Letter to the Editor; Reply To Wilson
Winnifred B. Cutler
Hormones and Behavior, 22, pp. 272-277., 1988.
© Copyright 1988.
To the Editor:
Thank you for the opportunity to reply to Dr. Clyde Wilson's comments regarding our paper.
Dr. Wilson questions the conclusions drawn through his elaboration of assumptions, many of which appear to be incorrect. Thus, it may be most appropriate to first correct these errors.
(1) Contrary to Dr. Wilson's statement defining our test hypothesis, we did not posit that aberrant cycle lengths would "change to normal" but rather that they would move toward normal as a consequence of treatment with male axillary extract.
(2) It is not true that the length of the "second" cycle for each subject was used to test for significant change. Rather, the length of the last complete cycle within the experimental protocol was used. For some women, this was the second cycle, but for others it was the first or the third. For example, all three of the women with short cycles (<26 days) at first cycle did indeed show last cycles that had moved toward the 29-day cycle length. In all three cases, an intermediate second cycle occurred before the last.
(3) Dr. Wilson said, "earlier work by Cutler and associates...found an association [emphasis added] between subject's weekly coital behavior and 29 day cycles." In fact, our developing sets of experimental data had provided the conclusion that the relationship was probably causal (Cutler, Preti, Huggins, Erickson, and Garcia, 1985) because the presence of regular weekly sex with men consistently produced the associate variables reflecting fertility (circa 29.5-day cycle lengths) (Cutler et al., 1985; Cutler, Garcia, and Krieger, 1979a), fertile-type basal body temperature graphs (Cutler et al., 1985), absence of deficiently short hyperthermic phase lengths indicative of luteal phase defects (Cutler et al., 1985, 1979a, Cutler, Garcia, and Krieger, 1979b; Cutler, Garcia and Kreiger, 1980; Cutler, Davidson, and McCoy, 1983), as well as substantially and statistically significantly higher levels of estrogen (Cutler et al., 1983; McCoy, Cutler, and Davidson, 1985; Cutler, Garcia, Huggins and Preti, 1986). In contrast, the presence of any of those indicators of fertility did not necessarily predict what kind of sexual behavior the woman would reflect. Therefore, since the behavior consistently predicted a physiological parameter while the physiological parameter did not necessarily predict the behavior, I had previously come to the conclusion (Cutler et al., 1985, 1986) that regular weekly sex with men produced fertile-type cycles in women.
(4) Dr. Wilson believes that “the decision to remove the cycles of the four subjects who were having regular weekly coital activity was not justified by evidence from this experiment.” This is true. The precedent had already been set that regular weekly sex with men was shown to produce circa 29.5-day cycles (Cutler et al., 1985, 1979a, b, 1980, 1983, 1986; McCoy et al., 1985; Cutler, Garcia and Kreiger 1979c). Because these data had been so consistent and compelling in suggesting that regular sexual behavior actually promoted fertile-type cycles, the decision to remove from the data pool the four subjects who were having weekly coital activity was justified.
(5) Dr. Wilson correctly notes that we did not have an observed preexperimental database from which to determine changes in the subject's cycle lengths. Unfortunately, the financial constraints of the funding process coupled with the length of the fall semester limited the experiment to 14.5 weeks. As explained in the original paper, the preexperimental questioning of subjects about their perception of their cycle lengths did allow us to create an experimental pool of aberrant cyclers. By randomly assigning these self-perceived aberrant cyclers into either the placebo or the male extract groups, an objective control group to test cycle length differences provided appropriate experimental objectivity. As we concluded at the end of the paper, the next step along this research path requires a longer experimental period to include both baseline data and subsequent crossover application of experimental essence. This next experiment will require 40 weeks rather than 14.5 available in this report.
(6) Dr. Wilson correctly notes that three normal cycles and one aberrant cycle were removed when weekly active women were removed from the data base. This fits the previously reported data showing that women who have regular weekly sex with men tend to have normal cycles (Cutler et al., 1985, 1979a, b, 1980, 1986). It is entirely possible that all four of these women were indeed aberrant cyclers at the beginning of the experiment, as per their self-perceptions, but that the onset of a regular sexual relationship promoted the high frequency of normal cycles among them. The decision to remove those subjects who showed regular weekly sex with men form an experimental analysis of data is sensible if the regular sex is either more powerful or as powerful as the extract. If so, regular sex would mask the influence of extract.
