Summary1. The incubation-limitation hypothesis states that clutch size in shorebirds is constrained by the ability to incubate more than four eggs successfully. In order to test predictions from this hypothesis, clutches of northern lapwings ( Vanellus vanellus ) were here experimentally increased and reduced by one egg at the onset of incubation. Parental behaviour, incubation time, hatching synchrony, egg viability, hatching success and body condition of hatchlings was compared with unmanipulated clutches. 2. Nest desertion and egg breakage was not observed in any of the experimental nests, showing that clutch enlargement did not alter the parental behaviour adversely. 3. There were no differences among the experimental groups in male and female attentiveness, but the total nest attentiveness was lower in reduced clutches than in the other clutch sizes; nor did the parents' frequency of nest visits differ between clutch sizes. 4. Enlarged clutches were incubated longer than controls and reduced clutches. There were, however, no significant differences among clutch sizes in partial or total clutch predation rates. Hatchability of the eggs (proportion of eggs hatching of those present in nests at the end of incubation) was marginally lower in the enlarged clutches (92·5% vs. 100% in the two other groups). 5. Contrary to predicted, hatching success was highest in enlarged clutches. The numbers of hatched eggs calculated from Mayfield estimates of daily nest and egg survival rates were 0·93, 2·10 and 3·14 in three-, four-and five-egg clutches, respectively. 6. When controlling for egg volume, mean body mass of newly hatched chicks was lower in clutches with five eggs than in other nests. 7. We suggest that the lowered body condition of chicks could outweigh the benefits of laying five instead of four eggs in northern lapwings. Future studies of the selective mechanisms of clutch size in shorebirds should not only focus on the number of young that hatch, but also their quality.