Abstract
• Background and Aims Fifty-two populations were sampled in order to establish the taxonomic delimitation and relationships of eight taxa belonging to the A. majus L. and A. siculum Miller groups.
• Methods Data on 13 allozyme loci were recorded after extraction of fresh leaves and electrophoresis on horizontal 10 % starch gels.
• Key Results Genetic distances between conspecific populations are lower than for other species of the genus.
• Conclusions These results support the recognition of A. majus, A. tortuosum, A. linkianum, A. cirrigherum, A. litigiosum and A. barrelieri at specific rank. The genetic distances, together with the lack of morphological differences and the sympatric distribution ranges, support the inclusion of A. australe into A. tortuosum, A. dielsianum into A. siculum, and A. latifolium subsp. intermedium as a synonym of A. latifolium. The results support separation of the taxa studied into two groups, coinciding with series Sicula Rothm. and Majora, but disagreeing with the arrangement of species into them. According to our results, Sicula consist of A. siculum and Majora consists of A. latifolium, A. majus, A. tortuosum, A. linkianum, A. cirrigherum, A. litigiosum and A. barrelieri.
Keywords: Antirrhinum, A. majus, A. siculum, A. australe, snapdragons, allozymes, systematics, taxonomy
INTRODUCTION
The genus Antirrhinum L. includes some 25 diploid (2n = 16) species mainly distributed in the Iberian Peninsula (Sutton, 1988). Antirrhinum majus L. is widely used as an ornamental and is one of the model species in genetic regulation research. Rothmaler (1956) divided the genus into three subsections with several series. Recently, Fernández Casas (1997) restructured the genus, mainly elevating the rank of infrageneric taxa from subsections to sections and from series to subsections. Here we follow Rothmaler. The morphological discrimination of Antirrhinum species is complex (Sutton, 1988), leading to different views both on the discrimination of taxa and in the assignment of rank. The group of species related to A. majus, with up to five subspecies, and other closely related species such as A. australe Rothm., A. latifolium Miller, A. barrelieri Boreau, A. dielsianum Rothm. and A. siculum Miller, is the most complex group in the genus (Table 1).
Table 1.
Comparison of the different taxonomic treatments of the studied taxa according to different authors
| Sections, Subsections, Series |
Consideration of the studied taxa |
||||||||
|---|---|---|---|---|---|---|---|---|---|
|
Rothmaler (1956) |
Fernández-Casas (1997) |
Rothmaler (1956) |
Webb (1971) |
Sutton (1988) |
Fernández-Casas (1997) |
||||
| Subsect. Antirrhinum Series Majora | Sect. Antirrhinum Subsect. Antirrhinum | A. majus subsp. majus var. pseudomajus var. striatum (=A. intermedium) A. majus subsp. litigiosum A. majus subsp. tortuosum A. majus subsp. linkianum var. linkianum var. ramossisimum (=A. cirrhigerum) A. majus subsp. latifolium | A. majus subsp. majus subsp. tortuosum subsp. linkianum subsp. cirrhigerum A. latifolium | A. majus subsp. majus subsp. litigiosum subsp. tortuosum subsp. linkianum subsp. cirrhigerum A. latifolium subsp. latifolium subsp. intermedium | A.majus subsp. majus var. pseudomajus var. striatum A. majus subsp. litigiosum A. majus subsp. tortuosum A. majus subsp. linkianum var. linkianum var. ramossisimum | ||||
| Subsect. Antirrhinum Series Sicula | Subsect. Sicula | A. siculum A. dielsianum A. barrelieri | A. siculum | A. siculum A. barrelieri | A. siculum A. dielsianum A. barrelieri | ||||
| Subsect. Antirrhinum Series Hispanica | Subsect. Hispanica | A. australe | A. australe | A. australe | A. australe | ||||
All of the taxa studied are included in Subsection Antirrhinum, together with A. graniticum Rothm. and A. boissieri Rothm., with which the taxa considered here show clear morphological differences in indument, plant size, and flower size and colour. All of them grow on limestones, sometimes on soil, and most of them have narrow and geographically separated ranges. Antirrhinum majus subsp. tortuosum (Vent.) Rouy and A. siculum have wider ranges. The former is widespread in the Mediterranean region while A. siculum reaches south-west Asia. With the exception of the cultivated plants of A. majus, all these taxa are self-incompatible outcrossers (Gruber, 1930, 1932; East, 1940).
The species of Antirrhinum are mostly allopatric or parapatric, sometimes sympatric, fully interfertile (Rothmaler, 1956), and data from ITS sequences (Oyama and Baum, 2004) suggest they are closely related. Differentiation of species in the genus is hypothesized based on the accumulation of differences encoded by single genes combined with geographic barriers (see several authors in Fernández Casas, 1997) together with the ability of pollinators to discriminate between species (Mather, 1947).