(7) The fact that removal of these four cycles did not impact the extract and blank samples in the same way seems to reflect the fact that women with regular weekly sex with men, even in the presence of a blank sample, do show normal cycles. Thus, the effect may not be chance distribution at all, but rather a direct cause and effect of sex behavior promoting the normal cycles.
(8) It is correctly noted that women who are engaging in regular weekly sex with men probably engage in sexual activity with men who have “violated all restrictions regarding use of deodorant, deodorant soaps, and perfumes” require of the extract donors. Although Dr. Wilson argues that the male extract treatments should have overridden other influences, in fact, we have concluded just the opposite.
The real thing, that is the presence of a man, appears to offer a more powerful influence than the chemical substitute of his extract. The consistent observation of redundancy in biological systems would compel the biologist to expect multiple stimulatory pathways leading toward fertility. Perhaps the extract is only one of these contributions which the man provides the woman in promoting fertility. The fact that male essence, as it is extracted, has an odor in no way suggests that the critical substance that promotes the effect is odorous. The odor may serve as a marker but is not necessarily the active substance. Therefore, the inhibition of that odor by American men may be irrelevant. Our experimental data have not yet permitted the critical experiments to enable defining which constituent within the extract pool is the active substance. We must not make the mistake of assuming that the active substance is the one that we can smell. In fact, the long history of the use of deodorants would argue just the opposite. Human beings are still fertile in spite of the making of odor. Regular weekly sex with men promotes fertile-type cycles in spite of the use of deodorants, perfumes, and so forth.
Wilson quotes the statement indicating that “most of the women whose cycles remained aberrant were recipients of the placebo…and most of the women whose cycles became normal in length were recipients of extract.” It is true that the experimental design employed here - one in which all women who reported themselves to be aberrant cyclers were assigned to the male extract experiment and then randomly assigned into the placebo or the extract groups - does assume that women who remained aberrant began aberrant and those whose how themselves to be normal became normal. This follows from the design of the placebo-controlled experiment and suggests the conclusions of the experiment - that those who became normal did so as a result of being in the male extract group and those who remained aberrant did so as a result of not having either a regular weekly sexual relationship or an active substance, whatever it may turn out be, that is in the male extract.
(9) Dr. Wilson states that “the only observed data available to the researchers were the first cycle lengths recorded past the midpoint of the experiment. “ This is not correct. Neither is it correct that in any way were the subjects predicting whether or not they would have aberrant cycle lengths. They were simply assigned to the group based on their perception of cycling status at the beginning of the experiments, not on any presumption about where they would continue to be as their lives continued.
Women with short cycles provided at least three cycles of data; although those with long cycles tended to have either one or two cycles. Therefore, the first cycle was often recorded well before midpoint of the experiment.
(10) Dr. Wilson reviews the data from our female essence experiment (Preti, Cutler, Krieger, Huggins, Garcia, and Lawley, 1986) as evidence that, despite the women entering the study claiming to have normal cycles, there were a good many aberrant-length cycles. Dr. Wilson uses this piece of information to argue that subjects cannot correctly predict whether their cycles will be aberrant or normal. We were not asking subjects to predict what kind of cycles they would have, but rather to tell us what kind of cycles they believed themselves to have at the start of the experiment. In fact, inspection of those data showed a very large proportion of these women having sporadic sexual behavior. It is entirely possible that these women entered the experiment believing themselves to be normal cyclers, while coincidentally engaged in a regular heterosexual relationship which subsequently ended. It is entirely possible that they changed their sexual activity during the course of the study. Single nulliparous women in their early 20s during the period between 1976 and 1983 at the University of Pennsylvania commonly fit such a behavioral pattern (Cutler et al., 1985, 1979a, b, 1980, 1986, 1979c; McCoy et al., 1985; Preti et al., 1986). High frequencies of variability in menstrual cycling, as Dr. Wilson correctly states, are well known to be typical in samples of young women. I would hasten to point out that they are coincidentally well known to exist in samples of women with either sporadically active sexual behavior or celibacy, precisely the circumstance of this age group at the time of data collection.
