We must, however, consider alternative explanations for the value on virginity whiting, Bur bank, and Ratner(1986) have proposed that premarital permissiveness or restrictiveness is re- lated to the age at which girls marry reasoning that restrictiveness reflects a concern with preg nancy more than with virginity per se. Given the well-recognized biological phenomenon of adolescent subfecundity as a result of which a pregnancy is rather unlikely in the first year or so after menarche, they claim that societies in which girls marry relatively late prohibit pre marital sexual relations in order to ensure that no pregnancy occurs, whiting and his collal rators support their argument by selecting 50 societies from the Standard Sample and rating them forduration of maidenhood, the period between menarche and marriage and for sex- ual permissiveness versus restrictiveness(see Table 3). They find that those societies with more than four years of maidenhood are restrictive, those with a maidenhood of about three years are permissive and those with shorter maidenhoods are mixed with 13 restricting and 17 per- mitting sexual relations. They estimate that if sex is permitted, the probability of a pregnancy about 30 percent when maidenhood lasts from one to two years, somewhat less than 50 percent hen it lasts from two to three years, and about 60 percent when it lasts longer than three years Whiting et al. 1986: 278). They relate both age of marriage and attitude toward premarital sex to concerns about fertility and control over it rearranging their subsample into categories de- fined by marriage transaction, we see that these societies follow a pattern similar to that of the larger sample from which they are drawn(Table 4). The exception is the gift-exchange category, where by chance the researchers have selected the three permissive societies that exchange gifts: Trobriand Islands, Omaha, and Samoa The most problematic category for the Whiting et al. hypothesis is the group of societies in which maidenhood is short, with 13 societies that value virginity and 17 that do not (see Table 3). Their hypothesis does not account for the 13 restrictive societies, 43 percent of this cluster It should also be noted that four of the six societies with long periods of maidenhood, all of which value virginity, are dowry-giving. It was impossible to test the distribution in this cluster of 30 societies with short maidenhood because one of the categories had no representatives and neither the Chi-square nor Fishers Exact tests allow for an empty cell. However, the dis- tribution follows that for the larger subsample and for the entire sample(see Table Preindustrial Europe is the world area best known for long-delayed marriage(Stone 1977), and virginity has been of concern in European cultures until very recently. Against this, how Table 3. The value on virginity according to the duration of maidenhood Medium ource: Whiting et al. (1986). Positive where coded as sex prohibited or restricted negative where coded as sex permitted or encour- Table 4. virginity and marriage transactions for the subsamp Bride. Bride. ed None wealth service exchangeexchange Dowry dowry Source of subsample: Whiting et al. (1986) 726 american ethnologistWe must, however, consider alternative explanations for the value on virginity. Whiting, Bur￾bank, and Ratner (1986) have proposed that premarital permissiveness or restrictiveness is re￾lated to the age at which girls marry, reasoning that restrictiveness reflects a concern with preg￾nancy more than with virginity per se. Given the well-recognized biological phenomenon of adolescent subfecundity, as a result of which a pregnancy is rather unlikely in the first year or so after menarche, they claim that societies in which girls marry relatively late prohibit pre￾marital sexual relations in order to ensure that no pregnancy occurs. Whiting and his collabo￾rators support their argument by selecting 50 societies from the Standard Sample and rating them for "duration of maidenhood," the period between menarche and marriage, and for sex￾ual permissiveness versus restrictiveness (see Table 3). They find that those societies with more than four years of maidenhood are restrictive, those with a maidenhood of about three years are permissive, and those with shorter maidenhoods are mixed, with 13 restricting and 17 per￾mitting sexual relations. They estimate that if sex is permitted, the probability of a pregnancy is about 30 percent when maidenhood lasts from one to two years, somewhat less than 50 percent when it lasts from two to three years, and about 60 percent when it lasts longer than three years (Whiting et al. 1986:278). They relate both age of marriage and attitude toward premarital sex to concerns about fertility and control over it. Rearranging their subsample into categories de￾fined by marriage transaction, we see that these societies follow a pattern similar to that of the larger sample from which they are drawn (Table 4). The exception is the gift-exchange category, where by chance the researchers have selected the three permissive societies that exchange gifts: Trobriand Islands, Omaha, and Samoa. The most problematic category for the Whiting et al. hypothesis is the group of societies in which maidenhood is short, with 13 societies that value virginity and 17 that do not (see Table 3). Their hypothesis does not account for the 13 restrictive societies, 43 percent of this cluster. It should also be noted that four of the six societies with long periods of maidenhood, all of which value virginity, are dowry-giving. It was impossible to test the distribution in this cluster of 30 societies with short maidenhood because one of the categories had no representatives and neither the Chi-square nor Fisher's Exact tests allow for an empty cell. However, the dis￾tribution follows that for the larger subsample and for the entire sample (see Table 5). Preindustrial Europe is the world area best known for long-delayed marriage (Stone 1977), and virginity has been of concern in European cultures until very recently. Against this, how￾Table 3. The value on virginity according to the duration of maidenhood. Virginity Duration of maidenhoodb valueda Long Medium Short Yes 6 0 13 No 0 14 17 Source: Whiting et al. (1986). aPositive where coded as sex prohibited or restricted; negative where coded as sex permitted or encour￾aged. bLong where Whiting et al. coded the duration of maidenhood as long or very long; medium where they coded it as medium; short where they coded it as short or absent. Table 4. Virginity and marriage transactions for the subsample. Marriage transaction Virginity Bride- Bride- Gift Women Indirect valued None wealtha service exchange exchange Dowry dowry Total Yes 1 7 2 0 0 4 3 17 No 8 9 8 3 2 0 3 33 Source of subsample: Whiting et al. (1986). aIncludes token bridewealth. 726 american ethnologist