Allozymes are molecular markers useful for establishing relationships at low systematic levels (Gottlieb, 1984; Crawford, 1985). They have been used to study several American genera of tribe Antirrhineae Dumort (Elisens and Crawford, 1988; Elisens, 1992; Elisens and Nelson, 1993) and to delimit species in Antirrhinum (Mateu-Andrés, 1999; Mateu-Andrés and Segarra-Moragues, 2000, 2003b). Here we have studied taxa related to A. majus and A. siculum in order to establish their taxonomic delimitation and relationships.
MATERIALS AND METHODS
Plant material
Seeds were sampled from 52 populations (Table 2, Fig. 1). Generally, seeds were sampled from between 25 % and 50 % of the total number of mature plants in each population. Up to 30 seedlings per population were raised under glasshouse conditions. In all, 881 individuals were studied.
Table 2.
Sampled taxa and populations
| Species |
Population code |
Country* |
Province |
Locality |
|---|---|---|---|---|
| A. siculum (AS) | AS1 | It | Sicily | Palermo |
| AS3 | It | Sicily | Nicolosi | |
| AS4 | It | Sicily | Taormina | |
| AS6 | It | Napoles | Napoles | |
| AS7 | Is | Haifa | Haifa | |
| AS8 | It | Grosseto | Santo Stefano | |
| A. dielsianum (AD) | AD2 | It | Sicily | Siracusa |
| AD5 | It | Sicily | Messina | |
| A. majus (AM) | AM2 | Hs | Gerona | Queralbs |
| AM3 | FR | Pyréneés Orientales | Colliure | |
| AM4 | Hs | Gerona | Banyolas | |
| AM5 | Hs | Gerona | Blanes | |
| A. tortuosum (AT) | AT1 | It | Sicily | Agrigento |
| AT2 | Hs | Sevilla | Alcalá de Guadaira | |
| AT4 | Hs | Córdoba | Cerro Muriano | |
| AT5 | Hs | Córdoba | de Posadas a Espiel | |
| AT8 | Hs | Córdoba | Lucena | |
| AT9 | It | Lacio | Roma | |
| AT10 | It | Latina | Norma | |
| A. australe (AA) | AA1 | Hs | Cádiz | Zahara de la Sierra |
| AA2 | Hs | Cádiz | Peñón de la Parra | |
| AA3 | Hs | Cádiz | Grazalema | |
| AA4 | Hs | Cádiz | Benaocaz | |
| AA6 | Hs | Málaga | Torcal de Antequera | |
| AA7 | Hs | Córdoba | de Lucena a Cabra | |
| AA8 | Hs | Córdoba | Carcabuey | |
| AA9 | Hs | Jaén | Alcaudete | |
| A. intermedium (ALI) | ALI2 | Fr | Pyréneés Orientales | Villefranche de Conflent |
| ALI3 | Fr | Pyréneés Orientales | Usson-les-Bains | |
| A. latifolium (ALL) | ALL2 | Hr | Gerona | Tossas |
| ALL3 | Fr | Provence | La Fontaine de Vaucluse | |
| ALL4 | Fr | Massif Central | Beaulieu | |
| ALL5 | Fr | Alpes Maritimes | Auron | |
| ALL6 | Fr | Alpes Maritimes | Valdeblore-Rimples | |
| ALL7 | Fr | Alpes Maritimes | Menton | |
| A. cirrhigerum (AC) | AC5 | Lu | Beira Litoral | San Pedro de Muel |
| AC6 | Lu | Beira Litoral | Figueira da Foz | |
| AC7 | Lu | Douro Litoral | Aveiro | |
| A. linkianum (ALK) | ALK1 | Lu | Estremadura | Serra da Arrabida |
| ALK2 | Lu | Estremadura | Trafaria | |
| ALK3 | Lu | Estremadura | Almada | |
| ALK6 | Lu | Estremadura | Livramento | |
| ALK7 | Lu | Beira Litoral | Coimbra | |
| ALK8 | Lu | Estremadura | Pernes | |
| A. litigiosum (ALIT) | ALIT6 | Hs | Castellón | Begis |
| ALIT9 | Hs | Tarragona | Mas de Barberans | |
| ALIT17 | Hs | Tarragona | Sant Vicenç de Calders | |
| ALIT19 | Hs | Valencia | Lliria | |
| ALIT20 | Hs | Valencia | Bugarra | |
| A. barrelieri (AB) | AB14 | Hs | Granada | Laroles |
| AB15 | Hs | Granada | Ugijar | |
| AB16 | Hs | Granada | Padul |
Fr = France; Hs = Spain; Is = Israel; It = Italy; Lu = Portugal.
Fig. 1.
Distribution maps of the populations studied.
Electrophoresis and analysis
Electrophoresis was carried out on horizontal 10 % starch gels. The extraction buffer consisted of 0·2 m Tris–HCl pH 7·5, 2 mm EDTA. 0·12 m Na2S2O5, 1 m Cl2Mg, 40 mg mL−1 (w/v) PVP, and 4 mL mL−1 mercaptoethanol. Material used consisted of young leaves of plants grown in the greenhouse. Extracts were absorbed onto 3 mm wicks of Whatman chromatography paper.
