Carrot cavity spot, caused by a complex of Pythium species, is characterized by sunken elliptical lesions on the taproot. Recent epidemiological studies of P. violae have demonstrated the occurrence of both primary and secondary infections, with two types of secondary infection, autoinfection and alloinfection. Investigating the mechanisms underlying alloinfection and the role of carrot lateral roots, we asked whether direct physical root contact plays a role in alloinfection and whether root exudates enhance mycelial growth in soil alone. A rhizobox system was designed to differentiate the effects of each mechanism: a buffer zone created by nylon mesh was used to test the first mechanism, and young carrots with a root system similar to lateral roots were used to test the second. Alloinfections were generated in rhizoboxes via diseased taproots transplanted close to healthy, mature carrots. The nylon mesh had no significant effect on disease intensity (reflecting alloinfection), providing evidence that mycelial growth in soil contributed more to disease spread than did physical contact among roots. Nor did young carrots significantly affect alloinfection; thus root exudates had little effect on mycelial growth