Silicon is present in plants in amounts equivalent to those of such macronutrient elements as calcium, magnesium, and phosphorus, and in grasses often at higher levels than any other inorganic constituent. Yet except for certain algae, including prominently the diatoms, and the Equisetaceae (horsetails or scouring rushes), it is not considered an essential element for plants. As a result it is routinely omitted from formulations of culture solutions and considered a nonentity in much of plant physiological research. But silicon-deprived plants grown in conventional nutrient solutions to which silicon has not been added are in many ways experimental artifacts. They are often structurally weaker than silicon-replete plants, abnormal in growth, development, viability, and reproduction, more susceptible to such abiotic stresses as metal toxicities, and easier prey to disease organisms and to herbivores ranging from phytophagous insects to mammals. Many of these same conditions afflict plants in silicon-poor soils-and there are such. Taken together, the evidence is overwhelming that silicon should be included among the elements having a major bearing on plant life.

Silicon is the second most abundant element in soils, the mineral substrate for most of the world's plant life.

The title of this essay declares that silicon does have roles in plants and all participants in this conference know that that is so. This knowledge, however, is not shared by the general community of plant biologists, who largely ignore the element. This baffling contrast is based on two sets of experience. First, higher plants can grow to maturity in nutrient solutions formulated without silicon. That has led to the conventional wisdom that silicon is not an essential element, or nutrient, and thus can be disregarded. Second, the world's plants do not grow in the benign environment of solution culture in plant biological research establishments. They grow in the field, under conditions that are often anything but benign. It is there, in the real world with its manifold stressful features, that the silicon status of plants can make a huge difference in their performance. The stresses that silicon alleviates range all the way from biotic, including diseases and pests, to abiotic such as gravity and metal toxicities. Silicon performs its functions in two ways: by the polymerization of silicic acid leading to the formation of solid amorphous, hydrated silica, and by being instrumental in the formation of organic defence compounds through alteration of gene expression. The silicon nutrition of plants is not only scientifically intriguing but also important in a world where more food will have to be wrung from a finite area of land, for that will put crops under stress.

earlier references are listed there. Also in the transport problem, in Momentum Transfer, Hre. is negligible; see Parikh, J.,4 Appendix B. 6 Allowing A to depend on w2 [see Bardeen, J., Rev. Mod. Phys., 34, 667 (1962), Appendix BI does not affect the main part of our argument, concerned with terms linear in w. 7 Note that, contrary to our notation, the momentum of a quasiparticle of "type 1" (spin-) is called-k in ref. 4. 8 The condition owkF << 1 in equation (16) sets an upper bound to Jv| but still admits fairly large values of In 1vi (.10, say).

Increasing salinity of soil and water threatens agriculture in arid and semiarid regions. By itself, the traditional engineering approach to the problem is no longer adequate. Genetic science offers the possibility of developing salt-tolerant crops, which, in conjunction with environmental manipulation, could improve agricultural production in saline regions and extend agriculture to previously unsuited regions.