Nine enzyme systems were assayed: aconitase (ACO, EC 4.2.1.3), aspartate aminotransferase aminotransferase (AAT, EC 2.6.1.1), diaphorase (DIA, EC 1.6.99), isocitrate dehydrogenase (IDH, EC 1.1.1.42), malate dehydrogenase (MDH, EC 1.1.1.37), menadione reductase (MNR, EC 1.6.99), phosphoglucoisomerase (PGI, EC 5.3.1.9), phosphoglucomutase (PGM, EC 5.4.2.2) and triose-phosphate isomerase (TPI, EC 5.3.1.1), and all gave banding patterns. AAT and IDH could not be scored due to inconsistent banding patterns. The electrophoretic buffer system II of Wendel and Weeden (1989) was employed to resolve IDH and MDH; system VI for DIA, PGI, PGM, and TPI; and system VII for AAT, ACO and MNR. All staining methods followed Wendel and Weeden (1989), and AAT was modified following H.C. Prentice (pers. comm.).
As reported in previous studies in Antirrhineae, both in other genera (Elisens and Crawford, 1988; Elisens, 1992; Elisens and Nelson, 1993) and in Antirrhinum (Mateu-Andrés, 1999; Mateu-Andrés and Segarra, 2000, 2003a, b), PGI1 and TPI1 showed differences in band thickness which were interpreted as duplicated comigrant loci, so they were not scored.
Routines in BIOSYS-1 (Swofford and Selander, 1989) were used to generate both UPGMA phenograms from Nei's (1972) and modified Rogers's (Wright, 1978) distances.
RESULTS
Seven out of the nine enzyme systems were interpreted, giving a total of 13 putative loci, 12 of them with two or more alleles and only one (MNR1) with one fixed allele. The 13 scored loci gave a total number of 39 alleles summarized in Table 3. Original data are available upon request (Isabel.Mateu@uv.es). PGM1-2 was fixed in 42 out of the 44 populations in which it was present, and four other alleles (DIA3-2, MNR2-3, PGM2-3, TPI2-2) were fixed in many populations (mean number of alleles across populations: 53·8 %, 46·1 %, 23·1 % and 67·3 %, respectively).
Table 3.
Summary of allele frequencies in the taxa studied, including means and ranges (in brackets) across populations for each taxon
| Loci |
Allele |
AS |
AD |
AM |
AT |
AA |
ALL |
ALI |
AC |
ALK |
ALIT |
AB |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ACO1 | 1 | 1·000 | 1·000 | 0·022 (0·000–0·063) | 0·08 (0·000–0·250) | 0·200 (0·060–0·481) | 0·000 | 0·000 | 0·147 (0·000–0·417) | 0·146 (0·000–0·250) | 0·241 (0·000–0·643) | 0·250 (0·250) |
| 2 | 0·000 | 0·000 | 0·978 (0·938–1·000) | 0·920 (0·750–1·000) | 0·800 (0·512–0·983) | 1·000 | 1·000 | 0·853 (0·583–1·000) | 0·854 (0·750–1·000) | 0·759 (0·357–1·000) | 0·750 (0·750) | |
| ACO2 | 1 | 1·000 | 1·000 | 0·074 (0·071–1·000) | 0·080 (0·000–0·333) | 0·065 (0·000–0·250) | 0·040 (0·000–0·250) | 0·000 | 0·606 (0·429–0·694) | 0·037 (0·000–0·091) | 0·128 (0·000–0·375) | 0·000 |
| 2 | 0·000 | 0·000 | 0·000 | 0·715 (0·167–1·000) | 0·793 (0·467–1·000) | 0·000 | 0·000 | 0·394 (0·306–0·571) | 0·923 (0·813–1·000) | 0·771 (0·625–1·000) | 0·500 (0·500) | |
| 3 | 0·000 | 0·000 | 0·926 (0·875–1·000) | 0·079 (0·000–0·318) | 0·142 (0·000–0·533) | 0·960 (0·750–1·000) | 1·000 | 0·000 | 0·040 (0·000–0·136) | 0·101 (0·000–0·300) | 0·500 (0·500) | |
| 4 | 0·000 | 0·000 | 0·000 | 0·127 (0·000–0·500) | 0·000 | 0·000 | 0 | 0·000 | 0·000 | 0·000 | 0·000 | |
| DIA1 | 1 | 0·000 | 0·000 | 0·000 | 0·106 (0·000–0·333) | 0·037 (0·000–0·250) | 0·042 (0·000–0·143) | 0·107 (0·000–0·214) | 0·020 (0·000–0·060) | 0·008 (0·000–0·050) | 0·000 | 0·014 (0·000–0·042) |
