I think the best answer is "we don't know" but I do think there's a biological reason. We do know that there are physiological changes responsible for a change in color preference as people become older (older people prefer drab colors). That isn't to say that culture isn't responsible at all. Even if only 51% of girls prefer pink, then companies will make more pink clothes, and because there are more pink clothes and people think girls prefer pink, girls are more likely to wear them. Culture can emphasize such differences easily.

20-odd years ago our sons were at a boys' prep school in England where two cousins of my mother had been in the 1920s. When I said our boys were going to the same school, the first question from relatives of that generation was always "Do they still wear pink?", so it must have made an impression even then. The answer was (and I believe still is) yes. Properly the colour is "Leander pink", a deep salmon-pink, but their rugby shirts were a more ordinary pink with white, and the grey blazers were trimmed with frankly baby pink ribbon. The boys did not seem to mind. And our boys' younger (male) cousins happily wore the outgrown pink rugby shirts at their home many miles from the school, where they must have really stood out.

Pink clothes have been around for millenia; as the first comment says, the association with 'girls' is around 100 years old, and is, as another commenter pointed out, specifically Western.There is simply no doubt that this is an example of the way gender is reproduced within our society. Other commenters describe making a conscious decision to avoid so gendering their children's clothes, and this being resisted by the child. However, this does *not* show that female children have a preference for pink. It shows that from the youngest age, children pick up on non-verbal communication around them, and use it to communicate their own preferred social identities. Not only are these children encountering pink garments and objects in the context of their female peers, they are also influenced by the fact that garments and objects designed to appeal to 'girls' (whose other characteristics are continually reinforced by advertising, television programmes, cartoons, etc.) are *also* pink. Pinkness iteself is not the source of the appeal, but is highly correlated with it in social and cultural terms, and therefore preferred.Primate studies of such behaviours and preferences are no more relevant than if they found that female primates consistently prefer the shape of the letter 'a' over other letters. Human beings have evolved the ability to use colour (both physically, and in thought and communication) as complex translinguistic systems. Primates have not, and their responses to colour in general are distinct from our own. Further, extensive research has shown that individual colour preferences in humans vary widely. Historical research is absolutely clear that colour 'meanings' change within cultures across time.The only colours for which there is any evidence of instinctual cues in any way over-riding such cultural considerations are bright red, and bright yellow, due to their warning functions in animals. Although we (in Western culture) characterise pink as closely connected to red, again, this is cultural, as there is little visual/perceptual resemblance.In short 'girls' prefer pink: female children, across cultures and times, do not.

As a young child I remember being obsessed with finding the perfect red (pencils, felt tips etc.). The colour red just seemed to resonate with me. Now I get the same feeling from looking at intense blues and purples. Some colours resonate so strongly with me they are like religious experiences. I would love to know the scientific reason for this.

There is a good bit of evidence that females prefer red shades, especially purple, more often than men do, and it cuts across cultural boundaries. In fact, there are females that are "tetrachromatic", having four sensory pigments in the eye compared to the more usual three. In tests done worldwide, there are always color preferences between genders. For more, check out the journal Current Biology, the Aug. 21, 2007, issue, for a paper by Anya Hurlbert and Yazhu Ling. In it, they report that Chinese boys and girls showed the same patterns of color preference that Americans did. They conclude that the preferences are innate.