Ogino's Discovery in his own words

Anthony Zimmerman
Studies of Human Corpora Lutea
Kyusaku Ogino, M.D.
Niigata Medical University, Pathological Laboratory
(Professor Kawamura, Director)
Vol. 38, No. 1, Feb. 20, 1923.
Translation by Fr. Anthony Zimmerman, 1980
Reproduced with Permission

[N.B.: This in an important paper, consisting of 92 pages and 9 charts, including 6 heliotype charts. Space does not allow us to print the entire paper; we re-produce here a part of the second section. The essential part of the historic paper is posted here because many have asked me about it.]


Ever since Von Baer discovered human ova (egg cells) in 1927, there have been many controversies among scholars about whether there is a relationship in time between ovulation and menstruation. The conclusions they have reached can be classified into the following categories:

  1. The time of ovulation corresponds to the time of menstruation either entirely or very nearly so.
  2. There is no fixed relationship of time between ovulation and menstruation.
  3. Ovulation occurs at some fixed time other than the time of menstruation.

The first of these opinions is the one which has received general acceptance up to the present time. The third opinion is a new theory of recent years. The second has always been existing alongside as an opposite view. But in the third theory, there are divergent views about the concrete problem of this fixed time. If we were to believe that all the views about this fixed time were factual, then we could not admit a fixed time. Further study is needed to learn which of the opinions holding to the third view is the correct one.

After this summary of the literature, I will now explain each of the views in detail. [Omitted.]


1. The Selection of the Subjects.

I selected subjects whose menstrual cycles were very regular from among the patients on whom I performed ventrotomy during the more than three years of time, May 1919 to November 1922. (Total number of subjects, 65.)

2. Observation during Operations.

When performing ventrotomy on these patients, I always exposed the ovaries on both sides very carefully after incising the abdominal wall. I then very carefully observed the condition of the follicles and of the corpora lutea, to ascertain whether ovulation had occurred, or whether not. Then I proceeded with the intended operation.

In case I would perform a hysterectomy, I usually also excised the appendage of the side which contained the corpus luteum, if this caused no harm, leaving the appendage on the other side in place. Then, immediately after the operation was finished, I re-examined the excised ovary to observe how the mechanical stimuli connected with the operation influenced the follicles and the corpora lutea. The comparative observations made before the excision of the ovary and after yielded the following data:

For the reason given above, when one is investigating the problem of time relationship of ovulation, if an investigator studies only stored specimens and does not make observations during operations, he does not exclude confusion arising from artifacts and so it is not possible to avoid errors; that is, if he does not make observations during operations, to examine how the corpora lutea in the initial proliferative phase are distributed in relation to menstruation, or how they line up in relation to the cycle of the uterine endometrium, and so - forth-,in his effort to estimate the time of ovulation. Therefore I took care to avoid these mistakes by making the observations during operations.




Employing the method indicated above, I proceeded to investigate the following problems:

In this way I endeavored to solve the problem about the time of ovulation.


Ex. 1 Name:
Age: 30 years
Childbirth: 5 times
Menstrual cycle: 45 days
Last menstruation, five days, beginning Oct. 5, 1919.
Diagnosis: Adhesive retroflexion of the uterus and endometritis.
Date of operation: Nov. 15, the 42nd day after, and 4th day before.
Kind of surgery: Alexander operation.
My views: The endometrium is in the pre-menstrual phase; the corpus luteum is in the development stage.
Remarks: The menstrual cycles of my cases were all constant, so I omitted recording in each case "menstrual cycle constant." I noted only the length of the menstrual cycle. The __th day after, and the__th day before refers to the number of the day after the last menstruation and the number of the day before the expected next menstruation.

The cases in which I noted only that in my view the corpora lutea were in the development stage (hatsuikki) are those in which the corpora were observed only macroscopically. When I made an examination through a microscope, I noted each phase specifically of the corpora lutea. [Rest is omitted.]


1. Time relationship between ovulation and menstruation.

(Impressions gained through the operations)

It is to the great merit of Fraenkel, Ruge II, and Schroeder that they demonstrated that ovulation and menstruation do not occur simultaneously. In addition to other proofs, the cases presented in charts 2 and 3 prove this clearly. There is no need from now on to debate about this problem any more.