| 2 | 1·000 | 1·000 | 1·000 | 0·800 (0·333–1·000) | 0·733 (0·367–0·944) | 0·786 (0·643–0·875) | 0·773 (0·714–0·833) | 0·454 (0·350–0·632) | 0·623 (0·182–0·975) | 0·953 (0·765–1·000) | 0·986 (0·958–1·000) | |
| 3 | 0·000 | 0·000 | 0·000 | 0·095 (0·000–0·333) | 0·230 (0·056–0·583) | 0·172 (0·056–0·333) | 0·120 (0·071–0·167) | 0·526 (0·368–0·650) | 0·369 (0·025–0·818) | 0·047 (0·000–0·235) | 0·000 | |
| DIA3 | 1 | 1·000 | 1·000 | 0·250 (0·250) | 0·010 (0·000–0·071) | 0·005 (0·000–0·038) | 0·010 (0·000–0·071) | 0·070 (0·000–0·143) | 0·000 | 0·085 (0·000–0·250) | 0·024 (0·000–0·063) | 0·028 (0·000–0·083) |
| 2 | 0·000 | 0·000 | 0·750 (0·750) | 0·990 (0·929–1·000) | 0·995 (0·962–1·000) | 0·980 (0·929–1·000) | 0·930 (0·857–1·000) | 1·000 | 0·915 (0·750–1·000) | 0·964 (0·875–1·000) | 0·945 (0·917–1·000) | |
| 3 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·010 (0·000–0·056) | 0·000 | 0·000 | 0·000 | 0·013 (0·000–0·063) | 0·028 (0·000–0·083) | |
| MDH1 | 1 | 0·000 | 0·000 | 0·255 (0·000–0·385) | 0·117 (0·000–0·306) | 0·200 (0·020–0·397) | 0·035 (0·000–0·214) | 0·226 (0·167–0·286) | 0·575 (0·447–0·778) | 0·853 (0·632–1·000) | 0·393 (0·267–0·472) | 0·204 (0·083–0·278) |
| 2 | 1·000 | 1·000 | 0·745 (0·615–1·000) | 0·883 (0·694–1·000) | 0·800 (0·603–0·980) | 0·965 (0·786–1·000) | 0·774 (0·714–0·833) | 0·425 (0·222–0·553) | 0·147 (0·000–0·368) | 0·607 (0·500–0·733) | 0·796 (0·722–0·917) | |
| MDH2 | 1 | 0·000 | 0·000 | 0·340 (0·000–0·500) | 0·145 (0·000–0·339) | 0·256 (0·100–0·414) | 0·035 (0·000–0·214) | 0·416 (0·333–0·500) | 0·930 (0·789–1·000) | 0·930 (0·727–1·000) | 0·519 (0·333–0·750) | 0·213 (0·000–0·389) |
| 2 | 1·000 | 1·000 | 0·660 (0·500–1·000) | 0·855 (0·661–1·000) | 0·744 (0·438–0·914) | 0·964 (0·786–1·000) | 0·584 (0·500–0·667) | 0·070 (0·000–0·211) | 0·071 (0·000–0·273) | 0·481 (0·250–0·667) | 0·787 (0·611–1·000) | |
| MDH3 | 1 | 0·000 | 0·000 | 0·295 (0·214–0·346) | 0·270 (0·033–0·500) | 0·330 (0·120–0·500) | 0·100 (0·000–0·278) | 0·310 (0·286–0333) | 0·009 (0·000–0·026) | 1·000 | 0·101 (0·000–0·300) | 0·449 (0·389–0·500) |
| 2 | 1·000 | 1·000 | 0·706 (0·654–0·786) | 0·730 (0·500–0·967) | 0·670 (0·500–0·880) | 0·900 (0·722–1·000) | 0·690 (0·667–0·714) | 0·991 (0·974–1·000) | 0·000 | 0·874 (0·700–1·000) | 0·551 (0·500–0·611) | |
| 3 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·025 (0·000–0·125) | 0·000 | |
| MNR1 | 1 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 |
| MNR2 | 1 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·116 (0·000–0·250) | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 |
| 2 | 0·005 (0·000–0·330) | 0·000 | 0·147 (0·000–0·375) | 0·000 | 0·225 (0·083–0·360) | 0·024 (0·000–0·143) | 0·035 (0·000–0·071) | 0·261 (0·000–0·500) | 0·066 (0·000–0·313) | 0·240 (0·111–0·542) | 0·097 (0·000–0·250) | |
| 3 | 0·995 (0·967–1·000) | 1·000 | 0·853 (0·625–1·000) | 1·000 | 0·775 (0·640–0·917) | 0·860 (0·750–1·000) | 0·965 (0·929–1·000) | 0·739 (0·500–1·000) | 0·934 (0·688–1·000) | 0·760 (0·458–0·889) | 0·903 (0·750–1·000) | |
| PGI2 | 1 | 0·000 | 0·000 | 0·078 (0·000–0·188) | 0·000 | 0·040 (0·000–0·132) | 0·000 | 0·000 | 0·000 | 0·096 (0·000–0·292) | 0·086 (0·000–0·133) | 0·000 |
