Comparison of two shrub rose genotypes using architectural analysis
Résumé
The aesthetic value of an ornamental plant is highly dependant on its shape. This is primarily determined by the characteristics of its axes at the following levels: (1) morphological (size of the different constituent elements) (2) topological (branching relationships) and (3) spatial (position of the axes in space). Architectural analysis therefore appears to be an interesting method for characterising plant shape and for assessing its aesthetic value. This method seems to be particularly well adapted to an ornamental shrub such as the rose bush that has a visibly complex architecture. The analysis took place on two cultivars of the shrub rose Rosa hybrida, Knock Out® cv Radrazz and Lovely Meilland® cv Meiratcan. Quantitative experimental data (morphological and topological) were obtained on plants grown from metamers cuttings in a greenhouse for five months. Fifteen plants per cultivar were fully described. Data was analysed using the Multi Tree Scale method (MTG) (Godin C. and Caraglio Y. 1998). This method made it possible to obtain a database for each cultivar that included the different related architectural scale levels, morphological entities and variables (attributes). By analysing this database, we were able to clearly identify the dominant architectural elements common to the two cultivars, such as the existence of two types of axes (long and short) and the position of these axes according to branching order. Moreover, it also allowed us to identify the architectural characteristics that differentiate the two genotypes, such as the number of branching orders, the number of flower-bearing branches and their length. This precise and quantitative analysis of shrub architecture is a powerful method for describing shrub shape. It may therefore be an invaluable tool for breeders, for example, who want to accurately characterise the architecture of different genotypes in order to impose selection pressure on objective criteria. The same is true for the researcher who would like to assess the effect of various treatments on plant shape, such as in the case of pruning or the application of a growth regulator.