The problem which remains is whether ovulation occurs with extreme irregularity in relation to menstruation as Ricker and Dahlman, Schichele and Iijima claimed; or whether it occurs within a limited and fixed time in relation to menstruation, as Fraenkel, Ruge II, and Schroeder concluded; and if the latter theory corresponds with the facts, then which of the of the conclusions of the three gentlemen is the correct one.

In order to recapitulate the findings: of the 65 cases, I calculated the amount of days lapsed since the beginning of the previous menstruation, following the method of Fraenkel, Ruge II and Schroeder, and arranged the cases in this order. On Chart 2, 1 recorded my findings at the time of the operation, indicating whether ovulation had occurred by the time of the operation; and if so whether I found a corpus luteum in the stage of development; if ovulation had not yet occurred at the time of the operation, this was recorded. I arranged them in the order of the number of days lapsed since the beginning of the last menstruation; it was a tentative summary of the cases.

Chart 2 shows that: I found that all cases before the 11th day after the onset of the last menstruation were not yet ovulated; and that there was not a single case of a developing corpus luteum. This finding does not agree with the conclusion of Ruge II that ovulation occurs during the time immediately following the onset of menstruation until the 14th day after.

From the 12th to the 28th days after the onset of menstruation I found both cases, namely those not yet ovulated and those with developing corpora lutea. At first glance this finding seems to agree with Fraenkel's theory; namely if we follow his theory that ovulation takes place after approximately 18.9 days in average, and consider that in my cases the average is also close to 18.9 days.

It is clearly illustrated in Chart 2, however, that the length of the menstrual cycles of my examples ranges from between 23 days to 45 days. Similarly, the examples of Fraenkel include a range from two and a half weeks to five and a half. Moreover, his individual cases show cycles which vary up to a week in length; that is to say, their cycles are quite irregular. Now I think that everyone will agree that in a situation like this, when you are using all kinds of cycles as data, with a range of variation of ovulation from the 12th day to the 28th day, it just doesn't make sense to say that ovulation occurs around the 18.9th day in average.

3. Schroeder's theory is that ovulation occurs on the 14th to 16th days in cycles of the 28 day pattern.

My cases with 28 cycles agree with this theory. But Schroeder gave no clear explanation about the time of ovulation in cycles which fall outside of the 28 day pattern.

In his book CYCLICAL CHANGES OF THE UTERINE ENDOMETRIUM Schroeder wrote that in cases where the menstrual cycle is constant and is not the 28 day type, each phase of the segmented cycle of uterine endometrium expands or shortens in proportion to the length of that menstrual cycle.

He also wrote in the paper "'Time Relationship Between Ovulation and Menstruation" that each single phase of the corpus luteum is closely related to the cyclical changes of the uterine endometrium.

We might theorize from these two facts that if ovulation occurs from the 14th to the 16th day in the cases of 28 day cycles, namely right in the middle of the cycle, then in cycles which fall outside of the 28 day pattern, ovulation might also occur during the 3 days which are in the middle between the menstruations.

If we look at Chart 2 with this in mind, we can interpret it to mean that ovulation occurred in the middle of the cycle, if we make an exception of case No. 3.

In the cases, however, where the cycle is short, corpora lutea are found a bit ahead of the dividing middle line; and in cases of long menstrual cycles beyond 30 days, one finds examples where ovulation has not yet occurred after the middle dividing line. It is even possible to infer that the longer the cycle is, the later are the cases which have not yet ovulated after the middle line. Admittedly the tendency is a slight one, but it is one which we must take into consideration.

4. Finally, when comparing my opinion with the interpretation of Iijima, the fact that I found both pre-ovulation cases and corpora lutea from the 12th to the 18th days of the cycle, does not necessarily lead to the conclusion of Iijima that ovulation is something which occurs in a quite irregular pattern.

As mentioned above, it is possible to interpret the time of ovulation in various ways from the arrangement of the cases in CHART 2.

