Boraginaceae
Description
Trees, shrubs, subshrubs, woody climbers, perennial or annual herbs usually covered by hairs or bristles on the herbaceous parts, woody species sometimes entirely glabrous.
Hairs are among the most generally distributed features in the family. Unicellular hairs are always present, but they are not the only type of hairs in many species. Typical for unicellular hairs are agglomerations of calciumcarbonate or cystoliths in their lumen. To maintain their flexibility, there are various devices, most often a specific distribution of thin and thickened segments of the cell wall. In many cases, most characteristically in members of tribe Lithospermeae, the hairs are surrounded by one to several circles of so called 'accessory cells' with calcified cell walls, that may also be different in the ratio of length to width from other epidermal cells. Widespread are bristle-like hairs with one longitudinal series of cells the terminal cell of which may be horizontal or oblique. These are glochids mainly found on fruits in order to attach them to the fur of animals. Other types of hairs less common are capitate hairs which are characterized by undivided heads in Boraginaceae, stellate hairs and hairs composed of more than one series of cells.
Leaves alternate, very rarely opposite (in Tournefortia), exstipulate, undivided, usually entire, in a very few species serrate, with reticulate venation of which only the main nerves are clearly detectable in most cases.
Inflorescence either a simple cyme or compound, cymes arranged dichotomously or in racemes or panicles, with or without bracts, terminal or lateral, sometimes single flowers in the axils of upper leaves.
Flowers hermaphroditic, rarely unisexual and plants monoecious, composed of calyx and corolla, pentamerous, rarely tetramerous, actinomorphous, in some genera slightly zygomorphous.
Stamens of the same number as corolla lobes, alternating with them, included in or exserted from the corolla; anthers sessile or on distinct filaments, sometimes with a sterile tip of connective tissue or produced to an awn-like structure, awns of one flower twisted or not, anthers bilocular, dorsally fixed to the corolla or the filament, opening longitudinally.
Ovary superior, often fixed either to a disc-like base or base produced to a conical or pyramidal structure (gynobasis), bicarpellate as a rule, undivided, entire or with four deep furrows, breaking into two biovulate halves or into four uni-ovulate mericarps ('nutlets'), sometimes part of the locules aborted, rarely a greater number of carpels developed (e.g., in Trigonotis procumbens); placentation axillary; ovules erect or nearly horizontal, rarely pendent.
Fruit drupaceous or a berry or divided into 2-4 mericarps surrounded by a hard outer wall, sometimes adherent to the style and separating with an awn; surface smooth, warty, or rugulose, often tuberculate or spiny, sometimes divided into a disc-like outer and a convex inner side, the outer disc surrounded by a thickened margin or not, sometimes glochids present on the whole or certain parts of the surface or nutlets winged.
Seeds usually without albumen.
Distribution
Asia-Tropical
About 100 to 115 genera with a total of about 2,400 species in all climatic regions of the earth. The woody genera of subfamilies Cordioideae, Ehretioideae and Heliotropioideae are predominant in tropical and subtropical regions, Heliotropium is widespread in tropical and temperate regions, while herbs and subshrubs belonging to Boraginoideae are most numerous in temperate regions, in the tropics mainly confined to higher altitudes or growing as weeds and also represented in the arctic with a few species (involves only Boraginoideae).
In the Malesian region, 12 genera are indigenous, two more introduced. Of these, Crucicaryum is of rather doubtful value and only known from one collection, which has been lost. Pteleocarpa, sometimes regarded as a member of Boraginaceae, is not included in this family by the present author. The number of species acknowledged in this account, is 77, of which 64 or 65 are indigenous, the other 12 or 13 having been introduced either for economical and/or ornamental uses or as weeds.
In the Malesian region, 12 genera are indigenous, two more introduced. Of these, Crucicaryum is of rather doubtful value and only known from one collection, which has been lost. Pteleocarpa, sometimes regarded as a member of Boraginaceae, is not included in this family by the present author. The number of species acknowledged in this account, is 77, of which 64 or 65 are indigenous, the other 12 or 13 having been introduced either for economical and/or ornamental uses or as weeds.
