I hate to quote patent literature to make a point but sometimes the background of the invention can give some useful references to begin a search. On the subject of human phermones I found this in patent application #20030049726, as yet unawarded.
"The existence of human pheromones, however, is controversial. Human reproductive behavior is largely independent of oestrous-promoting hormones. Maternal behavior may occur without pregnancy and sexual human behavior is also tempered by culture, learning and personal experience. Moreover, evolutionary enlargement of the human neocortex has enabled the rapid assimilation and integration of information from a number of senses. Hence, it has been argued that it is implausible that humans would be under significant behavior and endocrine regulation by pheromones. Nevertheless, the existence of human pheromones was first suggested by the observation that women living together can develop synchronized menstrual cycles under specific conditions (McClintock, Nature 291:244 (1971)). The causal agents were later attributed to odorless pheromone-like substances produced in female underarms (Stern and McClintock, Nature 392:177 (1998)). There are also reports suggesting short-chain fatty acids found in vaginal secretions isolated from vaginal secretion of sexually active human females can act as sex-attractants (Michael et al., Psychoneuroendocrinology 1:153 (1975)); Sokolov et al., Archives of Sexual Behavior 5:269 (1976)).
Much human pheromone research has centered on the 16-androstenes, which comprise a family of related steroids that have pheromone activity in animals. Androsterone (5-alpha-16-androst-16-en-3-one) and its alcohol form, androstenol (5-alpha-16-androst-16-en-3-ol) are porcine pheromones synthesized in the boar testes and submaxillary glands and, which induce recipient sows to adopt the mating stance (Reed and Melrose, Br. Vet. J. 130:61 (1974); Perry et al., Animal Production 31:191 (1980)). These and other related 16-androstenes are also synthesized in human testes and believed by many investigators to have pheromone activity in humans (see, for example, Gower and Ruparelia, J. Endocrinol. 137:167 (1993); U.S. Pat. No. 5,278,241; U.S. Pat. No. 5,272,134; U.S. Pat. No. 5,969,168; U.S. Pat. No. 5,965,552). 5-alpha-16-androst-16-en-3-ol is the most abundant of the 16-androstenes in human urine. Androsta-4,16-dien-3-one is the most abundant 16-androstene present in human semen, in male axillary hair and male axillary skin surfaces (Nixon et al., J. Steroid Biochem. Mol. Biol. 29:505 (1988); Rennie et al., In: Chemical Signals in Vertebrates, pages 55-60 (Oxford University Press 1990); Kwan et al., J. Steroid Biochem. Mol. Biol. 43:549 (1992)). Androstenes are also found in the human axillary sweat secreted by the apocrine glands, which are sites for pheromone production in lower animals (Brooksbank et al., Experientia 30:864 (1994))."
I haven't read any of these references but they seem to be in reputable, peer-reviewed journals.
You can read the entire patent application yourself at:
http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=3&f=G&l=50&co1=AND&d=PG01&s1=%22Holloway%2C+James%22.IN.&s2=%22Lok%2C+Si%22.IN.&OS=IN/"Holloway,+James"+AND+IN/"Lok,+Si"&RS=IN/"Holloway,+James"+AND+IN/"Lok,+Si"