Investigating physiological and biochemical

3.1.2.5. Ranking of genotypes for salt tolerance 
Usually in conventional screening methods, genotypes are screened on the basis 
of observations made for a single physiological or growth characteristics to allocate 
comparisons. The intervals are set on the basis of observation ranges. This method is 
often erroneous specifically when multiple growth and physiological characteristics of 
large number of genotypes are screened. In cluster group analysis, genotypes can be 
screened simultaneously on several physiological and ionic parameters. No score 
boundaries are set as genotypes are classified on the basis of cluster group rankings. 
Genotypes are ranked and grouped for their salinity tolerance. Ranking numbers are 
assigned to groups on the basis of cluster means. Genotypes are scored on the basis of 
ranking numbers (Khrais et al., 1998).
Salt tolerance indices were formed from the data for cluster analysis (Zeng et al., 
2002). Salt tolerance index was the observation under salinity divided by the mean of 
control. In cluster analysis, methods of Jolliffe et al., (1989) and Khrais et al., (1998) 
were followed. Group ranking was obtained on the basis of Ward’s minimum 
variance cluster analysis for salt tolerance indices of four growth parameters at the 
seedling stage, i.e., root length, root dry weight, shoot dry weight and shoot/root ratio, 
and three ionic parameters, i.e., Na
+
, K
+
and Na
+
/K
+
ratio of plant leaves. The 
procedures are described in the SAS User’s Guide (SAS Institute, 1994). Cluster 

---

56 
groups were ranked from cluster means, in order from highest to lowest means. All 
the numbers of group ranking at each salinity level in each genotype were added to 
find the sum. 
The final ranking of genotypes was done on the basis of these sums in the order 
that the genotypes with lowest sums were considered most salt tolerant and those with 
highest sums were considered salt sensitive when grown under salinity stress 
(El-Hendawy, 2011).