Floral scent is a key mediator in plant–pollinator interactions; however, little is known to what extent intraspecific scent variation is shaped by phenotypic selection, with no information yet in deceptive plants. We recorded 289 scent compounds in deceptive moth fly-pollinated Arum maculatum from various populations north vs. south of the Alps, the highest number so far reported in a single plant species. Scent and fruit set differed between regions, and some, but not all differences in scent could be explained by differential phenotypic selection in northern vs. southern populations. Our study is the first to provide evidence that phenotypic selection is involved in shaping geographic patterns of floral scent in deceptive plants. The hyperdiverse scent of A. maculatum might result from the plant’s imitation of various brood substrates of its pollinators.

Floral scent samples were collected from randomly chosen Arum maculatum individuals of a total of eleven populations (six located north of the Alps, i.e. Austria, Germany, Switzerland; five located south of the Alps, i.e. Italy) by dynamic headspace on the first day of anthesis during the female stage. Scent samples were later analysed by thermal desorption-gas chromatography/mass spectrometry (TD-GC/MS) and obtained data were handled using GCMSolution v.4.41 (Shimadzu Corporation, Kyoto, Japan).
Fruits were collected from the same individuals in summer.

A simulation study was performed to quantify impact of non-detects (0) in highly multivariate phenotypic data (scent data) on estimates of an elastic net regression

For further details, see:
Gfrerer E, Laina D, Gibernau M, Fuchs R, Happ M, Tolasch T, Trutschnig W, Hörger AC, Comes HP and Dötterl S (2021) Floral Scents of a Deceptive Plant Are Hyperdiverse and Under Population-Specific Phenotypic Selection. Front. Plant Sci.:719092. https://doi.org/10.3389/fpls.2021.719092

Data_Arum_scent_selection_Gfrerer_etal.xlsx
The excel file has a total of four data sheets. The first data sheet contains all quantitative scent data of Arum maculatum [ng per inflorescence per 5 min] for subsequent scent (and selection) analyses and the simulation study. Sheet 2 and 3 contain the z-score standardised datasets used for selection analyses with scent data only from the two most extensively sampled populations ("JOS" sheet 2, and "DAO" sheet 3), which can also be filtered from sheet one. The final sheet is a read.me sheet and gives all abbreviations and further information.

Rscript_Arum_zero_simulation_Gfrerer_etal.R
R code for simulation study to quantify impact of non-detects (zeros) on selection estimates, i.e. to assess how specificity and sensitivity change with zero-inflation for elastic net regression estimates.

read.me_Arum_Gfrerer_etal.txt
Read me file provides information for understanding and using the data.