Ecology
Pollination is performed by insects of various groups in a great majority of cases though special adaptations to particular pollinators are lacking. Faucal appendages prevent insects with a short proboscis from pollinating the flowers in many genera, however. These appendages serve as guide to the nectar in the basal part of the corolla. Selfing is observed in a number of taxa, but is prevented by heterostyly in others. Heterostyly is combined with self-incompatibility as a rule, in some genera also with pollen dimorphism. In large-flowered Cordia species bats may act as pollinators.
There are several ways of dispersal found within the family. Nutlets bearing glochids are well adapted to epizoochorous dispersal. Comparatively small, hard nutlets are eaten by birds and transported in their digestive tract over longer or shorter distances, while the fruits of Cordia are partly digested by mammals and the endocarp is excreted in a viable state. Species of the sea coast such as Tournefortia argentea with a spongy pericarp are dispersed by seawater. Winged nutlets are transported by the wind. The same is true sometimes in plants with single surviving nutlets that are shed together with the enlarged calyx.
There are several ways of dispersal found within the family. Nutlets bearing glochids are well adapted to epizoochorous dispersal. Comparatively small, hard nutlets are eaten by birds and transported in their digestive tract over longer or shorter distances, while the fruits of Cordia are partly digested by mammals and the endocarp is excreted in a viable state. Species of the sea coast such as Tournefortia argentea with a spongy pericarp are dispersed by seawater. Winged nutlets are transported by the wind. The same is true sometimes in plants with single surviving nutlets that are shed together with the enlarged calyx.
Taxonomy
Delimitation of Boraginaceae as a whole is still controversial. While a majority of authors regard them as a natural entity that can further be subdivided into five subfamilies, I.c., Wellstedioideae, Cordioideae, Ehretioideae, Heliotropioideae and Boraginoideae, others take some of these subfamilies as separate families which are even assigned to different orders by Hutchinson and his followers. As an example, Takhtajan (1987) may be cited, who excludes the primarily woody Cordioideae and Ehretioideae as well as the tetrame-rous Wellstedioideae from Boraginaceae in the strict sense, which are divided into Heliotropioideae and Boraginoideae. On the other hand, embryologists advocate recognition of a distinct family Heliotropiaceae including Cordioideae and Ehretioideae. There is little doubt that all these groups are closely related to each other. In the present author's opinion, there are no compelling reasons for dividing Boraginaceae into several separate families, and they are treated as an entity here.
Boraginoideae are divided into a number of tribes, some of which have been adopted in all newer treatments of the subfamily. These are Lithospermeae, Boragineae (= An-chuseae of older literature), Eritrichieae and Cynoglosseae. To include some genera ancestral to these tribes and probably closely related among each other, Riedl (1968) adopted an additional tribe originally proposed by Popov (1953) under the name Trigonotideae. Myosotis is different from all other genera of Boraginaceae by its contorted aestivation and is best regarded as the only member of tribe Myosotideae.
Synopsis of subfamilies, tribes and genera of Boraginaceae represented in Malesia:
Pteleocarpa is not considered as a member of Boraginaceae. It is treated here as a genus of uncertain position at the end of the family.
While there is little doubt that Boraginaceae are most closely related to Hydrophyllaceae and Lennoaceae, their assignment to a particular order is still controversial. Hutchinson (1973) included the woody Ehretiaceae in Verbenales ancestral to Verbenaceae, while Boraginaceae s.str. were considered as the only family of a monotypic order Boraginales derived from Polemoniales.
As has been argued, the woody groups are certainly closely related to the herbaceous ones and should not be placed into a different order. Cronquist (1981) and Thorne (1976) united them with Labiatae and Verbenaceae in the order Lamiales. The alternate leaves, actinomorphic corolla and number of stamens equal to that of corolla lobes along with the presence of pyrrolizidine and not-iridioid alkaloids are strong arguments against this decision as has been pointed out by Dahlgren (1977,1983) and Thorne (1983), who included the family in Solanales, among others. The soundest solution seems to keep Boraginales with the families mentioned above as a separate order as proposed by Takhtajan (1980, 1987) and Dahlgren. It is also justified for embryological and anatomical reasons according to Al-Shehbaz (1991). Boraginales probably are derived from Polemoniales in close vicinity of which they are also placed by Takhtajan (1987).
