Page 2 - Domina_alii_2017

ISRAEL JOURNAL OF PLANT SCIENCES 49

deliberately omitted from it the morphologically unfit- Table 1. Source of sampled populations (taxonomy following
Domina et al. 2016).
ting C. parlatoris. The complex in that cladogram
Elevation (m Coordinates (WGS84,
appears as sister to a clade of three taxa: C. aeolica Taxon and populations a.s.l.) decimal degrees)
Lojac. subsp. aeolica (Eolian Islands), C. aeolica subsp. Centaurea busambarensis
1. Busambra 1300 37.854759 N, 13.415954 E

pandataria (Fiori & B eg.) Anzal. (Ventotene, Pontian

2. Kumeta 930 37.970513 N, 13.256235 E

Islands of Latium), and C. papposa (Cape Bon in Tuni- 3. Pizzuta 1030 37.997086 N, 13.246412 E

4. Isnello, al castello 570 37.945732 N, 14.005906 E

sia). Together, these two clades are nested within C.
cineraria L. (coasts of Latium and Basilicata, peninsular C. erycina
5. Erice 720 38.035189 N, 12.591671 E

Italy).
However, the recent survey of the “Centaurea C. panormitana subsp.
panormitana
group” (C. sect. Centaurea)byHilpold et al.(2014), 6. Punta Mastrangelo 850 38.064298 N, 13.243363 E

7. Pizzolungo 40 38.063512 N, 12.570256 E

based on DNA sequence data and a much extended

8. Monte Pellegrino, S 150 38.153465 N, 13.360442 E
sampling of taxa from other areas, resulted in a poorly slope

9. Monte Pecoraro 750 38.162269 N, 13.123313 E
resolved tree that partly contradicts the earlier phylo- 10. Favignana, Monte 100 37.920730 N, 12.307299 E

genetic hypothesis. At present, relationships within C. Santa Caterina
sect. Centaurea are better considered as poorly under- C. panormitana subsp.
stood. The molecular data, while supporting the umbrosa
11. Monte Gallo 500 38.218151 N, 13.311306 E

assumption of widespread hybridization, correlate bet- 12. Monte Pellegrino, N 200 38.187822 N, 13.335711 E

ter with geographical distribution patterns than with slope

13. Monte Grifone 250 38.071647 N, 13.370249 E

morphology. These findings support our approach of
C. panormitana subsp.
considering, in the first place, diversity and diversifica-
todaroi
tion within geographically defined boundaries. 14. Mongerbino 30 38.110688 N, 13.538341 E

15. Monte Catalfano 100 38.114900 N, 13.513798 E

The Centaurea busambarensis complex on the
whole shows pronounced character variation, both C. panormitana subsp.
seguenzae
within and between populations, so that the delimita- 16. Capo Tindari 20 38.137394 N, 15.052898 E

tion of its constituent taxa is often uncertain (Giardina
C. saccensis
et al. 2007). We therefore undertook to complement 17. Gole della Tardara 130 37.614346 N, 13.052547 E

the studies done up to now (Raimondo & Bancheva
2004; Raimondo et al. 2004; Palla et al. 2005; Bancheva conditions, to observe intra-populational variability and
et al. 2006; Geraci et al. 2007) by sampling wild popu- interactions between morphology and environment.
lations and using statistical methods to analyze their Eleven populations of taxa attributed to the four
morphological variation patterns, so as to clarify their subspecies of Centaurea panormitana were compared
taxonomic status. In the process, we paid particular with six populations of the other taxa of the C. busam-
attention to plants from the island of Favignana (Egadi barensis complex (see Table 1 and Figure 1).
Archipelago, W Sicily), which show pronounced mor- As recommended by Guarino and Rampone (2006),
phological differences when compared with popula- we used the original descriptions of the taxa involved
tions of the other taxa described and named so far. to draw up a list of characters of potential diagnostic
usefulness. A total of 26 characters were used: 19 con-
Material and methods tinuous quantitative, four discrete quantitative and
three qualitative.
The specimens used for statistical analysis were newly
collected throughout the geographical range of the
Nineteen continuous quantitative characters (mm)
Centaurea busambarensis complex. For each popula-
tion, 10 measurements were taken for each quantita- 1. Whole plant, height
tive character from each of at least 10 different plants 2. Rosette leaves, length
(see Electronic Supplement 1, supplemental material). 3. Rosette leaves, width
Voucher specimens were deposited in PAL and FI. Her- 4. Lower stem leaves, length
barium studies were complemented with observation 5. Lower stem leaves, width
in the field and cultivation in uniform environmental 6. Upper stem leaves, length