Amorphous computingg1] is the study of programming ultra-scale computing environments of smart sensors and actuators that communicate locally via wireless broadcast. In such environments, where individual elements have limited resources, aggregation into groups is useful for specialization, fault-tolerance, and resource allocation. This paper presents a new algorithm , called clubs, that takes advantage of the local communication to eeciently aggregate processors into groups in an amorphous computer. Time taken is proportional to the local density of processors, even in an asynchronous setting. The physical embedding of the amorphous computer is used to derive an upper bound on the number and density of groups formed. The clubs algorithm can be extended to adapt to processor failures and to nd the maximal independent set (MIS) and + 1 vertex coloring in O(log N) rounds, where N is the total number of elements and is the maximum degree. Simulation results and example applications are presented.