This dissertation asks two questions: (1) how do pharmaceutical compounds shape the categories of human difference they are used to treat? And (2) how do divergent uses of compound classes change our principal understanding of their indication? Sociologists have largely explained pharmaceutical development and expansion in terms of medicalization, emphasizing mechanisms such as pharmaceutical advertising, changes in the political economy of healthcare, and corporate control over clinical trials. This dissertation argues that, between the synthesis of a new compound and the proliferation of new patient groups, crucial social processes occur. First, drug indications are refined through, and bounded by, intricate clinical trials. Second, the diagnostic profiles being targeted can change over the course of these trials. To make sense of this, pharmaceutical compounds are treated as the philosopher Ian Hacking calls an ‘engine for making up people.’ While Hacking describes pharmaceuticals in the context of medicalizing and normalizing people, this dissertation demonstrates how drugs also interact with human kinds in ways that modify both over time. To make this argument, I draw on three case studies examining the clinical history of amphetamines. I begin by looking at the work of psychiatrist Abraham Myerson in 1930s, whose clinical studies treating depressed housewives with Benzedrine sulfate led to a softening of clinical understandings of depression that would apply to a much broader patient pool. It is argued that Myerson’s work with Benzedrine sulfate informed the pharmacologic profile that subsequent generations of antidepressants would need to meet. Following this, I look at a series of pediatric clinical trials in the 1960s that led to the pairing of amphetamines with hyperactive patients. Here it is argued that the dominant description of psychostimulants as attention-enhancing drugs and the profile for attention-deficit/hyperactivity disorder (ADHD) emerged in tandem through a recursive process in a series of randomized-control trials (RCTs). Lastly, I trace the history of amphetamines and amphetamine-like drugs in the context of eating disorders, namely anorexia, bulimia, and binge eating disorder, demonstrating how the use of amphetamines in pharmacodynamic studies of eating behavior played an important role in the gradual partitioning of new disorders including bulimia and binge eating disorder.