JOURNAL OF AVIAN BIOLOGY 31: 567 – 575. Copenhagen 2000 How costly is clutch formation in the Audouin’s Gull Larus audouinii ? Xavier Ruiz, Lluis Jover, Vittorio Pedrocchi, Daniel Oro and Jacob Gonza ´lez-Solı ´s Ruiz, X., Jover, L., Pedrocchi, V., Oro, D. and Gonza ´lez-Solı ´s, J. 2000. How costly is clutch formation in the Audouin’s Gull Larus audouinii ? – J. Avian Biol. 31: 467–575. During the Audouin’s Gull’s breeding season at the Ebro Delta in 1993, 24 fresh eggs from eight three-egg clutches (modal clutch-size) were collected at the peak of the laying period. Eggs were processed to obtain formalin-fixed yolks, which were halved and stained using the potassium dichromate method. Digitized images of the yolks were examined to assess the daily rates of yolk deposition. We used these data in combination with egg compositional analysis to build a model of energy demands during the formation of an average clutch in Audouin’s Gull. To show how the different parameters of clutch formation affect the daily energy investment peak, we performed a simulation analysis in which the rapid yolk development (RYD) period, the follicle triggering interval (FTI), the laying interval (LI) and the albumen synthesis period (ASP) were allowed to vary simultaneously. In our sample, the mean RYD period was seven days with a range from six to eight days. There were no significant differences in yolk volume among eggs in a clutch, but albumen volume was significantly smaller in third eggs. According to our model the albumen synthesis of the a-egg coincides with the energy demand peak for clutch formation. This peak represents an increase by ca. 42% in female energy requirements. Values obtained from the simulation analysis showed that only the ASP of the a-egg and the RYD durations of the second and third follicles produced noticeable reductions in peak energy investment. We predict that in gulls, whose laying intervals seem to be kept constant, significant increases of the durations of the RYD periods of second and third eggs, or even significant reductions of yolk size of these eggs, may operate simultaneously to match the energy demands during clutch formation to the prevail- ing food conditions. X. Ruiz (correspondence), V. Pedrocchi, D. Oro and J. Gonza ´lez -Solı ´s, Departament de Biologia Animal (Vertebrats), Uniersitat de Barcelona, Ada. Diagonal 645, ES -08028 Barcelona, Spain. E-mail: xruiz@porthos.bio.ub.es. L. Joer, Departament de Salut Pu ´blica (Bioestadistica), Facultat de Medicina, Uniersitat de Barcelona, Casanoas 143, ES -08036 Barcelona, Spain. Present address of D. Oro: Instituto Mediterra ´neo de Estudios Aanzados IMEDEA, CSIC – Uniersitat de les Illes Balears, Ctra. Vallde - mosa, km. 7500, ES -07071 Palma de Mallorca, Spain. Reproductive effort per breeding attempt is one of the main parameters defining life-history strategies (Stearns 1992). In birds, clutch size is viewed as one of the traits showing close adjustments to individual capacity and local conditions (Lack 1968), but since many brood manipulations have shown that birds can rear more young than the number of eggs they lay (see Monaghan and Nager 1997 for review) the proximate mechanism by which clutch size is adjusted remains poorly under- stood. Recent studies suggest that costs of laying and incubating eggs might also play an important role in parental fitness (Monaghan et al. 1998). Laridae have been recognized among those birds which have comparatively high energetic demands for egg formation (Ricklefs 1974, Robbins 1981, Houston et al. 1983). It has been suggested that such costs may limit clutch size in gulls to three eggs (Reid 1987). However, experiments in which additional food was supplied to well-fed laying females without any further increase of clutch size (Hiom et al. 1991, Bolton et al. © JOURNAL OF AVIAN BIOLOGY JOURNAL OF AVIAN BIOLOGY 31:4 (2000) 567