Boraginoideae are divided into a number of tribes, some of which have been adopted in all newer treatments of the subfamily. These are Lithospermeae, Boragineae (= An-chuseae of older literature), Eritrichieae and Cynoglosseae. To include some genera ancestral to these tribes and probably closely related among each other, Riedl (1968) adopted an additional tribe originally proposed by Popov (1953) under the name Trigonotideae. Myosotis is different from all other genera of Boraginaceae by its contorted aestivation and is best regarded as the only member of tribe Myosotideae.
Synopsis of subfamilies, tribes and genera of Boraginaceae represented in Malesia:
- Subfamily Cordioideae
- Trees or shrubs. Fruit a drupe with one quadrilocular or four unilocular pyrenes. Style terminal, twice forked.
- In Malesia: Cordia.
- Subfamily Ehretioideae
- Trees, shrubs, climbers or rarely herbs. Fruit a drupe with two bilocular or rarely four unilocular pyrenes. Style terminal, simply forked or two styles.
- In Malesia: Carmona, Coldenia, Ehretia, Rotula.
- Subfamily Heliotropioideae
- Trees, shrubs, subshrubs, climbers, perennial or annual herbs. Fruit with a fleshy, rarely a spongy pericarp and two bilocular or four unilocular pyrenes or hard, usually breaking up into four unilocular nutlets, rarely into two mericarps or not at all. Style terminal, surrounded by the annuliform stigma, entire or forked.
- In Malesia: Heliotropium, Tournefortia (including Argusia).
- Subfamily Boraginoideae
- Annual or perennial herbs, rarely subshrubs. Fruit breaking up into four unilocular nutlets, rarely nutlets 1-3 by abortion. Style basifixed between the nutlets and coherent to gynoecial tissue that is covering the receptacle and separating from the inner layers of the fruit wall, sometimes on top of a pyramidate to subulate gynobasis formed by that gynoecial tissue, undivided or once to twice forked.
- Tribus Trigonotideae
- Gynobasis flat, slightly concave or slightly convex in its centre. Ventral (inner) side of the nutlets angular, with an open furrow upwards from the subbasal areola at the angle, or with a simple keel or angle and stipitate areola, or umbilicate ventrally, umbilicus with a thickened margin.
- In Malesia: Bothriospermum, Trigonotis.
- Tribus Boragineae
- Gynobasis flat or nearly flat. Nutlets with a basal areola often surrounded by a thickened margin.
- In Malesia: Borago.
- Tribus Cynoglosseae
- Gynobasis conical to pyramidate. Nutlets attached for their whole length or only slightly surpassing the gynobasis, free from an early stage.
- In Malesia: Trichodesma, Cynoglossum, Omphalodes (?Crucicaryum).
- Tribus Myosotideae
- Gynobasis flat, small. Nutlets smooth with a minute subbasal, oblique areola. Corolla-lobes contorted in aestivation.
- Only Myosotis.
Pteleocarpa is not considered as a member of Boraginaceae. It is treated here as a genus of uncertain position at the end of the family.
While there is little doubt that Boraginaceae are most closely related to Hydrophyllaceae and Lennoaceae, their assignment to a particular order is still controversial. Hutchinson (1973) included the woody Ehretiaceae in Verbenales ancestral to Verbenaceae, while Boraginaceae s.str. were considered as the only family of a monotypic order Boraginales derived from Polemoniales.