The question to be solved before we even begin to discuss about the relation of ovulation and menstruation is the following: to which of the menstruations is the ovulated cell, or the corpus luteum which develops after ovulation, closely related? Is it to the menstruation which preceded, or the one which is to follow? We should keep in mind:

  1. The fact of the occurrence of pregnancy without having even once seen menstruation; or of pregnancy without a preceding menstruation in the post-partum, which shows that a previous menstruation is not necessary for the occurrence of ovulation.
  2. In accordance with the views of Schroeder and Ruge II, (and I agree with them), the pre-menstrual stage of the uterine endometrium corresponds to the proliferative phase of the corpus luteum.
  3. After I had excised corpora lutea in the development stage, menstruation occurred within 3 days after the operation.

Such facts indicate that the ovum or the corpus luteum are etiologically related to the menstruation which is to come, and have no relation whatsoever to the preceding menstruation, the view which is now quite generally accepted. So when we discuss about the time relationship between ovulation and menstruation, it makes good sense to study the kind of connection which exists after ovulation with the coming menstruation; whereas to look for some kind of connection with the menstruation which has gone before, although there really is no connection with it, is unreasonable.

If, however, one limits the discussion to only cycles which have the 28 day pattern, as Schroeder did, then the result must come out to the same thing.

But if you take the materials such as Fraenkel and Ruge II used, in which the various cycle lengths differ from each other, then I am not able to make a calculation about the time of ovulation in relation to the preceding menstruation. Herein lies, I believe, one reason for the assortment of different conclusions drawn by various researchers. And for me it is not possible to come to a conclusion about the time of ovulation from the way the examples are lined up in CHART 2.

In CHART 3, therefore, I calculated the time corresponding to the number of days between the day of operation and the expected day of onset of menstruation. On this standard I then lined up each case in order to see whether some kind of connection can be discovered between ovulation and the expected menstruation.

My views about the data of CHART 3 is as follows:

  1. From the first day of the foregoing menstruation until the 17th day before the expected menstruation, follicles have not yet ovulated in all the cases, with the exception of cases No. 2 and No. 10. (No. of cases 21; exceptions ,2)
  2. During days 12 to 16 before the expected menstruation - 5 days - some of the follicles were ovulated, some were not. (No. of cases: 13)
  3. From the 11th day before until the onset of the expected menstruation corpora lutea were found in all the cases. There was not one case in the pre-ovulatory phase. (No. of cases:31)

I draw the following conclusion from the above views: The time of ovulation is during a span of 5 days from the 12th to the 16th days before the onset of the expected menstruation. This holds true throughout all the cases, whose menstrual cycles vary in length from 23 to 45 days. ...[Dr. Ogino then explains about the two exceptions, and thereafter draws on a detailed description of his cases to confirm the credibility of his theses.]

Fr. Zimmerman Comments

Ogino's 1923 paper published in Japan, followed by his visit to Germany in 1929 where he published his theory in German, where Dr. Hermann Knaus published a similar conclusion in 1929, is a monumental turn in human history. The publication of these papers made possible the scientific study of natural family planning.

The charts which Dr. Ogino mentions in the above writing can be posted only in photo format which would take some time to be accessed on this electronic page. The charts can be found in printed form in my book Catholic Teachings on Pro-Life Issues, which is available at HLI or One More Soul.

The long and short of the charts is that the day of ovulation is in a time zone related to the expected menstruation rather than to the previous menstruation. It may be presented as follows:

Short cycle of 23 days, counting days from the first day of menstruation:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23***

Cycle of 45 days:

1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45***

The possible days of ovulation (in bold letters) are consistently 12-16 days before the next expected menstruation. This is the heart and soul of scientifically based natural family planning. (But some women have consistently longer times, or consistently shorter times, between ovulation and the expected menstruation.)

If you know about when to expect the next menstruation, you can mark on the calendar the expected time of ovulation. Today's NFP is advanced over the Ogino-Knaus calendar method, observing the temperature shift, the mucus peak day, and other signs. But it was Dr. Ogino who first discovered the constant time relationship between ovulation and the expected subsequent menstruation. ***