The genetic study of ADHD has made considerable progress. Further developments in the field will be reliant in part on identifying the most appropriate phenotypes for genetic analysis. The use of both categorical and dimensional measures of symptoms related to ADHD has been productive. The use of multiple reporters is a valuable feature of the characterization of psychopathology in children. It is argued that the use of aggregated measures to characterize the ADHD phenotype, particularly to establish its pervasiveness, is desirable. The recognition of the multiple comorbidities of ADHD can help to isolate more specific genetic influences. In relation to both reading disability and conduct disorder there is evidence that genes may be involved in the comorbid condition that are different from pure ADHD. To date, progress with the investigation of endophenotypes for ADHD has been disappointing. It is suggested that extending such studies beyond cognitive underpinnings to include physiological and metabolic markers might facilitate progress.