Committee formation in early American legislatures happened when those assemblies were inundated with petitions, a relationship unexamined in institutional political science. We develop a model where a floor creates committees to respond to topic-specific petitions, predicting committee creation when petitions (1) are topically specific, (2) are spread across constituencies, and (3) have complex subject matter, and predicting committee appointments from petition-heavy constituencies. Analysis of case studies and with two original datasets – petitions sent to the Virginia House of Burgesses from 1766 to 1769, and over 100,000 petitions sent to Congress and recorded in the House Journal (1789-1875) – shows petitions, their complexity and their geographic dispersion predict committee creation. Our theoretical argument embeds asset specificity in legislative institutions, and helps reinterpret the entropy of political agendas and the origins of standing committees in American legislatures.