Drought resistant varieties as an experimental tool to prove a physiological hypothesis – a frequent pitfall


It has been and still is a common approach in physiological investigation of drought resistance to use two or more crop varieties divergent in their "drought resistance". When in the experiment the two varieties also differ for a certain physiological drought-related trait then this trait is concluded to be associated with or even explain drought resistance. For example, if the two varieties differ in their root length then long roots are taken as the explanation of drought resistance in that case. Two problems are recognized here:

1.    The a priori assertion that these varieties differ in their drought resistance is sometimes not well substantiated.

2.    Other possible physiological explanations of the difference in drought resistance between the test varieties might be overlooked.


A recent example is the paper in Functional Plant Biology 33:823–837, 2006* where two wheat varieties were stated to be drought resistant (Seri82) and drought susceptible (or ‘standard’) (Hartog), respectively. Data collected in root boxes on these two varieties was used to explain the difference in drought resistance between them on the basis of their different root architecture. However, in this case as in others, no information or proof is given why these two varieties were designated as different in their drought resistance and no other developmental or physiological factor besides root architecture was considered as a possible explanation of drought resistance if indeed it varied between these varieties. These two varieties could differ in their drought resistance by factors not investigated in this study, such osmotic adjustment or use of stem reserves for grain filling or any other contributing trait. These two varieties could differ in their drought adaptation due to plant developmental traits such as phenology, plant size, early vigor or other traits. Roots may or may not have been the explanation of drought resistance in this case – if indeed the two varieties differed in their drought resistance. The simulation performed in this study showing that greater soil moisture extraction increases yield under drought stress does not help convince the reader that root architecture is the reason for the stated “drought adaptation” of Seri82 and consequently of “wheat” as stated in the title.

Defining drought resistance is crucial for such studies. Varieties might differ in their performance under drought stress for different of reasons and if the definition of drought resistance ignores these reasons then the use of such varieties to dissect drought resistance becomes impossible. Poor definition of drought resistance is not uncommon in such studies. For example, in some cases differing varieties for drought resistance were simply stated as such by word of mouth. In another case they were designated by the token of their yield in a certain dry year. In another case the opinion of the local breeder was cited as proof. Even when “yield performance under water-limited conditions” is cited as the basis for defining drought resistance, there is a problem. We already realize now that plant production under mild stress can sometimes be fully accounted for by the yield potential of the variety without any bearing on its drought adaptation.

Hence, when two or more varieties are stated to be different in their "drought resistance" and are being used to dissect the physiology of drought resistance, satisfactory information should be produced to substantiate the basis of their defined difference in field resistance. For cereals such as wheat, the definition is ideally based on solid tests of yield performance and yield reduction under well quantified severity and timing of stress while accounting (or normalizing) for a possible role of phenology and gross developmental traits. Therefore the choice of such varieties is crucial and it should be based not only on their performance under drought but also on their being similar (not necessarily isogenic) in terms of whole plant development. A physiological dissection of drought adaptation in such carefully chosen and well-tested varieties is then most likely to reveal relevant traits associated with crop productivity under stress.


(*) The Role of Root Architectural Traits in Adaptation of Wheat to Water-Limited Environments.

By A.M. Manschadi, J. Christopher, P. deVoil and G.L. Hammer

(Functional Plant Biology 33:823–837, 2006)