Table 1. Summary of the experimental protocols that enable to account for allelic differences at loci other than those conferring resistance among compared resistant and susceptible plant genotypes.

Method	Direct methods							Indirect methods
	Single population	Pedigreed lines	Segregating population	Multiple populations	Segregating cross	NILs	Transgenic lines	Multigenerational	Cline
(a) Prevalencea	4.5	1.0	18.2	18.2	2.0	4.3	2.0	7.5	1.0
(b) Advantages/disadvantages
Mechanistic basesb	+	+	+	+	+	+	+	−	−
Fitness costc	−	−	−	−	−	−	−	+	+
Time consumption (generation)	2	2	2–3	0	2–3	6–7	3–4d	5–6	1
Genetic background controle	+	+	+	++	++	+++	+++	f	f
Assumptions	No sub-population structure	Quantitative traits only		No population structure
Genotyping and phenotyping costh	+	+g	+	+	+	+	+	−	−
(c) Recommendations
Model organism							x		x
Selfing species		x	x		x				x
Outcrossing species	x	?	x		x				x

Abbreviation: NILs, near-isogenic lines.

A number of advantages, disadvantages and recommendations are described for each protocol (for more details see text).

^aUse (%) of experimental designs in published studies (n=93) assessing pleiotropic effects associated with herbicide resistance alleles.

^b‘+' and ‘−' stand for methods that provide or not an understanding of the particular trait involved in the expression of cost, respectively.

^c‘+' and ‘−' stand for methods that evaluate or not the impact of costs on the evolutionary trajectories of resistance allele frequencies, respectively.

^dIt will depend on the reproductive system of the species.

^eMethods that examine costs within a random homogenous genetic background (+), provide an understanding of the effects of different genetic backgrounds on costs (++) and eliminate the effects of genetic background (+++).

^fThe genetic background control for the indirect methods depends on the plant material used.

^gDifficult to determine resistance genotype.

^h‘+' and ‘−' stand for low and high genotyping and phenotyping cost, respectively.