16 I EUROPEAN SEED I EUROPEAN-SEED.COM would reduce the number of live samples being shared around the globe. This will not completely solve the phytosanitary or logistical issues, but could reduce it to only the most necessary varieties. The issue of obtaining reference material that is of a comparable growth stage with the submitted candidate material should, in theory, be simple to resolve using DNA data as long as there is a good understanding of the species under consider- ation. Some crops can be sampled from any plant part, at any stage in the life cycle without affecting the genotyping results. However, species such as some conifers can have a different genotype depending on which part of the tree is sampled. This is also true for the morphology: taking a cutting from the top of the tree can give upright growth, whereas a cutting from the side branches tends to give prostrate growth. Therefore, as with the current DUS testing system, it is essential that there is a good understanding of the crop concerned. A DNA database is only as good as the data involved. ANY OTHER APPLICATIONS? Another function of molecular techniques in DUS testing could be to address issues relating to the set up of particular tests e.g. tests to evaluate one or two characteristics. Examples are tests for seasonal type or disease resistance which are costly to set up and time consuming to record. Where an identifiable molecular marker exists, a screening system can be used to evaluate these characteristics in a more efficient way. However, one of the limitations of a molecular test for replacing a phenotypic characteristic assessment is the current need to test for uniformity which requires the examination of many individuals. This means that a method is required for a single plant or part of a plant to be assessed to give a measure of uniformity. Looking at individuals increases the cost of this work and therefore reduces the benefits. The use of bulk samples is currently being addressed within UPOV to find ways round this current potential bottle neck SURELY THERE ARE LIMITATIONS? The overall attitude to using molecular techniques in DUS test- ing is positive, so there must be reasons as to why it has not yet been fully incorporated into the system. The development of a molecular system for a species which has relatively low international importance is often not a priority. It is easy to understand that more research (out with the DUS system) has been done on species like maize than for species like Kniphofia. It is clearly more cost effect from a DUS examination office point of view to develop a system using publically available markers than to start developing new ones. A common concern is the question of ‘how different is dif- ferent?’ There are cases where two varieties appear very simi- lar when grown out, but have very different quality traits that are easily recognised by molecular markers. For DUS testing, agronomic qualities are not considered under UPOV guidelines because of the genotype x environment interaction. Perhaps genetic markers could open up a host of quality traits that have previously been ignored because of the effect of the environment. On the other hand, in the case of mutations, varieties are clearly distinct morphologically but would appear very similar genetically. If these varieties were screened using molecular markers, they would always be included in a growing trial. It is understood that some outcrossing varieties have a wide genetic variation within a variety (e.g. in some Lolium sp.), which makes developing a molecular system for these species more difficult. All of these limitations can be minimised by knowing the physiology of the crop in question and adapting the system accordingly. A crop expert’s judgement would still be relied upon. The development of universally accepted standardised test methods is also an obstacle at present. The granting of Plant Breeders' Rights is not the end of the story. The final user of the DUS report should be considered. This is particularly impor- tant in agricultural crops where the official description produced during the DUS test is used for seed certification. Inspectors are required to ensure the crop being inspected does not differ from the description of the variety and to check the purity. Unless there is a major shift in seed certification policy, a vari- ety description will continue to play an important role in this process. FINAL THOUGHTS Research into molecular techniques continues at an incredible rate. Progress has been made in knowledge (and limitations) of markers, but also in methodology. Throughput of enough samples is no longer a problem; in fact, data sets have become so large that the analysis is the challenge. The use of molec- ular techniques in DUS testing is becoming more accessible. UPOV have had a working group for BMT issues for some time, however growing interest and the international importance of this work has been addressed by increasing the frequency of the meetings to encourage further sharing of information. At a European level, the CPVO has set up the IMMODUS working group to encourage the integration of biomolecular techniques into variety testing. UPOV, OECD and ISTA are all examining and co-operating on the possibilities for the use of biomolecular tests. The key to achieving progress in this area will be collabo- ration. We need to move from research to routine. Editor’s Note: Margaret Wallace is Technical Manager Agricultural Crops Characterisation at the National Institute of Agricultural Botany (NIAB) in the UK