The author comments:

This paper was composed as an exercise in criticizing the design of experiments. You may wish to test your scientific acumen by identifying as many fallacies as you can – then compare these with your favorite experiment in software engineering.


Recent work on contraception has begun to shift from exclusive attention to the female and to concentrate on the male role in conception and its prevention (1). The greater part of this work, like that on female methods, has been handicapped by an implicit dedication to the “sperm-egg theory” of conception (2). Since the sperm-egg theory has never been validated in human beings by direct observation of the supposed fertilization of egg by sperm, it would seem more prudent for workers in this field to confine themselves to the terminology, “sperm-egg hypothesis.” We shall use this terminology in this report. We will describe a series of experiments that that seriously challenge the correctness of this hypothesis . A great breakthrough in contraceptive technique awaits those researchers with the courage to discard antiquated scientific “paradigms” (3).

The sterility (as it were) of the sperm-egg hypothesis can be seen by tabulating the number of studies in a twelve-month samplings of articles in the Journal of Reproductive Fertility and similar journals which concern themselves with methods of contraception in the male and female human. In Table 1, we see the results of this tabulation broken down by male-female as well as by the time of application of the method – “before,” “during,” and “after.”





















Table 1. Fashions in
Research: The number of reported studies on various methods of

The usual bias of female over male studies is shown, but our interest here is in the entry “male/after” – the only entry for which there are no studies at all. This startling lacuna in the research interests of so-called scientists can be directly attributed to their blind devotion to the sperm-egg hypothesis. Under this dubious hypothesis, once conception had taken place within the female, the male would be unable to affect successful carriage of the embryo to term. Conversely, any evidence that “male-after” contraception was effective would prove disastrous to the sperm-egg hypothesis – which explains, perhaps, why the scientific establishment has so carefully avoided this promising area of research.

The Screening Experiment

In recent years, it has become standard practice for health researchers to bemoan the difficulties introduced into their work by restrictions on the use of new drugs on human subjects. We find these complaints unjustified, for with suitable scientific imagination and creativity, the researcher can design safe experiments which are unaffected by these restrictions. Since the restrictions are on the use of experimental drugs on humans, they may be circumvented by either:

1. using experimental drugs on animals
2. using non-experimental drugs on humans

For the initial screening, method 2 was used by making systematic search among drugs sold without prescription in a sample of 385 volunteer married couples. Since the risk of pregnancy was great in such a screening of drugs previously untested for contraceptive efficacy, subjects were informed of the possibilities and dropped from the experiment if they objected to the possibility of becoming pregnant.

Each couple in the experiment was given a supply of the drug they were to screen, with instructions as to use taken from the normal mode of use, but with the name of the drug removed. The husband was instructed to administer on standard dose to himself on the morning following each incident of intercourse. Multiple incidents of intercourse in one evening were to be followed by only a single dose (4). Incidents taking place during daytime hours were to be followe by a dose administered immediately after the next succeeding sleep period. In this way, no subject administered more than one dose per 24-hour period, regardless of the frequency and timing of incidents of intercourse.

At the end of one year, the number of pregnancies was tabulated. It was found that 34 out of the 385 couples had failed to have even one conception. This gave 34 non-prescription drugs for effective “male-after” contraception. Of these, 29 were in pill form, 3 were powders to be dissoved in water, and 2 were suppositories. The suppositories were from the European pharmacopia, and it was found that American male subjects were reluctant to use this mode of administration. In fact, interviews with the two subjects in question indicated that they had not actually used the suppositories at all. Therefore, the suppositories were eliminated from the follow-on study, leaving 32 candidates in all.

In the follow-on study, 160 (32 x 5) subject pairs were used in the same experimental design, to eliminate those drugs which might have resulted in zero pregnancies by chance. The number of pregnancies per drug were tabulated, with the results shown in Table 2.

Number of Pregnancies

(5 possible)

Number of Drugs

Giving This Number

of Pregnancies













NTable 2: Secondary Screening Results of 32 “male-after”

(Mean number of pregnancies = 0.5/couple)

As can be seen in the Table, one drug was 100 per cent effective at preventing pregnancy. (A second drug was 100 per cent effective at causing pregnancy – this drug will be the subject of a separate study on male morning-after fertility drugs.)

The effective drug was a chocolate-covered chewable pill known by the trade name “Ex-Lax.” Rather than attempt to isolate a single active ingredient in this multi-component drug, we decided to proceed immediately to a full-scale experiment on its contraceptive effectiveness.

