Abstract Genotype-environment interaction (GEI) intro- duces inconsistency in the relative rating of genotypes across environments and plays a key role in formulating strategies for crop improvement. GEI can be either quali- tative (i.e., crossover type) or only quantitative (i.e., non- crossover type). Since the presence of crossover-type in- teraction has a strong implication for breeding for specif- ic adaptation, it is important to assess the frequency of crossover interactions. This paper presents a test for de- tecting the presence of crossover-type interaction using the response-environment relationship and enumerates the frequency of crossovers and estimation of the cross- over point (CP) on the environment axis, which serves as a cut-off point for the two environments groups where different/specific selections can be made. Sixty-four bar- ley lines with various selection histories were grown in northern Syria and Lebanon giving a total of 21 environ- ments (location-year combinations). Linear regression of the genotypic response on the environmental index repre- sented a satisfactory model, and heterogeneity among re- gressions was significant. At a 5% level of significance, 38% and 19% of the pairs showed crossover interactions when the error variances were considered heterogeneous and homogeneous, respectively, implying that an appre- ciable number of crossovers took place in the case of bar- ley lines responding to their environments. The CP of 1.64 t/ha, obtained as the CP of regression lines between the genotype numbers 19 and 31, provided maximum genotype x environment-group interaction. Across all en- vironments, genotype nos. 59 and 12 stood first and sec- ond for high yield, respectively. The changes in the ranks of genotypes under the groups of environments can be used for selecting specifically adapted genotypes. Key words Crossover point · Genotype x environment interaction · Crossover genotype-environment interaction · Linear regression model · Barley Introduction Genotype-environment interaction (GEI) plays a key role in developing strategies for crop improvement. Numer- ous methods for exploiting GEI have been developed in the literature (Yates and Cochran 1938; Finlay and Wil- kinson 1963; Eberhart and Russell 1966; Byth et al. 1976; Gauch 1988; Singh et al. 1996b), among others) and reviewed (Freeman 1973; Lin et al. 1986; Westcott 1986; Kang 1990). In general, the selection of genotypes is based on their evaluation in a number of environ- ments. The relative response of the genotypes varies over the environment, indicating a change in the superi- ority of one genotype over the others with respect to the environment, including a change in the rank of the geno- types. Selection of genotypes with an objective of yield maximization in the case of rank changes over environ- ments is complicated (Haldane 1947) due to non-separa- bility of response behavior (Gregorius and Namkoong 1986). In clinical trial situations, Peto (1982) distin- guished qualitative interaction, where the direction of true treatment (genotype in the present case) differences varies among subsets of patients (environments in the present case) from quantitative interactions where treat- ment differences vary only in magnitude, but not in di- rection (i.e., in rank). Gail and Simon (1985) introduced the term crossover interaction for qualitative interaction and non-crossover interaction for quantitative interac- tion, and provided a test for crossover interaction be- tween clinical treatments and patients. The “crossover” here is different from the crossover term used in genetics (e.g., Sinnot et al. 1958). Baker (1988) applied the tests given by Azzalini and Cox (1984) and Gail and Simon (1985) on spring-wheat data for testing crossover geno- type-environment interaction (COGEI). The consequenc- es of COGEI on breeding strategies have been discussed Communicated by H. Becker M. Singh ( ✉ ) International Center for Agricultural Research in the Dry Areas, P.O. Box 5466, Aleppo, Syria e-mail: m.singh@cgiar.org Tel.: +963-21-2213477 Fax: +963-21-2213490 Theor Appl Genet (1999) 99:988–995 © Springer-Verlag 1999 ORIGINAL ARTICLE M. Singh · S. Ceccarelli · S. Grando Genotype x environment interaction of crossover type: detecting its presence and estimating the crossover point Received: 25 January 1999 / Accepted: 16 March 1999