When treated as a subfamily, the name Scilloideae is derived from the
generic name of the
type genus,
Scilla, and is attributed to
Gilbert Thomas Burnett in 1835. When treated as a family, the name Hyacinthaceae is derived from the type genus
Hyacinthus, and is usually attributed to
August Batsch from ("ex") a 1797 publication by
Moritz Borkhausen.
Phylogeny '' The
monophyly of Scilloideae is well supported by studies based on molecular data. These studies also give support to the exclusion of
Camassia,
Chlorogalum and related genera, i.e. the former Hyacinthaceae subfamily Chlorogaloideae, now placed in the subfamily
Agavoideae. The exact position of the Scilloideae within the broadly defined Asparagaceae is less clear. One possible phylogeny for the seven subfamilies recognised within the family is shown below. }} Although generally agreeing on the main division of the Asparagaceae into two clades, studies have produced slightly different relationships among the Agavoideae, Aphyllanthoideae, Brodiaeoideae and Scilloideae. For example, Seberg et al. (2012) present analyses based on parsimony and on maximum likelihood. In the first, the Scilloideae are sister to the Agavoideae; in the second, they are sister to the Brodiaeoideae.
Early classifications Detailed historical accounts of taxonomic issues relating to the modern subfamily Scilloideae have been provided by Pfosser & Speta (1999) and Chase et al. (2009). The
lilioid monocots have long created classification problems. At one extreme, e.g. in the
Cronquist system of 1968, they have been regarded as one large family (
Liliaceae sensu lato). At the other extreme, e.g. in the
Dahlgren system of 1985, they have been divided between orders and split into many often small families. Dahlgren divided the lilioid monocots in search of
monophyly, but in practice he was unsuccessful. His major contribution was to split the Liliaceae into two families, the true Liliaceae, Liliaceae
sensu stricto, and the Hyacinthaceae (families which are now placed in separate orders,
Liliales and
Asparagales). Splitting off the Hyacinthaceae from the Liliaceae was originally suggested by
Batsch in 1786. Batsch's version of the family only superficially resembles the modern version, but did include
Hyacinthus and
Lachenalia. The group was reduced to a tribe by
Endlicher in 1836, and included
Camassia. In 1866
Salisbury redistributed the genera into several families. In the 1870s,
Baker used tribes to divide up the Liliaceae
s.l.. introducing the Hyacintheae, Scilleae, Massonieae, and Chlorogaleae. In 1887
Engler divided the Liliaceae
s.l. into two tribes, Lilieaoe and Scilleae. In the twentieth century,
Fritsch proposed the division of Liliaceae
s.l. into smaller more homogeneous families. In the 1930s the Viennese school elevated Engler's tribes to subfamilies. They questioned the inclusion of such different groups as Lilioideae and Scilloideae within the same family, and even Scilloideae was considered to be composed of at least three groups. By 1969,
Huber was recognizing the Scilloideae as the family Hyacinthaceae, and dividing it into tribes. How many tribes were recognised and how the genera were distributed within those tribes depended on the diagnostic characters chosen. Huber used seeds, while Schulze in 1980 used pollen. Morphology and chromosome analysis were supplemented by chemotaxonomy, due to the presence of cardiac steroids, such as the bufadienolids in the Urgineoideae and cardenolids in Ornithogaloideae. Even
Linnaean genera such as
Hyacinthus,
Scilla and
Ornithoglum proved heterogeneous and characters useful in other families failed to define satisfactory taxa.
Modern classifications '', glory-of-the-snow Modern classification systems for plants are largely derived from
molecular phylogenetic analysis. The initial molecular analysis of the Liliaceae
s.l. was based on the Dahlgren system, as for example in the work by Chase et al. in 1995. When it was discovered that the Dahlgren families were not
monophyletic, the tendency was to create new families out of each identified
clade, as in the first
Angiosperm Phylogeny Group system of 1998, the
APG system. This placed many lilioid families and genera in the order
Asparagales (a term derived from Dahlgren, and the largest monocot order). One of the 29 families into which the Asparagales were divided was the Hyacinthaceae. With further work it was evident that these 29 families, some of which had few genera, could be grouped into larger clades. The
APG II system of 2003 was a compromise. It divided the Asparagales into 14 broadly defined families, while allowing an alternative system in which some of the larger families could be replaced by smaller ones. The Hyacinthaceae was one of these optional smaller families, which could alternatively be sunk into a broadly defined Asparagaceae. This compromise approach was abandoned in the
APG III system of 2009, which allowed only the broader families. The paper presenting the system states "The area around Asparagaceae is difficult from the standpoint of circumscription. Although Asparagaceae s.l. are heterogeneous and poorly characterized, Asparagaceae s.s., Agavaceae, Laxmanniaceae, Ruscaceae and even Hyacinthaceae have few if any distinctive features." At the same time, Chase et al. provided subfamilies to replace the alternative narrowly defined families of APG II. The Hyacinthaceae became the subfamily Scilloideae of the family Asparagaceae. Many sources have adopted the APG III system; for example, the
World Checklist of Selected Plant Families places genera such as
Hyacinthus only in the broadly defined Asparagaceae. Other sources prefer to retain the narrower families of APG II; for example, Seberg et al. say that it "remains a moot point whether the difficult-to-recognize bracketed families of APG II are a worse or a better choice than the equally difficult-to-recognize subfamilies of APG III", and in their analyses of the phylogeny of the Asparagales they continue to use families such as Hyacinthaceae.
