination, intramorph pollination (between plants belonging to the same morph) and intermorph pollination (between plants belonging to the opposite morph). In Hildebrand's experiment, seed set was only observed when pollen derived from the opposite morph type (‘legitimate pollen’) was used for pollination. Darwin (1877), however, noticed that some English pin plants of P. officinalis were able to set seed regardless of the presence of surrounding thrum plants and that these plants were even capable of self-pollination. Brys et al. (2008a) found similar results in Belgium regarding the higher compatibility of pin plants of P. officinalis. In thrum plants, self- and intramorph pollination resulted in, respectively, only 5.2% and 11.5% of the pollen tubes entering the ovary, and only 1.6% (0.06 seeds per fruit) and 8.6% (0.34 seeds per fruit) of the ovules setting seed. In contrast, 41.9 and 48.3% of the pollen tubes reached the ovary of pin flowers, and on average, 10.3% (0.41 seeds per fruit) and 23.0% (0.92 seeds per fruit) of the ovules set seed after self- and intramorph pollinations, respectively. Intermorph pollination resulted in significantly lower pollen tube growth and less successful fertilization of ovules in thrum flowers compared with pin flowers, as on average 68.1% vs. 84.0% of the pollen tubes entered the ovary and on average 43.6% (1.74 seeds per fruit) vs. 50.4% (2.01 seeds per fruit) of the ovules set seed in thrum and pin flowers, respectively. Pollen tubes typically showed a callose expansion at the top of the tube when they were arrested in the stylar tissue, regardless of the pollination treatment applied or the receiving morph.
Seed production in natural P. officinalis populations is determined by a complex interplay of morph type, morph ratio, population density and spatial distribution of the morphs in the population, floral display size and the quality of pollen (self- vs cross-pollen) deposited on the stigma (Brys, Jacquemyn & Beeckman 2008b; Brys et al. 2008a; Brys & Jacquemyn 2010). Investigating seed set in 35 natural populations in Belgium, Brys, Jacquemyn & Beeckman (2008b) showed that average seed production per flower increased significantly with increasing population size and was also significantly larger in pin plants than in thrum plants (on average 2.38 ± 0.06 and 1.44 ± 0.07, respectively). Mean seed set per flower in thrum plants significantly declined with decreasing population size, whereas seed set per flower did not depend significantly on population size in the pin morph. Biased morph ratios, however, induce shifts in reproductive success between the morphs due to negative frequency-dependent fertility. Brys, Jacquemyn & Beeckman (2008b) reported equal morph fertility at a frequency of 66% pins in the population, and Meeus et al. (2012a) found the highest within-population genetic diversity in slightly pin-biased populations (61% pins). Because of the incompatibility system's selectivity, it is essential for both morphs to occur in the population in order to maintain large seed set. Reproduction, however, is further affected by the spatial distribution of the morphs in the population because the seed set of an individual plant decreases as the distance to the nearest compatible mate, a flowering individual of the opposite morph increases (Brys & Jacquemyn 2010). However, reproduction of thrums is more affected by increasing distance because of the low compatibility with pollen of the same morph type. Seed set in pins, on the other hand, is more affected by floral display size as the pin-eyed morph benefits from geitonogamous pollination through its partial self-compatibility (Brys & Jacquemyn 2010).
Seed dispersal in P. officinalis is mediated by ants (myrmecochory) (Hermy et al. 1999; Lengyel et al. 2010). The diaspores of P. officinalis dispersed by ants are the nutlets with an attached elaiosome derived from the lignified pericarp base (fruit tissue) of the nutlets (Fig. 7a; Forget et al. 2005).
(D) Seed germination and viability
In natural conditions, few P. officinalis seeds are able to germinate immediately after seed dehiscence in summer, whereas the majority of seeds overwinter and germinate between February and April (F. Vandelook, unpubl. results; Fig. 8). These observations suggest that P. officinalis seeds need cold stratification to break dormancy before germination can begin, which is common for shade-tolerant herbs of temperate deciduous forests (Bierzychudek 1982). A cold stratification of 5 °C in the laboratory of 2 months appeared to be less effective in breaking dormancy with a germination success of on average 4.7% compared with 5 months of stratification which yielded 46% of germinated seeds (F. Vandelook, unpubl. results). A pre-treatment of scarification through cutting into the pericarp enhances germination success considerably because about 97% of seeds germinated at an alternating temperature of 12 h at 15 °C and 12 h at 6 °C without any cold stratification (S. Meeus, pers. observ. 2011). The removal of the elaiosome during ant dispersal may provide scarification under natural conditions (Gorb & Gorb 2003).
(E) Seedling morphology
Stages in the development of seedlings in natural conditions are shown in Fig. 7. Germination starts by the partial rupture of the pericarp at the top of the nutlet (opposite to the elaiosome) with the emergence of the radicle (Fig. 7a,b). Germination is epigeal. During the elongation of the primary root axis of the plant, the pericarp and testa are shed by the expanding and separating, green, ovate cotyledons (Fig. 7c). After approximately 1 week from the separation of the cotyledons, the first real leaf emerges from between the cotyledons (Fig. 7d,e).
IX Herbivory and disease
(A) Animal feeders or parasites
Leaf miners, such as Agromyza abiens (Zetterstedt), are frequently (c. 10% of all plants affected; H. Jacquemyn, pers. obs.) observed to feed on P. officinalis leaves leaving brown necrotic spots or a cluster of narrow corridors (Fig. 9; Spencer 1972). The larvae of Dialectica imperialella (Zeller), a moth (Lepidoptera, Gracillariidae) is another common miner of P. officinalis leaves whose mines eventually form brownish blotches (Weeda et al. 1988; Fitter & Peat 1994). Bumblebees sometimes become nectar robbers when they feed on P. officinalis nectar from the outside by perforating the corolla tube without transferring the pollen (Frankum 1999; S. Meeus, pers. observ.).
