Tree diseases may be as simple to diagnose as glancing at a common species with a common problem. I call this disease identification. But much more frequently disease diagnosis is far more complicated and requires a process that is far removed from simple identification. In fact, the training a plant pathologist receives is only a little short of that required of a medical doctor, and the process of plant diagnosis is similar. As in medicine, plant disease diagnosis is not an exact science, and there can be considerable uncertainty accompanying difficult cases. I tell students in my classes that any diagnostician who is 100 percent confidant of his diagnoses doesn?EUR??,,????'???t know enough to be uncertain. I strive for complete confidence about 70 percent of the time and accept some level of uncertainty in the other 30 percent of the time. Disease diagnosis is difficult for a number of reasons. Chief among them is that the number of ways that leaves can express symptoms is limited, so that diseases with widely different causes (and consequences for the plant) may look similar. For example, marginal leaf scorch, or death of a uniform margin of a leaf can be caused by such factors as drought, root injury from construction, root rot, Xylella fastidiosa (a bacterium that causes Pierce?EUR??,,????'???s disease of grapes and scorch of many ornamentals), or possibly even a nematode. Frequently there are subtle differences in distribution or appearance of symptoms that can be key clues, but to distinguish between most diseases you need more information than you can gain just by looking at the plant. One of my friends who spent years working on cabbage black rot and who had diagnosed literally hundreds of cases of this bacterial disease told me that he never makes an official diagnosis without looking for bacteria under his microscope. Why? Because he has encountered cases with what he regarded as ?EUR??,,????'??classic?EUR??,,????'?? symptoms, yet there were no bacteria present. The disease wasn?EUR??,,????'???t black rot. I rarely feel confident of a diagnosis without some form of laboratory confirmation such as by observation of microscopic signs (an element of the pathogen) in a plant part or in culture, or by soil testing. For most landscape professionals this means that when an accurate diagnosis is important you need the help of a diagnostic lab. You can help the lab enormously by knowing what material the lab needs and how to collect and ship it so the technicians have solid material to work with. Diagnosticians have some new tools that make the job of diagnosis a bit simpler and less uncertain. These include ELISA testing, which is based on immunology, and DNA screening, but with only a small number of exceptions, these are not field techniques and require plant material to be sent to diagnostic labs for analysis. In one case I was involved in, a diagnosis literally had to wait for 10 years for development of a commercially available ELISA test for X. fastidiosa. Although the pathogen could have been cultured, it is a fairly difficult process and is not performed in most plant disease clinics. Now that the ELISA test is widely available, it takes just a day to run the samples. Tests such as this can be pricey, however, and run-of-the-mill problems may not be able to justify the expense. Diagnosis is a process with distinct steps. The answer at each step determines the next step. I begin by finding out what the healthy plant is supposed to look like. If a leaf has a bluish cast, you may suspect early symptoms of a powdery mildew, or it may be normal for a cultivar. Death of interior branches may excite customers, but is normal and expected due to shading. The next step is to analyze the pattern and distribution of symptoms. With some important exceptions (don?EUR??,,????'???t you hate that phrase), symptoms of most diseases caused by non-biological factors are very ordered and uniform in their appearance. They tend to affect the same parts of each leaf and are uniformly distributed within the tree. Symptoms of diseases caused by biological pathogens are distributed randomly. If the symptoms lead me in the direction of non-biological factors I start thinking about what could have gone wrong in the tree?EUR??,,????'???s culture and pay special attention to the plant history questionnaire for the answer. If the symptoms suggest a soil related problem, the key factors are pH, salinity and, in the western US, calcium content. Diagnoses of micronutrient deficiencies are generally based on pH, and deficiencies of N, P and K very rarely appear as a ?EUR??,,????'??clinical?EUR??,,????'?? problem. Those pesky exceptions include root diseases, which will cause uniform symptoms, and soil applied herbicides such as Prometone and Spike that will cause symptoms that appear random in the crown if they are applied non-uniformly. Additional clues can separate competing possibilities. For example, soil applied herbicides and iron deficiency both cause interveinal chlorosis and scorch, but symptoms of the herbicides, the injury developed more strongly in older leaves while iron deficiency appears in younger leaves. If the symptoms appear to be random, a biological pathogen is likely, and the next step is to identify the location where the pathogen is actually working. Leaf spot, leaf blight, powdery mildew and leaf rust diseases will be fairly obvious, but they are also the least important of tree diseases because the leaves are rapidly replaces after the disease episode is passed. Very rarely a leaf-limited problem can violate that principle such as the recent episode we in California experienced with the red gum lerp psyllid, an insect that attacks Eucalyptus camaldulensis. When the insect first appeared in the fall of 1998 no one appreciated that it could cause repeated rounds of defoliation throughout the growing season. After two years of withering attacks exhausted trees began to die. Fortunately, leaf problems rarely cause such damage. When whole twigs or branches quickly die, suspect a canker disease. Look for sections of the branch or twig where bark has been killed. The damage may be obvious including callus roles or gumming, but frequently there are no local symptoms visible on the outside of the bark. Scrape the bark starting below the dead portion of the branch with a stout knife or chisel to see if the interior bark has become darkened. If this darkened interior bark forms a ring around the branch, it is a canker and is very likely to be the cause of branch death. Lab analysis is usually needed to determine which canker disease it might be, but the lab can?EUR??,,????'???t do anything without the right plant part. If on investigation there is no indication of a canker, you may be dealing with one of the wilt diseases such as Verticillium wilt, Fusarium wilt, Dutch elm disease or oak wilt. Don?EUR??,,????'???t be misled by the name of the disease. Foliage on affected branches rarely wilts, but sudden yellowing and leaf death on individual branches are most common symptoms. In the case of wilt diseases there will be no bark symptom but the wood will be streaked and darkened throughout the affected branch. Any section of wood with the discoloration will yield a good culture if sent to the lab. In nearly every case I have encountered of individual branches dying, it was due either to a canker or a wilt disease. Although it is written in some texts that individual branches may die when individual roots are destroyed, I have not seen this in southern California where communities routinely prune roots to restore damaged pavement. Only once have I seen one-sided decline of a root pruned tree but by the second year it recovered. If the entire tree dies suddenly, look first for root disease that has crept into the root collar area and has girdled. It will look very much like a canker, but will extend up from below the soil line. If the injured area harbors white, felt-like fans of mycelium, think Armillaria root rot. If not, species of Phytophthora should be suspected. There will frequently be associated clues such as poorly draining and waterlogged soil, which is highly favorable to this fungus. Massive trunk cankers can also kill trees suddenly. The disease known as ?EUR??,,????'??sudden oak death?EUR??,,????'?? that is currently destroying native oaks in coastal areas of California is such a disease. General decline, the slow thinning and weakening of a tree, is among the hardest of conditions to diagnose. It can be caused by a wide array of biological pathogens and non-biological factors. Virus diseases are very difficult to diagnose because they do not culture and there has been little study of them in woody ornamentals. However, laboratories can runs tests for some specific virus diseases as they do for Xylella. Nematodes are usually not considered because the damage they do is usually not dramatic on woody plants. However, you should consider a soil analysis for them in cases of persistent and puzzling decline.