1 The tiller dynamics of the two dominant species in a grassland, Agro
stis stolonifera and Lolium perenne, were followed for two years in a
field grazing experiment. Sheep grazing was applied in three seasons a
t two grazing intensities; winter, spring (in both seasons grazed or u
ngrazed) and summer (hard or light grazed). 2 Tiller densities of both
species showed peaks in late summer and troughs in spring due to vari
ation in tiller birth rates. Tiller death rates were not seasonal. Yea
r-to-year weather variation also affected the dynamics. 3 Harder summe
r grazing increased tiller birth and death rates of both species, incr
easing tiller turnover, but did not affect tiller densities. The tille
r death rates were higher and the summer peak in birth rates was earli
er in the winter-grazed paddocks. Therefore, winter grazing decreased
tiller densities over the two years by reducing densities in the non w
inter-grazed paddocks during the late summer months. There were no mai
n effects of spring grazing treatment or interactions among grazing tr
eatments. Tiller birth and death rates were density-dependent, indicat
ing that the populations were regulated. 4 Agrostis generally had high
er tiller birth and death rates and therefore had a more rapid tiller
turnover than Lolium. Fewer tillers flowered and there was less seedli
ng establishment in the Agrostis populations. However, both species sh
owed similar seasonal dynamics, density-dependent regulation, effects
of between year climatic variation and responses to spring and summer
grazing treatments. Agrostis showed a significantly greater response t
o winter grazing. 5 The tiller populations of both species were regula
ted by density-dependent birth and death rates but showed seasonal dyn
amics upon which were superimposed effects of grazing treatment and ye
arly variation in the weather. However, their tiller population number
s were little affected by grazing level in spring or summer. Only graz
ing in winter decreased their densities and this is the treatment most
likely to bring about changes in species composition in the grassland
.