Evolution and Human Behavior 19: 299–312 (1998)  1998 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010 1090-5138/98/$19.00 PII S1090-5138(98)00022-1 Birth Order and Familial Sentiment: Middleborns are Different Catherine A. Salmon and Martin Daly Department of Psychology, McMaster University, Hamilton, Ontario, Canada Effects of birth order on several aspects of family relations and self-identity were exam- ined in three studies. In Study 1, first and lastborn undergraduates were more likely than middleborns to refer to kinship in characterizing themselves. In Study 2, subjects were asked to whom they would turn under two scenarios of duress. First and lastborns were more likely to nominate parents, whereas middleborns were much more likely than other respondents to nominate siblings. In Study 3, analyses of historical archives and of an Internet questionnaire indicated that genealogical research attracts many more first- borns and many fewer middleborns than expected by chance. In all three studies, first and lastborns were much more likely than middleborns to nominate their mothers as the person to whom they felt closest. These substantial effects support Sulloway’s claim that birth orders constitute significant family “niches,” which differ with respect to the per- ceived dependability of parental investment and therefore also differ in the social orien- tations that they engender. © 1998 Elsevier Science Inc. KEY WORDS: Birth order; Genealogy; Kinship; Parental investment. T Received January 21, 1998; revised April 6, 1998. Address reprint requests and correspondence to: Dr. Catherine A. Salmon, Department of Psychology, McMaster University, Hamilton, Ontario, Canada L85 4K1. Fax: 905-529-6225; E-mail: salmon@ sympatico.ca Acting Editor for this paper was Charles Crawford, Department of Psychology, Simon Fraser University. heoretical models of the evolution of parental inclinations predict that parents will often treat their offspring differently. There are grounds for predicting discriminative parental solicitude in relation to a number of variables including offspring age, parental age, birth order, offspring sex, cues of phenotypic quality, and cues of parentage (Clutton-Brock 1991; Daly and Wilson 1987, 1995; Trivers 1974; Trivers and Willard 1973; Wilson and Daly 1994). The unifying notion behind these theories is that natural selection has shaped parental psychologies to function as if they “value” individual offspring and invest-