(B) Plant diseases
The powdery mildew Golovinomyces (‘Erysiphe’) cynoglossi (Wallr.) V. P. Heluta is an obligate ascomycote parasite of the Erysiphaceae that infects the leaves, stems, flowers and even the fruits of P. officinalis. Both the anamorph and the teleomorph infect hosts exclusively from the Boraginaceae, such as Forget-me-nots (Myosotis spp.), Houndstongue (Cynoglossum spp.) and Lungworts (Pulmonaria spp.) (Ellis & Ellis 1997; Glawe 2008). De Clerck-Floate (1999) assessed the effect of this fungus on growth and reproduction of Cynoglossum officinale, an introduced rangeland weed in North America, and found that infection led to a reduction in percentage seed set of 20%, together with reductions in nutlet number, nutlet size and seed germination. Ramularia cylindroides Sacc. is an anamorph hyphomycete fungus that parasitizes on the leaves of several Pulmonaria, Symphytum and Alkanna spp. (Braun 1998).
Pulmonaria officinalis is a neophyte in the British Isles. The species was cultivated in Britain before 1597 and first recorded as a rare occurrence from the wild in 1793 (Grinling, Ingram & Polkinghorne 1909; Preston, Pearman & Dines 2002; Hill, Preston & Roy 2004). Because of its ornamental and medicinal properties, the species became cultivated in Western European countries, especially during medieval times (Preston, Pearman & Dines 2002). In the sixteenth century, the spotted leaves were believed to resemble diseased lungs and were therefore in accordance with the ‘Doctrine of signatures’ traditionally used as a treatment for lungs and respiratory diseases (Scott 2008). Ornamental or medicinal use of P. officinalis in Western Europe during the Middle Ages appears, from its occurrence in ancient forests whose names refer to them, to be associated with the former presence of a monastery or castle (Tack, Van den Brempt & Hermy 1993). Although the exact history of this species in Western European countries such as the Netherlands and Belgium is not known, some authors (e.g. Crépin 1863a,b) have stated that the species is native, whereas others believe (e.g. Tack, Van den Brempt & Hermy 1993) that the species was introduced during the Middle Ages. For centuries, plants from the genus Pulmonaria have also been used in gardens. Pure Pulmonaria species have often been crossed to obtain ornamental cultivars such as Pulmonaria ‘Beth's Blue’ and ‘Monksilver’, which are presumed to be hybrids between P. longifolia and P. affinis (Bennett 2003).
From its first record from the wild in 1793, in which P. officinalis was found to be very rare and naturalized in copses between Knockholt, Cudham and Downe (West Kent; SE outskirts of Greater London) (Grinling, Ingram & Polkinghorne 1909), the abundance of P. officinalis has increased during the last two centuries throughout Britain, especially in England (Fig. 10). Although the increase partly reflects more intensive recording, a Change Index of +1.77 indicates a relative range expansion between the period 1930–1960 and 1987–1999 independent of survey effort (Preston, Pearman & Dines 2002; Telfer, Preston & Rothery 2002; Hill, Preston & Roy 2004). Since the year 2000, P. officinalis has been found in 42 additional 10-km grid squares in Britain (Fig. 10).
In contrast to the Suffolk lungwort (P. obscura), P. officinalis is not listed in the UK as a vulnerable, rare, priority species but as an alien (neophyte) species that is increasing in total cover (Fig. 10) (Hitchmough & Woudstra 1999; Preston, Pearman & Dines 2002; Hill, Preston & Roy 2004; Dines et al. 2005). Pulmonaria officinalis is also not included in the European red list of vascular plants (Bilz et al. 2011), although the species is reported to be rare and/or declining in a few European countries such as Belgium and Sweden, where it is located at the margin of its distribution range. In southern Sweden, Falkengren-Grerup (1986) found that P. officinalis had disappeared from a majority of sites in which it had been found 15–35 years previously and that the species decreased proportionally more in the most acid soils, whereas it colonized neutral soils that became more alkaline.
In northern Belgium, the species is classified as rare although Van Landuyt et al. (2006) found a Change Index of +0.52, which indicates a slight increase in the range of P. officinalis between the survey periods 1939–1971 and 1972–2004. Meeus, Honnay & Jacquemyn (2012b) compared genetic diversity and differentiation between highly fragmented (Belgium) and more connected populations (Germany) of P. officinalis. They found less genetic variation and higher genetic differentiation between the fragmented populations than between highly connected populations, indicating that habitat fragmentation can have a major impact on the genetic diversity of P. officinalis populations. Gene exchange between P. officinalis populations in fragmented landscapes appeared to be limited and contributed to the reduced genetic diversity (Meeus, Honnay & Jacquemyn 2012b). Gene flow appeared to occur primarily between adjacent populations, leading to high genetic differentiation in fragmented landscapes and a significant pattern of isolation by distance, whereas in strongly connected landscapes, no isolation by distance was observed. Additionally, Jacquemyn et al. (2013) showed that fragmented populations of P. officinalis contained specific communities of nectar-inhabiting micro-organisms that showed low similarity, indicating that foraging by pollinators between populations is limited. These results show that P. officinalis may be susceptible to the deleterious effects of habitat fragmentation and that the long-term conservation of this species calls for specific actions that maintain population sizes, for example by regular opening of the canopy and increasing gene flow between populations.
This research was funded by the Flemish Fund of Scientific Research (F.W.O.) (project G.0500.10). We are grateful to Keith Kirby for providing useful information on plant communities. We also thank Filip Vandelook for sharing his experience on the germination of Pulmonaria officinalis seeds.