(11) Although Treloar et al., did show that the greatest variability in menstrual cycle length intervals occurs before the age of 21, a more important discovery was demonstrated by Cutler (see Cutler and Garcia, 1980; Cutler, Garcia, and Edwards, 1985; Cutler and Garcia, 1984). They showed that it was not the chronological age but the gynecological age that was the critical factor for attaining gynecological maturity. Seven years after the onset of the menarche, gynecological maturity had been reached, and with it, an increased propensity toward a circa 29.5-day cycle length. We did control for this youthful immaturity of menstrual cyclicity by including only subjects who had reached gynecologic maturity (i.e., menstruating more than 7 years) (Cutler, Preti, Huggins, Garcia, and Lawley, 1986).
For these reasons, “Male extract does appear able to affect female menstrual pattern; but the effect may require some priming time since 14 weeks were required to reveal the effect." I heartily agree with Dr. Wilson that the work needs to be continued, and that a longer period of study needs to be undertaken. I hope to see the budgetary capacity to perform the necessary work.
In conclusion, the decision to remove the cycles of the four women who had weekly coital activity is justified, not simply by the evidence in this experiment but by the evidence obtained from ten experimental papers cited as the basis for that decision. Furthermore, the absence of a preexperimental data base from which to measure changes in the women's cycle length, while limiting the strength of the conclusions that could have been drawn, is balanced by the use of placebo control (versus the male extract) in random assignment of subjects who self-reported aberrant cycles (Cutler et al., 1986).
It is hoped that subsequent funding will allow “a more carefully designed experiment with observed data on preexperimental cycle lengths…”
Cutler, W.B., Davidson, J.M, and McCoy, N. (1983). Sexual behavior, steroids, and hot flashes are associated during the perimenopause. Neuroendo. L 5, 185.
Cutler, W.B., Garcia., C.R., and Edwards, D.A. (1985). Menopause: A Guide for Women and the Men who Love Them. W.W. Norton & Co., New York.
Cutler, W.B., and Garcia, C.R. (1980) The psychoneuroendocrinology of the ovulatory cycle of women. Psychoneuroendocrinology.
Cutler, W.B., and Garcia, C.R. (1984) The Medical Management of the Menopause and Premenopause: Their Endocrinologic Basis. Lippincott Co., Philadelphia, PA.
Cutler, W.B., Garcia, C.R. Huggins, G.R., and Preti, G. (1986). Sexual behavior and steroid levels among gynecologically mature premenopausal women. Fertil, Steril. 45, 496-502.
Cutler, W.B, Garcia, C.R., and Krieger, A.M. (1979a). Sexual behavior frequency and menstrual cycle length in mature premenopausal women. Psychoneuroendocrinology.
Cutler, W.B, Garcia, C.R., and Krieger, A.M. (1979b). Luteal Phase defects: A possible relationship between short hyperthermic phase and sporadic sexual behavior and menstrual cycle length in women. Horm. Behav. 13, 214-218
Cutler, W.B., Garcia, C.R. and Krieger, A.M., (1979c). Infertility and age at first coitus: A possible association. J. Biosoc. Sci. 11, 425-432.
Cutler, W.B., Garcia C.R., and Krieger, A.M. (1980). Sporadic sexual behavior and menstrual cycle length in women. Horm Behav. 14, 163-172
Cutler, W.B., Preti G., Huggins, G. Erickson, B., and Garcia, C.R. (1985). Sexual behavior frequency and fertile-type menstrual cycle. Physiol. Behav. 34, 805-810.
Cutler, W.B.,, Preti, G., Krieger, A., Huggins, G.R., Garcia, C.R., Lawley, R.J. (1986) Human axillary secretions influence women's menstrual cycles: The role of donor extract from men. Horm Behav. 20, 463-473.
MCCoy, N., Cutler, W., and Davidson, J.M. (1985). Relationships among sexual behavior, hot flashes, and hormone levels in perimienopausal women. Arch Sexual Behav. 14, 385-394.
Preti, G., Cutler, W.B., Krieger, A., Huggins, G.R., Garcia, C.R., and Lawley, R.J., (1986). Human axillary secretions influence women's menstrual cycles: The role of donor extract from women. Horm. Behav. 20, 474-482.