| 2 | 0·000 | 0·000 | 0·008 (0·000–0·033) | 0·055 (0·000–0·200) | 0·000 | 0·050 (0·000–0·167) | 0·000 | 0·000 | 0·127 (0·000–0·429) | 0·000 | 0·104 (0·000–0·313) | |
| 3 | 1·000 | 0·970 (0·938–1·000) | 0·905 (0·813–1·000) | 0·920 (0·857–1·000) | 0·940 (0·862–1·000) | 0·900 (0·727–1·000) | 0·960 (0·917–1·000) | 0·896 (0·750–1·000) | 0·695 (0·429–0·864) | 0·914 (0·867–1·000) | 0·840 (0·688–1·000) | |
| 4 | 0·000 | 0·030 (0·000–0·063) | 0·008 (0·000–0·033) | 0·025 (0·000–0·143) | 0·020 (0·000–0·138) | 0·050 (0·000–0·182) | 0·040 (0·000–0·083) | 0·104 (0·000–0·250) | 0·083 (0·000–0·177) | 0·000 | 0·056 (0·000–0·167) | |
| PGM1 | 1 | 1·000 | 1·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 |
| 2 | 0·000 | 0·000 | 1·000 | 1·000 | 0·995 (0·983–1·000) | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | 1·000 | |
| 3 | 0·000 | 0·000 | 0·000 | 0·000 | 0·005 (0·000–0·028) | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | |
| PGM2 | 1 | 0·000 | 0·234 (0·000–0·469) | 0·084 (0·050–0·133) | 0·000 | 0·000 | 0·000 | 0·000 | 0·006 (0·000–0·019) | 0·014 (0·000–0·081) | 0·104 (0·000–0·250) | 0·208 (0·167–0·250) |
| 2 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 1·000 | 0·000 | 0·000 | 0·044 (0·000–0·222) | 0·000 | |
| 3 | 0·000 | 0·765 (0·531–1·000) | 0·916 (0·867–0·950) | 0·090 (0·000–0·286) | 0·135 (0·000–0·267) | 0·923 (0·667–1·000) | 0·000 | 0·994 (0·981–1·000) | 0·987 (0·919–1·000) | 0·583 (0·528–0·694) | 0·792 (0·750–0·833) | |
| 4 | 0·000 | 0·000 | 0·000 | 0·910 (0·714–1·000) | 0·865 (0·680–1·000) | 0·077 (0·000–0·333) | 0·000 | 0·000 | 0·000 | 0·269 (0·000–0·433) | 0·000 | |
| 5 | 0·166 (0·000–1·000) | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | |
| TPI2 | 1 | 0·000 | 0·000 | 0·025 (0·000–0·100) | 0·000 | 0·000 | 0·000 | 0·100 (0·000–0·167) | 0·000 | 0·043 (0·000–0·258) | 0·390 (0·111–0·813) | 0·083 (0·000–0·167) |
| 2 | 1·000 | 1·000 | 0·960 (0·900–1·000) | 0·980 (0·857–1·000) | 1·000 | 0·915 (0·667–1·000) | 0·900 (0·833–1·000) | 1·000 | 0·957 (0·742–1·000) | 0·604 (0·188–0·889) | 0·889 (0·750–1·000) | |
| 3 | 0·000 | 0·000 | 0·016 (0·000–0·063) | 0·020 (0·000–0·143) | 0·000 | 0·085 (0·000–0·333) | 0·000 | 0·000 | 0·000 | 0·000 | 0·028 (0·000–0·083) | |
| 4 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·000 | 0·006 (0·000–0·028) | 0·000 |
Ranges are given by two values (minimum–maximum). Where there is a single value, that is the average, meaning that all values are equal.
The most frequent alleles were, in general, shared by all the studied taxa (Table 3). ACO1-2, and DIA3-2 were the most frequent alleles in all taxa except A. siculum and A. dielsianum, in which ACO1-1 and DIA3-1 were fixed; MDH3-2 was the most frequent allele in all the taxa except A. linkianum in which MDH3-1 was fixed; and PGM2-4 was the most frequent allele in A. tortuosum, while MDH3-2 and PGM2-3 were the most frequent alleles in all of the other taxa. Two alleles were shared by couples of species as most frequent, A. majus and A. latifolium shared ACO2-3, and A. cirrigherum and A. linkianum shared MDH2-1 (Table 3). Antirrhinum siculum, A. dielsianum and A. cirrigherum shared ACO2-1.
One allele (PGM1-1) was shared both by A. siculum and A. dielsianum, being fixed in all the populations. Three alleles were exclusive to one taxon, ACO2-4, MNR2-1 and PGM2-5 to A. tortuosum, A. latifolium and A. siculum, respectively (Table 3).
