Yogi Berra, the legendary catcher of the New York Yankees, was famous for coining bons mots that have come to be known as ‘Yogi-isms’. He was born Larry Berra, the son of Italian immigrants, and got his nickname because a colleague thought that a yogi he had seen in India walked like Berra. Hanna and Barbera's Yogi Bear was named after him. One of Berra's sayings was that you can observe a lot by watching, which is one of the benefits of writing the Editor's View. Reading my way carefully through each issue of the Journal helps to keep me up to date with recent developments. But the task of comprehensively keeping up to date is an impossible one. As a clinical pharmacologist my regular reading should include at least the British Journal of Clinical Pharmacology, the European Journal of Clinical Pharmacology, and Clinical Pharmacology and Therapeutics; as a general physician my minimum allowance should encompass the BMJ, Lancet, JAMA, and New England Journal of Medicine, not to mention the QJM, Clinical Medicine (the erstwhile Journal of the Royal College of Physicians), the Annals and Archives of Internal Medicine, and the American Medical Journal; my interests in ion transport biology should take me to the many physiology and cell biology journals in which such matters feature; and my work in adverse drug reactions adds to my reading list the long list of publications in which they are reported and discussed [1]. Given that there isn’t time enough, we all read what we can, accepting that we can pick some things up but have to let others fall by the wayside. But often our individual choices of what to read are random; with a more concerted approach we could perhaps ensure that the information that we neglect would not be important information. And that's what 26 members of the Clinical Section of the British Pharmacological Society have done for specialists in the field; they have produced the material for a website in which the essentials of specialist clinical pharmacology are summarized. The co-ordinators of the project, John Mucklow, Nigel Baber, Simon Maxwell, and Munir Pirmohamed, describe the background and the resulting website in this issue of the Journal (pp. 349–350). Keeping it up to date will be a challenge. Visit the website (accessed through http://www.bps.ac.uk) and let us know what you think. ‘It's tough to make predictions, especially about the future.’ This pithy saying has also been attributed to Yogi Berra, although other suspects include Mark Twain, Niels Bohr, and Albert Einstein. But whoever said it, it has that certain ring of truth that the best Yogi-isms have. Nevertheless, there are numerous publications on predicting in vivo events from in vitro studies, including studies of hepatic drug biotransformation, in which human liver cells are used as in vitro surrogates of their in vivo counterparts. It is therefore surprising to learn that the interindividual variability in the precise amounts of protein in hepatocytes or hepatic microsomes has not hitherto been known. Now Wilson et al. (pp. 433–440) have shown that there are 107 (range 65–185) million liver cells per gram of liver and 33 (range 26–40) milligrams of microsomal protein per gram of liver. By my calculations that makes an average of about 300 picograms of microsomal protein per cell, with considerable variability. These data were based on studies of only seven samples of healthy livers from patients with hepatic metastases, and other studies have involved fewer samples still. We are still a long way from understanding the true variability in hepatocyte microsomal protein content in healthy and diseased livers, in different parts of the liver, at different ages, and in different ethnic populations. Prediction is going to continue to be difficult. When Berra was compared with his son, who played for the Pittsburgh Pirates, he said ‘Our similarities are different.’ He might have been thinking of stereoisomers. The word for a bunch of grapes In Latin was racemus. So, when in 1822 Kestner isolated an acid from grapes, Gay-Lussac called it racemic acid. Racemic acid was in all respects chemically identical to tartaric acid, except that it did not rotate polarized light, a phenomenon described by Jean-Baptiste Biot, who postulated molecular asymmetry. Later Louis Pasteur crystallized racemic acid and saw in his microscope that it contained two types of crystal, left-handed and right-handed. He physically teased them apart and found that a solution of one behaved like ordinary tartaric acid, while a solution of the other rotated light in the opposite direction. Racemic acid, being a mixture of the two, was neutral to polarized light. So the term racemate was adopted to describe a mixture of equal amounts of two stereoisomers. Of all synthetic drugs used in clinical practice, about 40% are chiral and about 90% of those are marketed in the racemic form (i.e. as an equal mixture of the two stereoisomers). And sometimes knowing about the differences in the pharmacology of the different isomers can be useful. For example, citalopram, a selective serotonin reuptake inhibitor commonly used for the gripes of depression, is a racemic mixture of two isomers, S and R; the former is thought to be the active moiety, while the latter has little activity [2]. And S-citalopram is inactivated by two demethylation steps catalysed sequentially by CYP2C19 and CYP2D6. Now Herrlin et al. (pp. 415–421) have studied the differences in the disposition of racemic citalopram in poor and extensive metabolizers of both isozymes, genotypically and phenotypically confirmed. The poor metabolizers had higher ratios of the S to R enantiomers and tended to have more adverse effects, particularly the one subject who was a poor metabolizer via both isozymes. However, subjects with this double whammy of poor metabolism are too rare (about 1 in 500 in the UK) to warrant routine screening. We await studies of escitalopram in similar panels of subjects. Citalopram and other serotonin reuptake inhibitors feature in another paper in this month's Journal. Moclobemide is a selective inhibitor of monoamine oxidase type A. It was developed in the expectation that it would not pose the problem of the so-called ‘cheese reaction’ experienced when monoamines (such as tyramine in foods) are taken with the non-selective monoamine oxidase inhibitors that were once commonly used in the treatment of depression. This expectation has been largely fulfilled at the usual dosage range of moclobemide (150–600 mg day−1), although the interaction with tyramine-containing foods may be a risk at higher dosages [3]. Isbister et al. (pp. 441–450) now show that although in overdose moclobemide is relatively safe, adverse reactions can occur when it is taken in overdose with inhibitors of the reuptake of serotonin. Of 106 subjects who took an overdose of moclobemide, 21 had taken another drug with serotonergic action, sometimes in therapeutic dosage; 11 of those developed evidence of serotonin toxicity. There has also been an earlier report of an interaction of moclobemide with clomipramine [4], although combinations with amitriptyline and desipramine have been reportedly well tolerated [5]. When non-selective monoamine oxidase inhibitors were widely used to treat depression most psychiatrists felt that it was dangerous to coadminister them with tricyclic antidepressants, because of numerous reports of what we would now recognize as the serotonin syndrome, although some brave individuals, notably William Sargent [6], were advocates of the combination. This latest study suggests that caution is indeed warranted, particularly in those who are thought to be at high risk of self-poisoning. As Yogi Berra might have said ‘It's déjà vu all over again’.