Breeding bird surveys in Drenthe (2681 km2) in 1975-80 and 1990-95 revealed resp. 800- 1000 and 300-500 Whinchat pairs, i.e. >50% of the Dutch population. Whereas 75% of the breeding pairs in 1975-80 was residing in farmland, this proportion had declined to 10-15% in 1990-95. Earlier censuses covered a series of years (totals being obtained by summing regions and highest annual counts/region), therefore likely obscuring the degree of the decline. To counter such effects, a survey was instigated in 1998, during which 68% of the entire province was covered (Fig. 1). Of Whinchat-suitable habitats, 91% was covered, i.e. heaths, moors, valleys of rivers and streams, and nature reserves. The survey revealed 267 territories (Fig. 2). Given the quality and area of non-covered regions, this total amounts to an estimate of 300-350 pairs for Drenthe in 1998. Grasslands (49%) and heaths and moors (45%) accounted for the majority of pairs, with much smaller numbers in fallow land (3%) and forestry plantations (3%) (Appendix 1). Single-species grassland was largely avoided; hayfields (35%) and grasslands with high botanical diversity (26%) were favoured. Intensively farmed grasslands were thinly occupied when sufficient herbage-rich ditches and roadsides were available (26%). Whinchats had almost disappeared from farmland by 1998; 86% of all pairs were located in nature reserves, the remaining pairs mainly in wet parts of regular farmland where intensity of farming practices was relatively low. This is strikingly different from findings in 1975- 80, when the majority of Whinchats still bred in regular farmland. Even in nature reserves, Whinchats favoured moist and wet parts (Appendix 2). The presence of sitting posts high seemed to be important as well (Appendix 2); open stretches of farmland were avoided. In 112 out of 166 territories, alarm-calling adults or fledglings were recorded. Several local studies seem to indicate that reproduction is poor, and probably does not suffice to guarantee a stable population. This needs further substantiation.