Antirrhinum siculum and A. dielsianum shared particular patterns (Table 3), with ten alleles fixed in all of the populations (ACO1-1, ACO2-1, DIA1-2, DIA3-1, MDH1-2, MDH2-2, MDH3-2, MNR1-1, PGM1-1 and TPI2-2) and three in the vast majority of them (MNR2-3, PGI2-3 and PGM2-3). In addition, one allele (PGM2-5) was found exclusively in taxon AS and was fixed in population AS1.
A summary of Nei's genetic distances, both within and between taxa, is presented in Table 4. These values are lower than those previously reported for other Antirrhinum species for these taxa based on allozymes (Mateu-Andrés, 1999; Mateu-Andrés and Segarra-Moragues, 2003b).
Table 4.
Summary of Nei's genetic distances within (diagonal) and between taxa. Both average and range across populations are given for each taxon (in brackets)
| AS |
AD |
AM |
AT |
AA |
ALL |
ALI |
AC |
ALK |
ALIT |
AB |
|
|---|---|---|---|---|---|---|---|---|---|---|---|
| AS | 0·02 (0·000–0·077 | ||||||||||
| AD | 0·02 (0·020) | 0·01 (0·02–0·057) | |||||||||
| AM | 0·29 (0·257–0·375) | 0·29 (0·257–0·375) | 0·01 (0·000–0·035) | ||||||||
| AT | 0·36 (0·314–0·386) | 0·35 (0·302–0·382) | 0·13 (0·073–0·174) | 0·03 (0·000–0·066) | |||||||
| AA | 0·35 (0·318–0·377) | 0·34 (0·311–0·374) | 0·18 (0·074–0·154) | 0·04 (0·001–0·076) | 0·01 (0·000–0·045) | ||||||
| ALL | 0·33 (0·277–0·395) | 0·31 (0·278–0·334) | 0·03 (0·000–0·048) | 0·12 (0·074–0·181) | 0·13 (0·071–0·174) | 0·01 (0·00–0·014) | |||||
| ALI | 0·32 (0·318–0·375) | 0·32 (0·318–0·334) | 0·08 (0·000–0·027) | 0·10 (0·102–0·184) | 0·12 (0·090–0·156) | 0·07 (0·000–0·033) | 0 | ||||
| AC | 0·36 (0·326–0·444) | 0·36 (0·328–0·383) | 0·13 (0·085–0·195) | 0·17 (0·115–0·269) | 0·16 (0·109–0·198) | 0·15 (0·097–0·216) | 0·13 (0·084–0·145) | 0·03 (0·028–0·041) | |||
| ALK | 0·50 (0·424–0·638) | 0·50 (0·424–0·576) | 0·18 (0·119–0·296) | 0·21 (0·130–0·332) | 0·20 (0·135–0·277) | 0·23 (0·158–0·337) | 0·17 (0·129–0·198) | 0·13 (0·097–0·157) | 0·04 (0·003–0·08) | ||
| ALIT | 0·32 (0·250–0·420) | 0·32 (0·250–0·376) | 0·10 (0·053–0·125) | 0·08 (0·050–0135) | 0·10 (0·035–0·118) | 0·12 (0·088–0·163) | 0·08 (0·068–0·128) | 0·10 (0·050–0·157) | 0·15 (0·086–0·243) | 0·06 (0·023–0·125) | |
| AB | 0·28 (0·259–0·420) | 0·27 (0·252–0·280) | 0·02 (0·014–0·036) | 0·07 (0·043–0·098) | 0·06 (0·041–0·097) | 0·03 (0·022–0·055) | 0·05 (0·010–0·039) | 0·12 (0·066–0·177) | 0·13 (0·061–0·213) | 0·05 (0·025–0·094) | 0·001 (0·000–0·002) |
Ranges are given by two values (minimum–maximum). Where there is a single value, that is the average, meaning that all values are equal.
The phenograms derived from Nei's and Rogers's distances showed high congruence and only the former is shown (Fig. 2; cophenetic correlation = 0·922). Two main clusters were found, one comprising A. siculum and A. dielsianum whose populations are intermingled, and the other comprising the remaining taxa. In this second cluster, A. linkianun, A. cirrhigerum and A. litigiosum are clearly defined, while populations initially ascribed to A. australe and A. tortuosum are mixed together and similarly those of A. majus and both subspecies of A. latifolium. Finally, all three populations of A. barrelieri form a cluster into that of A. majus–A. latifolium.
Fig. 2.
UPGMA clustering phenogram derived from Nei's genetic distance coefficients among the 52 populations studied. The cophenetic correlation is 0·922.
DISCUSSION
The allozyme data support the systematic subdivision of the studied taxa into two different groups, series Sicula and Majora (Rothmaler, 1956) (subsections according Fernández Casas, 1997), but our data do not support the arrangement of species within these groups, with A. siculum, A. dielsianum and A. barrelieri into series Sicula and the remaining taxa into series Majora (Table 1). The mean distance between A. siculum and A. dielsianum to A. barrelieri is much higher than those between A. barrelieri and any other taxa of series Majora, indicating that A. barrelieri is closer to these taxa and, consequently, should be included into series Majora instead of seriesSicula.
