1. The functional and numerical responses of two generalist raptors (hen harrier and peregrine) were studied on one moor for 6 years and on five other moors for 5 years.
2. Neither species showed numerical responses to grouse abundance. Harrier densities were highest in areas and years where their small prey (meadow pipits and small mammals) were most abundant. Peregrine densities were highest on southern study moors, probably in association with high abundance of racing pigeons.
3. For harriers preying on grouse chicks, the available data fitted a sigmoidal or type III functional response. Peregrines showed a type II response to adult grouse densities around eyries.
4. The proportion of grouse chicks taken by harriers was estimated to have been highest at densities of 67 chicks km–2 (equivalent to a mean of about 12 broods km–2). The proportion of adult grouse taken by peregrines appeared to be inversely density dependent, such that an increasing proportion of grouse was taken at grouse densities below 20 km–2.
5. In the absence of persecution, the impact of harriers on grouse populations is most likely to be greatest on moors where alternative prey and thus harriers are abundant. The question of whether harriers may dampen grouse population cycles at low grouse density is discussed.