SCIENTIFIC CORRESPONDENCE Clutch size and malaria resistance SIR - It was suggested more than a decade ago that the cost of egg production may be a major factor in the evolution of clutch size in birds 1, and that although this cost is a fundamental assumption of theories on the evolution of life-history patterns, it had not yet been adequately tested 1 • 2 , Man- ipulation of the rate of egg-laying is a pre- requisite for demonstrating this cost, but has been considered technically difficult 1 , Here we provide the first experimental evidence for a trade-off between egg pro- duction and parasite defence in the great tit (Parus major), a classic model species in the study of life-history evolution. We removed the first two eggs laid in half the nests of our breeding population. Females in this experimental group there- fore compensated by laying, on average, one more egg than females in a control group (a in the figure), but raised, on average, one offspring Jess (b in the fig- ure). Fourteen days after their chicks had hatched, we assessed the prevalence of Plasmodium in the peripheral blood- stream of females. Plasmodium is a malarial parasite that is common in the population of birds we are studying, and which potentially impairs survival and future reproduction 3 ,4 of its host. If there is a trade-off between egg production and parasite defence, one a, Mean number of eggs (± s.e.) laid by females of a control group (n = 41 broods) and an experimental group in which the first two eggs were removed the day of lay- ing. (n = 35 broods). Females of the test group produced a larger clutch than females of the control group ( t = 1.99; 74 d.f.; P 2 .,ailed = 0.05) . There was no diff er- ence between the two groups in mean hatching date (mean difference = 0.5 days; t = 0.30; 67 d.f.; P 2 .tailed = 0.76). b, Mean number of nestlings (± s.e.) 14 days post-hatching in the control group (n = 37) and in the test group (n = 33) . Females of the test group raised significantly fewer chicks than did females of the control group (t = 3.06; 68 d.f.; P2.tailed = 0. 003). c, Prevalence of Plasmodium spp. (± 95% confidence limits of binomial di stribution 12 ) in male (open bars) and female (shaded bars) great tits in the test group ( n = 30) and the control group (n = 35): Manipu- lation significantly affected the prevalence of Plasmodium spp. in females (x 2 = 6.49; 1 d.f.; P = 0.011), but not in males (x2 = 0.30; 1 d.f .; P = 0.59). Parents were caught with a door trap at the nestbox when nest lings were 14 days old. A drop of blood was t aken from t he brac hia! vein and a thin smear produced on a glass slide. Slides were air-dried, fixed wit h absolute methanol and stained with Giemsa stain. Each slide was screened with a light microscope under oil immer- sion for 10 min, and Plasmodium spp. would predict a higher prevalence of para- sites in the females of the experimental group, as they laid a larger clutch. Prev- alence in males of the experime ntal group should not increase, because the number of offspring to be raised is no greater than in the control group, We found that an increase in the clutch size by one egg leads to an increase in the prevalence of Plasmodium from 20% in the control females to 50% in the experimental females (c in the figure), thus supporting the hypothesis of a trade-off between egg production and parasite defence. As expected, there was no difference in the prevalence of Plasmodium in males of the two groups (c in the figure). Evolutiona ry theory predicts that the optimal clutch size is the result of a trade- off between current and future reproduc- tion2·5·6, At the physiological level, this trade-off predicts that parents investing heavily in their current offspring will have fewer resources to allocate to parasite defence, thereby impairing their future reproduction 7 • For birds, the studies in which clutch or brood size were experi- ment ally manipulated after egg-laying have shown that the clutch size giving rise to the most recruits is larger than the most common clutch, but has produced only equivocal evidence for a general trade-off a 10 -0 Q) " :, 9 -g a 8 Q) 'a 7 0 z b <I) .~ 6 Q) C: 0 5 z C 70 -0 50 .s <I) 1" :0 30 10 Control group Egg removal group Control group Egg re moval Control group g roup Egg removal group scored as present or absent. The experiments were performed in a stable breeding pop- ulation of great tits around the campus of the University of Lausanne, in spring 1995. NATURE · VOL 3 81 · 13 JUNE 1996 The cost of egg producti on in birds suc h as the great tit ( Parus major) shown here may be an important factor in t he evolution of clutch siz e. in life-history between current and future reproductionB-- 11 • It was stated recently that "all brood size manipulations ignore costs incurred earlier in the life history - in particular the cost of producing eggs, which might well affect the results of brood manipu- lation experiments" (p.158 of ref. 2). Our results support this contention, and pro- vide evidence that the cost of egg produc- tion may be an important factor in the evolution of optimal clutch size in birds, Anne Oppliger*, Philippe Christe Zoology Department, /ZEA, University of Lausanne, CH-1015 Lausanne, Switzerland Heinz Richner Z oo/ogy Department, Universit y of Bern, CH-3032 Hinterkappe/en, Switzerland 1 . Partridge, L. & Har vey, P. H. Nature 316, 20 ( 1985). 2. Stearns, S. The Evolution of Life Histories (Oxford Univ. Press, 1992). 3. Hayworth, A. M., van Riper, C. Ill & Weathers, W. W. J. Parasit. 73, 850-853 (1987) . 4 . Richner, H. , Christe, P. & Oppliger, A. Proc. natn. Acad. Sci. U.S.A . 92, 1192-1194 (1995) . 5. Williams, G. C. Am. Nat. 100, 687-690 (1966). 6. Charnov, E. L. & Krebs, J. R. Ibis 116, 217-219 ( 1974). 7. Gustafsson, L. et al. Phil. Trans. R. Soc . B346, 323-331 (1994). 8 . Gust af sson, L. & Sutherland, W. J. Nature 335, 813-815 ( 1988). 9 . Linden, M . & M0ller, A. P. Tr ends Ecol. Evol. 4, 367-371 (1989). 10. Lessel s, C. M. in Behavioural Ecology (eds Krebs, J. R. & Davies, N. B.) 32-68 (Blackwell, Oxford, 1991). 11. Pettifor, R. J. Anim. Ecol. 62, 145- 159 (1993). 12. Krebs, C. J. E col ogical Methodology (Harper Collins, New York, 1989). * The order of t he authors was determined by tossing a coin. Correspondence should be addressed to H.R. Erratum IN the Scientifi c Correspondence "Roots in mixotrophic algae" publi shed in the 30 May issue (Nature 381, 382; 1996), the authors' affiliations were transposed. They are shown correctly bel ow. J. R. M. Chisholm· , C. Daugat E. Ageront, P.A. D. Grlmontt J. M. Jauberf *0bser vatoire 0ceano/ogique Europeen, Centr e Scientifique de Monaco, MC 98000 Principality of Monaco t Unite des Enterobacteries, Unite 386 INSERM, lnstitut Past eur, 75724 Paris Cedex 15, France 565