The genetic distances between populations of A. siculum and A. dielsianum supports Webb's (1971) and Sutton's (1988) opinion of recognizing A. siculum as a single species including A. dielsianum as a synonym, in contrast to Rothmaler (1944, 1956) and Fernández-Casas (1997). Conversely, the genetic distances between the subspecies of A. majus (Sutton, 1988) (Table 4, Fig. 2), allow us to consider all of them at specific rank, as originally described for most of them (see Rothmaler, 1956; Sutton, 1988), A. linkianum and A. cirrhigerum being clearly separated from any other in series Majora.
Webb (1971) considered A. litigiosum as conspecific to A. barrelieri. However, for most authors (Rothmaler, 1956; Sutton, 1988; Fernández-Casas, 1997) both species are different, and belong to different series, with the former included as a subspecies into A. majus as explained earlier (Rothmaler, 1956; Sutton, 1988; Fernández-Casas, 1997). Our data support the recognition of both taxa at the specific level and, interestingly, show a closer relationship of A. barrelieri to A. majus than to A. litigiosum.
After the description of A. australe by Rothmaler (1956), who included it into series Hispanica, authors dealing with the genus Antirrhinum have agreed both in its specific status (Webb, 1971; Sutton, 1988; Fernández-Casas, 1997) as well as its close affinity to A. graniticum (Fernández-Casas, 1997). However, allozyme data suggest that populations of A. australe and A. tortuosum are mixed together. Together with the morphological similarity between both species and their sympatric ranges, it is impossible to distinguish among them even as subspecies or varieties, allowing us to consider them as a single species, A. tortuosum.
Populations of A. majus and both subspecies of A. latifolium also cannot be distinguished. The ranges of these taxa are parapatric and they are morphologically quite similar, with the colour of flowers being the main difference between them (purple in A. majus, yellow in A. latifolium). Our data do not support the separation of subspecies within A. latifolium, leading us to consider A. latifolium subsp. intermedium sensu Sutton (1988) as a synonym of A. latifolium.
Despite interfertility (Rothmaler, 1956), the rarity of mixed populations and the ability of pollinators to discriminate between the different species (Mather, 1947) leads to a gradual divergence among species, as was found in other species of Antirrhinum (Mateu-Andrés and Segarra-Moragues, 2003b). This fits with a geographical model of speciation, as previously observed in other group of species of Antirrhinum (Mateu-Andrés and Segarra-Moragues, 2003b), and proposed in Galvezia (Elisens, 1992) and Linaria (Valdés, 1970).
TAXONOMIC SYNTHESIS FOR THE STUDIED TAXA
Subsection Antirrhinum
Series Sicula Rothm.
A. siculum Miller. Gard. Dict., ed 8 (1768).
Synonyms: A. siculum var. glandulosum Chav. Monogr. Antirrh. (1833). A. dielsianum Rothm. Feddes Reppertorium 54 (1944).
Series Majora Rothm
A. majus L. Sp. Pl. ed 1 2: 617 (1753).
A. latifolium Miller Gard. Dict. Ed 8: no. 4 (1768).
Synonyms: A. intermedium Debeaux Bull. Soc. Bot. Fr.20 (1873). A. latifolium ssp intermedium (Debeaux) Nyman, Consnp. 3: 537 (1881). A. majus ssp. latifolium (Mill.) Rouy, in Rouy et Fouc., Fl. Fr. XI (1909). A. majus ssp. majus var. striatus (D.C.) Rothm., Feddes Reppert. (Beih.) 136 (1956).
A. tortuosum Bosc., in Bosc. Ex Lamk., Encycl. 4 (1797).
Synonyms: A. australe Rothm., Feddes Reppert. (Beih.)136 (1956). A. majus ssp. tortuosum (Bosc.) Rouy, Fl. Fr. 11 (1909).
A. linkianum Boiss. et Reut., Diagn. Pl. 2 (1856).
Synonyms: A. majus ssp. linkianum (Boiss. et Reut.) Rothm., Feddes Reppert 54 (1944).
A. cirrhigerum Welw. ex Rothm., Bol. Soc. Brot. 13 (1939).
Basionym: A. latifolium var. cirrhigerum Ficalho, Apont. (1880).
Synonyms: A. linkianum var. ramosissimum Wk. in Willk et Lge., Prodr. Fl. Hisp. 2 (1870). A. majus ssp. cirrhigerum (Ficalho) Franco in Heywood, Bot. Jour. Linn. Soc. 64(3) (1971).
A. litigiosum Pau, Not. Bot. VI (1895).
Synonyms: A. majus ssp. litigiosum (Pau) Rothm., Feddes Reppert. (Beih.)136 (1956).
A. barrelieri Boreau, Cat. Graines Rec. Jard. Bot. Angers 1854.
Synonyms: A. controversum Pau, Not. Bot VI (1892).
