In a meta-analysis of 150,000 Americans aged 13–22, those whose scores fell in the top 5–10% in math, geometry, mechanical reasoning and engineering were overwhelmingly male (e.g., Hyde, Fennema, & Lamon, 1990). Questionnaire studies in several other countries yielded similar results (e.g., Mann, Sasanuma, Sakuma, & Masaki, 1990). Most young adult men express approximately eight to ten times more testosterone than most women; and it is widely hypothesized that this sex difference in hormone expression contributes to the above sex differences in cognition. Although adult sex differences in cognitive tasks are not sufﬁ- cient evidence for hormonal effects, biological data support this hypothesized correlation. The brain architecture associated with these spatial/mathematical skills has been associated with fetal testosterone (e.g., Geschwind & Galaburda, 1985; Grimshaw, 1995); and bodily levels of testosterone also contribute to spatial/mathematical dexterity across the life span (Janowsky et al., 1994). Further, it has been proposed that greater understanding of spatial, mechanical, mathematical, engineering and other rulebased systems is the result of more short-range and less long-distance neural connectivity, due to the exposure to prenatal androgens (e.g., Manning, 2002). Other traits that may be linked with prenatal testosterone expression are heightened attention to detail, intensiﬁed focus, and restricted (narrow) interests (e.g., Baron-Cohen, Knickmeyer, & Belmonte, 2005; Knickmeyer, Baron-Cohen, Raggatt, & Taylor, 2005); and elevated adult levels of testosterone have also been associated with these traits (Dabbs & Dabbs, 2000). It has been suggested that prenatal testosterone expression is also linked with less emotion recognition, eye contact and social sensitivity (e.g., Baron-Cohen et al., 2005), a poorer ability to judge what others are thinking or feeling (Baron-Cohen, 1995); and lack of empathy (Baron-Cohen et al., 2005). Adult expression of testosterone has been correlated with being less polite, respectful, considerate or friendly (Dabbs, 1997; Harris, Rushton, Hampson, & Jackson, 1996); and being more conﬁdent, forthright and bold (Nyborg, 1994). Adult levels of testosterone have also been positively correlated with sensitivity to social dominance, drive for rank, the tendency to create dominance hierarchies (e.g., Mazur, Susman, & Edelbrock, 1997), and aggressiveness (e.g., Dabbs, 1990; Mazur et al., 1997). Poor verbal ﬂuency and other language deﬁciencies have been associated with testosterone priming in the womb (e.g., Baron-Cohen et al., 2005; Knickmeyer, Baron-Cohen, Raggatt, Taylor, & Hackett, 2006; Knickmeyer et al., 2005; Geschwind & Galaburda, 1985; Manning, 2002). Last, a lower (masculinized) second to fourth digit ratio has been associated with high prenatal testosterone (Manning, 2002). Women excel at several linguistic skills in US populations (e.g., Halpern, 1992; McGuinness & Pribram, 1979) and other countries (Mann et al., 1990). Cross-culturally, women also excel at recognizing emotions in faces (e.g., Hall, 1984; McClure, 2000), reading a person’s emotions from voice, gestures and other non-verbal cues (e.g., Hall, 1984; McGuinness & Pribram, 1979) and interpreting a range of mental states (Baron-Cohen, Jolliffe, Mortimore, & Robertson, 1997). Although these sex differences are not sufﬁcient evidence for hormonal effects, studies suggest that nurturance and other prosocial skills are associated with adolescent and adult estrogen activities in human females (Nyborg, 1994) and females of many other mammalian species (Carlson, 2001). Adult estrogen facilitates memory for emotional experiences (Canli, Desmond, Zhoa, & Gabrieli, 2002). Estrogen replacement therapy increases verbal memory (Hogervorst, Williams, Budge, Riedel, & Jolles, 2000). In addition, estrogen receptor modulators can elevate adult working and episodic memory, executive function and verbal skills (Goekoop et al., 2005). Data suggest that fetal steroids affect several adult behaviors (Manning et al., 2000; Nyborg, 1994). Data also suggest that a higher (feminized) digit ratio is positively correlated with fetal estrogen (Manning, 2002). At present, however, there is insufﬁcient evidence that fetal testosterone enhances the growth of the 4th digit or that fetal estrogen facilitates the growth of the 2nd digit.