Seeds of the common bean, Phaseolus vulgaris, contain two inhibitors of mammalian and insect α-amylases (αAls) that show specificity towards the amylases of different insect species. Expression in pea (Pisum sativum) and azuki bean (Vigna angularis) of a chimeric gene consisting of the cDNA of bean αAl-1 and a seed-specific promoter makes the seeds of these legumes resistant to three species of Old World bruchids whose amylases are inhibited by αAl-1. This was the first successful genetic engineering of insect resistance in seeds. To understand the basis of the specificity between amylases and inhibitors we cloned a second bean inhibitor (αAl-2) with different specificity, and we cloned the cDNA of the New World bruchid, Zabrotes subfasciatus. The amylase of this bruchid is inhibited by αAl-2, but not by αAl-1. Knowledge of the amino acid sequences and of the three-dimensional structure of the pancreatic α-amylase–αAl-1 complex allows us to predict the peptide domains and amino acids of the proteins that are important for protein–protein recognition and inhibition of enzyme activity.

Growers expect high-quality, genetically pure seed. As a result, seed companies maintain quality control programmes that monitor seed from harvest to purchase. An array of ‘traditional’ seed quality tests, including mechanical tests and tests of genetic purity, seed germination and vigour, and seed health tests, is used, and seed quality assessment techniques continue to evolve. Advances in molecular genetics are allowing the release of new varieties differing essentially in one gene. New molecular biology approaches offer the potential to identify these subtle genetic differences. Advances in seed enhancements, such as pelleting, priming and pregermination, require increased scrutiny of seed quality before and after the enhancement process. New developments in computer imaging for improved purity and germination/vigour analyses are being developed. These novel approaches to seed quality assessment become important as new genetic improvements are conveyed in the seed at increased cost to the grower.

New techniques for the accurate and feasible detection of many seedborne pathogens have been developed over the last 25 years. These include conducive environmental incubation and identification (many seedborne fungi), liquid plating assay (seedborne bacteria), enzyme-linked immunosorbent assay (seedborne viruses) and polymerase chain reaction (PCR) technology. Regardless of the detection methodology, the specificity, sensitivity, reliability, efficiency of the assay and an understanding of pathogen tolerance in a seedlot need to be considered before a technique is an acceptable clinical seed health test. Acceptable seed health tests are tools for disease risk-management and routinely used in seed quality assessment. Seed health issues are increasingly important in international seed trade. With the advent of free trade, many countries are redefining their phytosanitary requirements with the goal of preventing introduction of a devastating pathogen into their country. However, many phytosanitary regulations are implemented without fully understanding the economic threat of a pathogen or a complete scientific analysis of a pathogen's presence in that country. This has led to confusing regulations, unnecessary tests or inspection requirements and unjustified trade barriers. To provide a means to answer scientifically the problems encountered in the worldwide movement of seed, an international movement has emerged to standardize seed health tests and inspection practices for international seed trade. The impact of standardization on the free trade of seed and the goals of the International Seed Health Initiative (ISHI) will be discussed.

A high level of genetic purity in crop varieties must be achieved and maintained for agronomic performance as well as to encourage investment and innovation in plant breeding and to ensure that the improvements in productivity and quality imparted by breeders are delivered to the farmer and, ultimately, to the consumer. Traditionally, morphological comparisons have formed the basis for genetic purity evaluations. However, replicated field observations are time-consuming, expensive and unreliable. Morphology cannot provide information on the purity of specific genetic attributes that relate to grain quality or to pest or herbicide resistance bred into varieties. Biochemical assays, including isozymes, can distinguish varieties within several species. Isozymes have been routinely used in checking seed-lot purity in maize (Zea mays L.) for the past 20 years. Newer DNA-based technologies such as restriction fragment length polymorphisms and more recently developed methods that use the polymerase chain reaction can allow even more discriminative and faster identification of varieties. However, none of the DNA methods have replaced biochemical methods for seed purity assays, other than in a relatively select group of crops with very high seed value, due to their high datapoint cost. It will require further miniaturization, automation and enhanced capabilities to process numerous samples simultaneously before newly developed methods supplant biochemical methods for routine usage in purity testing. New varieties that have major genes for herbicide or insect resistance incorporated within them require purity assays during product development and following seed production of the commercial variety. Immunological or DNA sequence assays can be developed and automated systems are required to process hundreds of thousands of individuals. Ultra-high, micro-array technologies and single-molecule detection systems are now under development. These technologies offer the promise that adequate distinction and high sample throughput will be combined. New methods may eclipse the capabilities of biochemical methodologies, thereby potentially raising genetic purity standards and enabling farmers and consumers better to utilize and benefit from increasingly productive varieties that are bred from a more diverse base of genetic resources.

The literature suggests that the establishment of seed yield potential in cool-season perennial grasses is based on developmental processes that take place before floral induction. The relationship between vegetative plant development to flowering and seed yield can be influenced by numerous factors, such as genotype, stand age, post-harvest regrowth environment, pests, management practices and juvenility. The objectives of our investigations and survey of literature were to identify characteristics of vegetative plant development present before floral induction (autumn regrowth period) that were related to flowering (yield potential) and seed yield (utilization of seed yield potential). The number of fertile tillers present at anthesis is a function of the number and developmental state (or biological age) of vegetative tillers present prior to floral induction. The number and size of vegetative tillers before induction was correlated with flowering and seed yield in young stands of orchardgrass (Dactylis glomerata L.) and tall fescue (Festuca arundinacea Schreb.) (r = 0.93 and 0.95 for first-year stands of orchardgrass and tall fescue, respectively), but older stands did not exhibit this relationship. The basal diameter of vegetative tillers in autumn was related to flowering and seed yield in young stands of Kentucky bluegrass (Poa pratensis L.) and creeping red fescue (F. rubra L) As stands aged, basal diameter was not a primary indicator of subsequent flowering and seed yield. Tiller height at the end of the regrowth period was consistently related to flowering and seed yield in Kentucky bluegrass(r = −0.81, P < 0.01) and creeping red fescue (r = −0.87, P < 0.01), regardless of stand age. Increasing the rhizome:root ratio caused reduction of flowering in creeping red fescue. Vegetative characteristics were not related to flowering and seed production in perennial ryegrass (Lolium perenne L.) and dryland bentgrass (Agrostis castellana Boiss. and Reut.). Green leaf number was a poor indicator of flowering and seed yield potential in all species tested. Ethylene production during autumn vegetative development has been implicated as a potential cause of reduced flowering in creeping red fescue. As much as 92% of the seed yield potential in several cool-season perennial grasses is set before the onset of conditions favouring vernalization of the crop. An increased understanding of developmental gateways in the formation of flowering and seed yield potential may be key to enhancing the economic and environmental sustainability of seed production enterprises.