The Controlled Experiment

Because of the difficulties in finding further non-pregnant subject pairs in a small Midwestern college town, the controlled experiment had to be performed using a different subject group. Because of the effectiveness of the Ex-Lax in the screening tests, however, we were confident that the risk of unwanted pregnancy was small – a prediction that was borne out by the eventual results. For subjects, we selected 54 Freshman college students from a single fraternity. Each of the subjects had a “date” for the annual fraternity homecoming supper and dance, and these dates formed the other half of the the experimental couples.

The morning after the dance, each male subject was interviewed and asked the question, “Did you have one or more incidents of sexual intercourse with your date sometime in the previous 24 hours?” Where necessary, the question was rephrased into various vernacular forms more familiar to the subject population. Eventually, affirmative responses were obtained from 53 out of 54 subjects, and these 53 subjects were then instructed to take orally one (1) Ex-Lax tablet (unlabeled).

The Control Experiment

As a control in this experiment, the use of human subjects was out of the question, since without the administration of any contraceptive whatsoever, the risks of pregnancy were finite. Therefore, mice were used as controls. In order to obtain an exactly equivalent number (53) of matings, a male mouse was placed in a cage with a female mouse and observed until mating took place. If no matings were observed during and eight-hour period, the mice were separated and another pair was used. This procedure was repeated until 53 mating pairs were obtained – a total of 82 trials in all, as shown in Table 3.

Experimental Group




Without Ex-Lax (mice)




With Ex-Las (men)




Table 3: Controlled Experiment Results on Effectiveness of
“Male-After” Pill

Since the mice were controls, they were given no Ex-Lax after mating. As a result, 46 pregnancies resulted out of 53 matings, as compared with none in the human subjects in the same number of matings. The effectiveness of the non-Ex-Lax treatment, therefore, was only 13.2 per cent, as shown in the Table (5) The effectiveness of the Ex-Lax “morning-after” treatment is seen in the Table to be 100 per cent, which ranks it with the most effective of the female contraceptives, such as estrogen pills and abstinence.


There are two major conclusions that can be drawn from this experiment, one theoretical and one practical. First, in the light of these results, the sperm-egg hypothesis, so long held as central dogma in biology, is no longer tenable – at least for human beings. In a way, this profound result is not surprising, since we should long ago have learned that results obtained from animal experimentation are seldom directly applicable to human beings. In the light of these paradigm-shattering results, what new theoretical paradigm must be constructed for human reproduction? The answer to that question would take us beyond the scope of this paper, and will be the subject of a forthcoming review (6).

The practical implications of our experiments are much clearer. Naturally, other researchers will want to corroborate our results, but in view of the absolute effectiveness of the treatment, it is unlikely that any serious divergence will be found. Since the method involves the use of a drug that has been extensively field-tested and approved for dispensation without prescription, we can se no reason – except for possible male-chauvinistic reaction – for withholding our method from the general public.

Naturally, we are continuing our studies, particularly the attempt to isolate the active principle in the drug and to derive its full chemical structure and mode of action. By isolating the single active principle, we hope to be able to improve the drug by eliminating its one slight side effect.


(1) Jackson, Harold
Chemical Methods of Male Contraception
American Scientist, Vol. 61, pp. 188-193

(2) See, for example,
Pincus, G
Control of Fertility
New York and London: Academic Press

(3) Kuhn, Thomas
The Structure of Scientific Revolutions
Chicago and London: The University of Chicago Press

(4) Professor Dennis Geller (personal communication) has questioned the reliability of this procedure, but it has the distinct advantage over other regimes in that it can be correctly applied even by subjects who cannot remember (due to consumption of alcohol, etc.) how many incidents of intercourse took place.

(5) Professor Thomas Plum (personal communication) has pointed out that this may not be the proper figure for the effectiveness of the method. He suggests that the proper denominator is not the number of actual matings but rather the number of mating opportunities. His reasoning is that any reduction of libido consequent to the particular method is also a factor in its reproductive effectivenss. We believe, however, that any method, such as our control, that produces a marked reduction in libido in human males would be unsatisfactory as a contraceptive technique to most of the population.

(3) Weinberg, G. M., and B. Huntington Snark
The Homunculus Theory – A Case of Scientific Character Assassination?
(in Press)

Note by author, in 2001: I am republishing this slightly revised version of the classic 1973 study, perhaps to overcome the prejudice of the scientific establishment, who for three decades have studiously avoided any mention of these experiments, let alone conducted the follow-up experiments so obviously called for.
Because of this slight by the biology community, I have switched my own field of research to software engineering, where this study has served as a model for so many other studies.