As has been argued, the woody groups are certainly closely related to the herbaceous ones and should not be placed into a different order. Cronquist (1981) and Thorne (1976) united them with Labiatae and Verbenaceae in the order Lamiales. The alternate leaves, actinomorphic corolla and number of stamens equal to that of corolla lobes along with the presence of pyrrolizidine and not-iridioid alkaloids are strong arguments against this decision as has been pointed out by Dahlgren (1977,1983) and Thorne (1983), who included the family in Solanales, among others. The soundest solution seems to keep Boraginales with the families mentioned above as a separate order as proposed by Takhtajan (1980, 1987) and Dahlgren. It is also justified for embryological and anatomical reasons according to Al-Shehbaz (1991). Boraginales probably are derived from Polemoniales in close vicinity of which they are also placed by Takhtajan (1987).
Cytology
Polyploidy is confined to a certain limited number of genera within the family, while aneuploidy seems to be of great importance. In Cordioideae, only diploids are known with n = 8, 14, 16. In Ehretioideae, the basic number x may be 7-11, 13 and 21. In Heliotropioideae and Boraginoideae together, all basic numbers are known from x = 4 to 13. While it is often argued that the base number of the family as a whole is 12 because of its wide distribution within Boraginoideae, this is certainly not the case for the other subfamilies, in which x = 8 is much more likely the primary base number. There is no clear correlation between aneuploidy and systematic position within subfamily Boraginoideae, though x = 12 is most common in Cynoglosseae and probably also in Eritrichieae, while it is much less common in Lithospermeae. The lowest diploid number known in the family is 2n = 8, while the highest is 2n = 144.
Uses
Alkannin has been used as a dye in widely distant parts of the world. Species of Cordia provide edible fruits and timber. For the latter, South American species have been introduced into culture in Indonesia such as C. alliodora and C. gerascanthus. Recently, alkannin esters of various organic acids have been introduced into pharmacology for their wound healing properties. Borago officinalis is used as a spice.
Phytochemo
Pyrrolizidine alkaloids and quinonoid or phenolic compounds biogenetically derived from C-prenylated or -geranylated or -farnesylated 4-hydroxybenzoic acid are the most striking secondary metabolites of this family. The isomeric red pigments alkannin and shikonin are the best known representatives of such hydroxybenzoic acid derivatives. Synthesis and accumulation of pyrrolizidines and of alkannin-related compounds occur in all four subfamilies commonly accepted within the Boraginaceae', they form a biochemical link between Cordioideae, Ehretioideae, Heliotropioideae and Boraginoideae Boraginoideae and Heliotropioideae were most thoroughly studied by plant physiologists and phytochemists. This resulted in the detection of several metabolic features which should also interest plant taxonomists. In this respect phenolics, triterpenes and triterpene-based saponins, triglycerides of seeds, carbohydrate storage products and some N-containing compounds other than pyrrolizidines have to be mentioned. Phyto-chemical reviews of the family are available in Hegnauer (1964, 1989). In the following summary references are usually only given if not available in these two chemotaxonomic treatments.
1. R. Hegnauer
Embryology
Embryology has played a major part in the discussion whether Boraginaceae should be divided into at least two separate families or not. In subfamily Boraginaceae the endosperm may be nuclear, cellular or of an intermediate type, in the remaining subfamilies (as far as is known) it is always cellular. In Boraginoideae, an endosperm haustorium is lacking, while it is present in the remaining subfamilies. The suspensor is short in Boraginoideae, long in the rest. As already mentioned, pollen characters are not necessarily correlated with embryology, as Trichodesma likewise has two nuclei instead of three as in most of the other Boraginoideae. Mainly for embryological reasons, Heliotropiaceae Schrader as a separate family are advocated by several authors. Position of the embryo is correlated to attachment of nutlets and not always easy to interpret, therefore.
Citation
Brand 1921: pp. 1-183. – In: Engl., Pflanzenr., fam. IV.252
Van Royen 1975: pp. 79-98. – In: Pac. Sc.
C.B. Rob. 1909: pp. 687-698. – In: Philipp. J. Sc., Bot.
Brand 1931: pp. 1-236. – In: Engl., Pflanzenr., fam. IV.252
I.M.Johnston 1924: pp. 42-73. – In: Contr. Gray Herb.
Heine 1976: pp. 95-118. – In: Fl. Nouv.-Caléd
Ridl. 1923: pp. 438-442. – In: Fl. Malay Penins.