Spatial distribution of flower colour polymorphism in Iris lutescens

ABSTRACT

Iris lutescensis a common species occurring mainly in dry limestone habitats in Western Italy, Southern France, and Spain. The species shows a remarkable polymorphism for flower colour, and yellow and purple flowers can be found in the same population. As the species is a deceptive one, the previous studies on the maintenance of such a polymorphism were linked to its pollination ecology. Here, I reported on the spatial distribution of the polymorphism and showed that Spanish populations are mostly purple monomorphic. In contrast, populations in the South of France and Italy show the complete range, from 0 to 1, for the frequency of yellow morph, and the spatial autocorrelation for morph frequencies is very low. To go further, correlations between morph frequencies and abiotic factors, such as bioclimatic variables, UV irradiance, and aridity, were studied. On the whole, the spatial distribution of the frequency of yellow morph can be hardly explained by the contributions of these abiotic variables, and historical contingencies, including the phylogeography of the species have to be invoked, in particular, to explain the absence of polymorphic populations in Spain.

KEYWORDS:

• Deceptive pollination
• anthocyanins
• spatial autocorrelation
• Iris

Acknowledgments

I thank P. Aymerich, A. Bonet, R. Calatyud, R. Leras, L. Minuto, J. Pedrol, G. Pellegrino, L. Serra, M. Talavera, and S. Talavera for helping in finding localities. I also thank M. Arista, P. Ortiz, M. Talavera, and H. Wang for fruitful discussions about flower colour polymorphism, and F. Rousset for his help with data analyses. Two anonymous reviewers made some suggestions that contributed to clarify this manuscript. Finally, I am particularly indebted to Béatrice Hervouet for supporting me during field works. The first draft of this manuscript was written during a sabbatical period in the department of Botany of the University of Sevilla.

Disclosure statement

No potential conflict of interest was reported by the author.

Supplementary material

Supplemental data for this article can be accessed here.