Tribes In 1990, Pfosser and Speta stated that their earlier classification of the Hyacinthaceae into the subfamilies Hyacinthoideae, Ornithogaloideae, Oziroeoideae and Urgineoideae continued to be supported by ongoing studies. (They further divided the subfamilies Hyacinthoideae and Ornithogaloideae into tribes.) A part of reducing the Hyacinthaceae to the subfamily Scilloideae, Chase et al. (2009) suggested dividing it into four tribes, corresponding to Pfosser and Speta's four subfamilies: Hyacintheae Dumort., Ornithogaleae Rouy, Oziroëeae M.W.Chase, Reveal & M.F.Fay and Urgineeae Rouy. Hyacintheae was further divided into three subtribes: Pseudoprosperinae, Massoniinae and Hyacinthinae. The possible relationship of the tribes and subtribes is illustrated in the following
cladogram, which has, however, only "moderate" statistical support. }} The exact boundaries between genera within these tribes remains controversial; the situation has been described as being in a "state of flux".
Oziroëeae Species are found only in western South America. They have flowers with stamens which are joined to the petals, rounded seeds and the embryo as long as the seed. The basic chromosome numbers are
n = 15, 17. The tribe contains only the genus
Oziroë.
Ornithogaleae '' In terms of the number of species, this is the second largest tribe. Its species are distributed in Europe, western Asia and Africa. They have flowers with three stamens which have flattened filaments. Their seeds are flattened and angular. The basic chromosome numbers range from
n = 2 to
n = 10. In the treatment by Manning et al. (2009) and Stevens at the
Angiosperm Phylogeny Website, the tribe contains four genera,
Albuca (about 110–140 species),
Dipcadi,
Ornithogalum (about 160 species, including
Galtonia and
Neopatersonia) and
Pseudogaltonia. By contrast, Martínez-Azorín et al. (2011) divide the tribe into 19 genera.
Urgineeae Species within this tribe contain
bufadienolides and are distributed mainly in Africa, Madagascar, and the Mediterranean through to India. The seeds are flattened and winged with the head barely attached to the
endosperm. The basic chromosome numbers are
n = 6, 7 and 10. Depending on the source, the tribe may include the genera
Bowiea,
Drimia (including
Urginea),
Schizobasis (sometimes included in
Drimia) and
Fusifilum (also sometimes included in
Drimia).
Hyacintheae '' In terms of the number of species, this is the largest tribe. Its species have leaves with pustules or spots, rounded seeds and contain
homoisoflavanones. The tribe can in turn be divided into three subtribes: •
Pseudoprosperinae Speta : A monotypic subtribe genus with a single species,
Pseudoprospero firmifolium, is from eastern South Africa. It has two ovules per carpel with one seed per locule and a basic chromosome number
n = 9. • Massoniinae Bentham & Hooker f. : Species are distributed in Africa south of the Sahara and India. There are two or more ovules per carpel. The seeds have
elaiosomes. The basic chromosome number is 5 to 10+ (many 20). The subtribe contains about 13–20 genera (depending on the treatment), including
Daubenya,
Drimiopsis,
Eucomis,
Lachenalia (about 110 species),
Ledebouria (about 80 species),
Massonia (including
Whiteheadia),
Merwilla,
Schizocarphus and
Veltheimia. • Hyacinthinae Parlatore : Species are distributed in Europe, the Mediterranean and North Africa and the Middle East, and then again in the Far East. There are two to eight ovules per carpel; elaiosomes are present in the seeds; and the basic chromosome number is 4 to 8+. The subtribe contains about 14–25 genera (depending on the treatment), including
Bellevalia (about 50 species),
Brimeura,
Hyacinthoides,
Muscari (about 50 species),
Scilla (about 30 species) and
Prospero (about 25 species).
Genera and species Some genera that were formerly placed within the Scillioideae (as Hyacinthaceae), e.g.,
Chlorogalum and
Camassia, are currently placed in the Agavoideae. Both historically and , there has been "considerable disagreement over generic limits" in the remaining Scilloideae, with different sources listing from 15 to 45 genera for
sub-Saharan Africa alone. The total number of genera has been given as anything between about 30 (with about 500–700 species) and 70 (with about 1000 species).
List of genera Unless otherwise noted, the list below is based on genera accepted by the
World Checklist of Selected Plant Families as in the family
Asparagaceae (with synonyms from the same source), with assignments to the subfamily Scilloideae based on the Germplasm Resources Information Network. As noted above, other sources divide up some of these genera, creating a significantly larger number; thus the genus
Ornithogalum as conceived by Manning et al. (2009) is divided by Martínez-Azorín et al. (2011) into a more narrowly circumscribed
Ornithogalum plus an additional 11 genera. == Distribution and ecology ==