Sugarcane varieties, resistant and susceptible to Eldana saccharina Walker (Lepidoptera: Pyralidae) were grown in a pot trial, in silicon (Si) deficient river sand with and without calcium silicate as a source of silicon (Si). At nine months, the plants were artificially infested with E. saccharina and the larvae harvested seven (rind of sugarcane drilled; harvest 1) and 21 (rind not drilled; harvest 2) days after inoculation. For Harvest 2, fertilisation with Si significantly increased stalk length, total number of internodes, rind hardness and stalk mass for the susceptible variety N26 compared with the control. The same trend was observed for N11 (susceptible), N21 and N33 (both resistant), although differences in these varieties were not significant. The length of stalk bored was significantly less for the treated sugarcane compared with the control, although this was significant only for N26. Larvae were weighed prior to inoculation and at harvest. Mass of larvae at harvest was significantly less in N26 treated with Si compared with the control, and the same trend, although not significant, was recorded for N11, N21 and N33. Larval survival did not differ significantly between control and treatment (all varieties). Similar, but non-significant, trends were recorded for harvest 1. The application of Si shows a general trend of increased resistance of sugarcane to attack by E. saccharina consistent with the 'mechanical barrier hypothesis'.