The discovery of herbicidal properties for 2,4-D, MCPA, 2,4,5-T (see Table 1 for chemical names), and related chemicals in the early 1940s 1 is still possibly the most significant breakthrough in the field of chemical weed control. One of the main reasons for the continued significance of these chemicals is their selectivity for controlling broadleaf weeds in grasses or unwanted hardwoods within conifers at relatively low rates of application. These discoveries were some of the most significant "spin-offs" from the early research on plant growth regulators. During the three decades since these discoveries, we have witnessed the development of more than 100 organic chemicals as selective herbicides. The resulting gains in crop yields have been immense. During the peak of this era in the 1960s, many agriculturalists may have believed that it was only a matter of time before the perfectly selective herbicide was developed for every major weed problem in every major crop. However, our experiences during the 1970s proved that this was false optimism. First of all, we have not yet discovered the perfectly selective herbicide and we probably can't expect to do so. Repeated use of even the most effective herbicides will normally cause shifts within weed populations to weed species nearly as tolerant to the herbicide as the crops themselves. In some cases resistant biotypes of normally sensitive weed species have become problems after repeated use of particular herbicides. 2 Furthermore, we can't expect to keep solving new weed problems with new herbicides. With the increased development costs to document the toxicological and environmental safety of new chemicals, it is often wiser to make better use of the chemicals that we already have. We can often achieve broader spectrum weed control by employing combinations of two or more herbicides as long as the crop can tolerate all herbicides involved. This has been a common approach but it is still difficult to control weeds that are botanically related and physiologically similar to the crops. Thus, we eventually end up with the same problem — the need for greater herbicide selectivity.