Flower morphology of the Geraniales is rather conserved. They are usually perfectly pentamerous and pentacyclic without fused organs besides the carpels of the superior gynoecium. The androecium is obdiplostemonous. Only a few genera are tetramerous (Francoa,Dimorphopetalum,Melianthus). In some genera some stamens (Pelargonium) or a complete whorl of stamens are reduced (Erodium,Melianthus). In the generaHypseocharis andMonsonia there are 15 instead of the usual ten stamens. Most genera bear nectariferous flowers.[2] The nectary glands are formed by the receptacle and are localised at the bases of the antesepalous stamens.[2][3]
The economic importance of Geraniales is low. Some species of the genusPelargonium (Geraniaceae) are cultivated for theiraromatic oil used in theperfume industry. Some other species, also mostly within Geraniaceae, havehorticultural or medicinal uses. APaleobotanic record is missing.
Thebotanical authority for Geraniales is given toJussieu,[4] but since the original description did not fulfill all the rules for avalid publication and was subsequently validly published, attribution is given to both Jussieu and the subsequent publication, hence the designation GeranialesJuss.exBercht. &J.Presl Jussieu, who developed the concept of botanical families, described the Gerania, as a grouping of five genera, includingGeranium. Although Jussieu used the termOrdo this did not correspond to current understandings of the term Order.[5][6] The subsequent attribution occurred in 1820, in the Czech textO Prirozenosti Rostlin, byFriedrich von Berchtold andJan Svatopluk Presl, henceex Bercht. & J.Presl.[7] However, Berchtold and Presl also only described arad (ordo) of five genera, which they called Geraniae.[8] Other authorities have given the authority toDumortier who described the family Geraniaceae, consisting of two tribes, Pelargonieae and Geranieae, each with three genera.[9]
While theDahlgren system (1980) was much larger in circumscription with 16 families, only two of which were in Cronquist's construction, and placed the order in thesuperorderRutiflorae:[13]
Other modernsystems include those ofTakhtajan (1987) with nine families, andThorne (1992). Thorne's system was the same as Cronquist's except that Biebersteiniaceae, Dirachmaceae, Ledocarpaceae, and Vivianiaceae were reduced tosubfamilies of Geraniaceae.
Molecular phylogenetics: Angiosperm Phylogeny Group
The elucidation of the relationships within the order bymorphological orcytological methods alone had proven difficult as demonstrated by the widely different treatment by various authorities. For instance Cronquist andThorne immersed the families Biebersteiniaceae, Dirachmaceae, Ledocarpaceae, Rhynchothecaceae and Vivianiaceae within Geraniaceae (Geraniaceaesensu lato), whereas Dahlgren and others maintained them as separate taxa, maintaining a "core" Geraniaceae (Geraniaceaesensu stricto).[11] Price and Palmer (1993) were among the first investigators to applymolecular phylogenetics to this order, using thechloroplastgenerbcL.[11][14][a] This disassembled the traditional morphologically defined grouping ofdicotyledons, replacing it with a series of nestedclades. The Geranialessegregated in theeudicotclade, specifically in therosidsubclade.
Hypseocharitaceae were a small family of eight species of the genusHypseocharis found in the tropical mountainous regions of theAndes.[17] The APG provided the option of considering them as a separate family or subsumed into Geraniaceae. By 2003, when the APG was published, it was apparent that the small families Francoaceae, Greyiaceae and Melianthaceae were closely related and were collapsed into one family as Melanthiaceae with Francoaceae as an optional synonym. Thus the number of families was reduced to four.[18]
However, Considerable rearrangements took place in the 2016APG IV system.Francoaceae was substituted forMelianthaceae,due tonomenclatural priority. The latter subsumedVivianiaceae based on the work of Sytsma, Spalink & Berger (2014). However, there remains substantial uncertainty regarding the relationships within Francoaceaesensu stricto (s.s.), Melianthaceae (BersamaFresen. and Melianthus L.) and Ledocarpaceae. Here, Vivianiaceae is used as a latersynonym for Ledocarpaceae. This due to conflicting evidence (see Palazzesiet al., 2012). The APG chose to follow the broader circumscription for the time being till these differences are resolved.
The Vivianiaceae and Ledocarpaceae were included within the Geraniaceae, and the Hypseocharitaceae within the Oxalidaceae, which are now treated in the orderOxalidales. The Melianthaceae were placed within theSapindales, the Greyiaceae and Francoaceae within theRosales, the latter subsumed within theSaxifragaceae.
Recent comparison of DNA-fragments from species within the order resulted in the following phylogenetic tree.[20]
Chase, Mark W.;Soltis, Douglas E.; Olmstead, Richard G.; Morgan, David; Les, Donald H.; Mishler, Brent D.; Duvall, Melvin R.; Price, Robert A.; Hills, Harold G.; Qiu, Yin-Long; Kron, Kathleen A.; Rettig, Jeffrey H.; Conti, Elena; Palmer, Jeffrey D.; Manhart, James R.; Sytsma, Kenneth J.; Michaels, Helen J.; Kress, W. John; Karol, Kenneth G.; Clark, W. Dennis; Hedren, Mikael; Gaut, Brandon S.; Jansen, Robert K.; Kim, Ki-Joong; Wimpee, Charles F.; Smith, James F.; Furnier, Glenn R.; Strauss, Steven H.; Xiang, Qui-Yun; Plunkett, Gregory M.;Soltis, Pamela S.; Swensen, Susan M.; Williams, Stephen E.; Gadek, Paul A.; Quinn, Christopher J.; Eguiarte, Luis E.; Golenberg, Edward; Learn, Gerald H.; Graham, Sean W.; Barrett, Spencer C. H.; Dayanandan, Selvadurai; Albert, Victor A. (1993)."Phylogenetics of Seed Plants: An Analysis of Nucleotide Sequences from the Plastid GenerbcL"(PDF).Annals of the Missouri Botanical Garden.80 (3): 528.doi:10.2307/2399846.hdl:1969.1/179875.JSTOR2399846.
Dahlgren, R. M. T. (February 1980). "A revised system of classification of the angiosperms".Botanical Journal of the Linnean Society.80 (2):91–124.doi:10.1111/j.1095-8339.1980.tb01661.x.