Acknowledgments
We are indebted to Dr M. Iberite who provided us with seeds of several Italian populations of A. majus, A. siculum and A. tortuosum, and Dr A. Dafni who provided seeds from the Haifa population of A. siculum.
LITERATURE CITED
- Crawford DJ. 1985. Electrophoretic data and plant speciation. Systematic Botany 10: 405–416. [Google Scholar]
- Crawford DJ. 1990.Plant molecular systematics: Macromolecular approaches. New York: J. Wiley and Sons. [Google Scholar]
- East EM. 1940. The distribution of self-sterility in the flowering plants. Proceedings of the American Philosophical Society 82: 449–517. [Google Scholar]
- Elisens WJ. 1992. Genetic divergence in Galvezia (Scrophulariaceae): Evolutionary and biogeographic relationships among south american and Galapagos species. American Journal of Botany 79: 198–206. [Google Scholar]
- Elisens WJ, Crawford DJ. 1988. Genetic variation and differentiation in the genus Mabrya (Scrophulariaceae-Antirrhineae): systematic and evolutionary inferences. American Journal of Botany 75: 85–96. [Google Scholar]
- Elisens WJ, Nelson AD. 1993. Morphological and isozyme divergence in Gambelia (Scrophulariaceae): species delimitation and biogeographic relationships. Systematic Botany 18: 454–468. [Google Scholar]
- Fernández Casas J. 1997. De Antirrhinis notulae. Fontqueria 48: 195–202. [Google Scholar]
- Gottlieb LD. 1984. Isozyme evidence and problem solving in plant systematics. In: Grant WF, ed. Plant biosystematics. Orlando: Academic Press, 343–357. [Google Scholar]
- Gruber F. 1930. Über Selbststerilität und Selbstfertilität be Antirrhinum. PhD Thesis. Berlin. [Google Scholar]
- Gruber F. 1932. Über die Verträglichkeitsverhältnisse bei einigen selbststerilen Wildsippen von Antirrhinum und über eine selbstfertile Mutante. Zeitschrift für Indukt. Abstammungs-Vererbungsl. 62: 426–462. [Google Scholar]
- Mateu-Andrés I. 1999. Allozymic variation in three species of Antirrhinum L. Scrophulariaceae-Antirrhineae) and divergence among them. Botanical Journal of the Linnean Society 131: 187–199. [Google Scholar]
- Mateu-Andrés I, Segarra-Moragues JG. 2000. Population subdivision and genetic diversity in two narrow endemics of Antirrhinum L. Molecular Ecology 9: 2081–2087. [DOI] [PubMed] [Google Scholar]
- Mateu-Andrés I, Segarra-Moragues JG. 2003. Patterns of genetic diversity in related taxa of Antirrhinum L. assessed using allozymes. Biological Journal of the Linnean Society 79: 299–308. [Google Scholar]
- Mateu-Andrés I, Segarra-Moragues JG. 2003. Allozymic differentiation of the Antirrhinum graniticum and the Antirrhinum meonanthum species groups. Annals of Botany 92: 647–655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mather K. 1947. Species crosses in Antirrhinum I. Genetic isolation of the species majus, glutinosum and orontium Heredity 1: 175–186. [Google Scholar]
- Nei M. 1972. Genetic distance between populations. American Naturalist 106: 283–292. [Google Scholar]
- Oyama RK, Baum DA. 2004. Phylogenetic relationships of North American Antirrhinum (Veroniceae). American Journal of Botany 91: 918–925. [DOI] [PubMed] [Google Scholar]
- Rothmaler W. 1944. Systematische Einheiten in der Botanik. Feddes Reppert. 54: 1–22. [Google Scholar]
- Rothmaler W. 1956. Taxonomische Monographie der Gattung Antirrhinum Feddes Repertorium (Beitraege) 136: 1–134. [Google Scholar]
- Sutton DA. 1988.A revision of the tribe Antirrhineae. London and Oxford: Oxford University Press, 67–97. [Google Scholar]
- Swofford DL, Selander RB. 1989.Biosys-1. A computer program for the analysis of genetic variation in population genetic and biochemical systematics, Version 1·7. Champaign: Illinois Natural History Survey. [Google Scholar]
- Valdés B. 1970. Revisión de las especies europeas de Linaria con semillas aladas. Anales de la Universidad Hispalense. Publicaciones de la Universidad de Sevilla No. 7. [Google Scholar]
- Webb DA. 1971. Taxonomic notes on Antirrhinum L. Flora Europaea Notulae Systematicae 10 64: 271–275. [Google Scholar]
- Wendel JF, Weeden NF. 1989. Visualization and interpretation of plant isozymes. In: Soltis DE, Soltis PS, eds. Isozymes in plant biology. Portland, OR: Dioscorides Press, 5–45. [Google Scholar]
- Wright S. 1978.Evolution and the genetics of populations, Vol. 4 Variability within and among natural populations. Chicago: University of Chicago Press. [Google Scholar]