Insect cells have been propagated in monolayers in T-flasks or in suspension culture in spinner flasks, the latter being conducted over a range of spinner speeds. In both configurations, the cells were also infected with either wild or recombinant beta-galactosidase baculovirus at MOI of 0.1, 1 and 10. The strength of both uninfected and infected cells was also measured by a micro-manipulation technique. No significant difference in growth rate was obtained between monolayer culture and suspension culture at the spinner rate which was optimum for growth. This optimum was quite sharp. At the lowest speeds cells settled, whilst above the optimum speed the spinner action led to significant cell damage. The maximum infectivity was obtained at this optimum speed which also gave maximum survival after infection. There were significant changes of cell survival and infection, even over relatively small changes of speed, and presumably energy dissipation rate. As changes in growth in turbine-agitated bioreactors have been shown to be much less, even when the energy inputs varied by two orders of magnitude, these findings throw doubt on the usefulness of spinner flasks for assessing "shear" sensitivity of cell lines. The percentage of infected cells and beta-galactosidase production were significantly lower in the monolayer culture compared to that in the suspension culture at MOI values below 10 pfu/cell. This difference is explained as being due to the reduced movement of released virus particles from infected to non-infected cells in the T-flasks.