Some Effects of Unilateral Ovariotomy in Rabbits

This work was carried out with a view to extending research into the causes of fertility and fœtal atrophy in the domestic animals. During its course it was found possible to obtain data relating to the problem of compensatory hypertrophy of a surviving ovary, of the migration of ova, and of the alleged abnormal sex ratio after unilateral ovariotomy.

Mr Hammond kindly supplied me with a number of rabbits in half of which one ovary had been removed. These rabbits which had been operated on and used by him in his work on factors underlying the age of puberty were divided into three age groups. All were below the age of puberty at the time of the operation, and would not mate with the bucks when tried.

Series X, born on the same day; five controls, four operated on at the age of 22 weeks. In three the right ovary was removed and in one the left. In all cases the corresponding Fallopian tube was ligatured.

Series Y, 18 to 20 weeks old at the time of operation ; five controls, five with the left ovary removed and the corresponding Fallopian tube ligatured.

Series Z, 15-19 weeks old at the time of operation ; six controls, six operated. In this series besides the removal of the left ovary and the ligature of the corresponding Fallopian tube, part of the right ovary was removed. This was performed by removing a portion of ovary along its whole outer length, without touching the line of attachment of the ligament or that part of the ovary lying nearest it. The blood supply to the portion remaining was thus left as little impaired as possible. From one-third to two-thirds of the organ was removed in each case.

Every rabbit was placed with at least two bucks on the 20th or 21st of the month following that in which the operation was performed. If copulation did not take place, this was repeated on the 20th of each month until coitus occurred. The cages were examined at 10 A.M. each day, and any young which had been born were removed. The does were again mated on the fourth day after the removal of their young. In the few cases in which does copulated but did not become pregnant, recopulation took place on the thirty-fifth day after the previous coitus. All does were killed on the twentieth day of their fourth pregnancy. The feeding and housing conditions were kept uniform throughout the experiment; cages of operated and unoperated rabbits each containing one animal were mixed and not collected together in the classes of their occupants.

Direct determinations of the number of young following unilateral ovariotomy appear to be rare. Hunter ¹⁰  finds that the semi-spayed sow gives nearly as many piglings as the normal animal, while Bond ²  finds this to hold for the rabbit. Arai,¹  working on the rat, finds that semi-spayed females produce nearly as many young as do controls. He estimates this fertility by counting the corpora lutea, but his conclusion does not follow, as Hammond ^8,9  has shown that not all the ova shed develop to the stage of parturition. What Arai shows is that the fecundity is the same in operated animals and controls.

Table I. shows the results of our experiments on the size of the litters born:—

In the case of the X and Y rabbits there is a reduction in the size of the litter of 5.3 per cent, and 19.5 per cent, respectively, while in the Z rabbits, in which besides the removal of one whole ovary part of the remaining ovary was also removed, a decrease of 17.3 per cent, results. In no case is the reduction in the size of the litter commensurate with the proportion of ovarian tissue removed.

There are two possibilities before us which would account for this almost complete maintenance of fertility.

• (a) That semi-spaying does not affect the fecundity (i.e. the number of eggs shed) of the rabbits. In other words, that the number of eggs maturing at one time is not a definite percentage of the oöcytes present, but that a definite number of oöcytes mature irrespective of their total number.

• (b) That a reduced number of eggs is shed, but that a greater proportion of them develop to parturition.

In connection with (a) we have Pearl’s ¹⁴  observation that fecundity in the domestic fowl does not depend on the number of oöcytes present in the ovary. He finds no correlation between the number of visible oöcytes and fecundity as measured by the winter egg production of the fowls under observation. Bond ²  finds the number of corpora lutea in semi-spayed rabbits to be approximately the same as that in normal rabbits. Arai ¹  confirms this in the case of the rat, and adds the observation that the number of unripe follicles in the single ovary is a little less than that in the corresponding ovary of controls.

The results of our observations on this point are embodied in Table II.:—

These figures are taken from counts made on killing the rabbits on the twentieth day of the fourth pregnancy.

The number of eggs shed is determined by making a count of the recent corpora lutea, and does not allow for the possibility of some follicles discharging more than one egg. No figures are available concerning the frequency with which this occurs in the rabbit, but it appears to be very rare indeed.

The number of eggs lost is determined by subtracting the sum of the live and degenerate fœtuses from the number of eggs shed. In addition to eggs lost before implantation, it includes those which degenerate at an earlier stage than is visible at the twentieth day of pregnancy.

The two series X and Y are classed together as they represent animals which have undergone a similar operation. The average number of eggs shed by these rabbits show an increase of 0.9 per rabbit in favour of the operated rabbits, but the margin of significance is 0.2 only, which in the small number of cases under observation would not justify a statement that semi-castration has the effect of increasing the number of eggs shed.

In the case of the Z rabbits a decrease of 2.7 eggs with a margin of significance, after deducting the probable errors, of 1.6 per rabbit is shown on comparing operated animals with the controls. In other words, there is a decrease of 20.8 percent. in fecundity after the removal of one whole and a part ovary. As the birth decrease in these rabbits was 17.3 per cent., it follows that the combined loss by fœtal atrophy and loss of ova occurs on the same scale in operated and unoperated rabbits.

We are left then with the alternative which is confirmed by actual count of the corpora lutea, that removal of oöcytic material hardly affects the number of follicles which mature at any one time. The expression of fecundity then rests not on the oöcytic part of the ovary, provided that it be in a healthy condition, but on somatic conditions. It may be that nutrition can only be provided for the maturation of a certain number of follicles at one time.

Turning now to the fœtal atrophy figures we find in each series a significant rise from unoperated rabbits to operated ones. The combined figures show a rise from 0.4 ± 0.1 for unoperated rabbits to 1.4 ± 0.1 for the operated series. In the latter case all the fœtuses are crowded into one horn, for one Fallopian tube was ligatured and internal migration of ova did not take place. In view of the fact that nutritional conditions were the same for each series, we conclude that this extra crowding is responsible for the increased amount of atrophy. While this may be so under the conditions of this experiment, the increase in atrophy is so small with an approximate doubling of the number of occupants of the uterine horn that lack of accommodation can hardly be a factor involved in the atrophy of fœtuses in normal rabbits. The power of adaptation to the quantity of its contents possessed by the uterus is remarkable, and under normal conditions the uterus is not called upon to its utmost limit.

Taking the weight of each litter at the twentieth day of pregnancy, finding the average weight of the individual in the litter, and thence the average weight of fœtus for each series, we obtain a value of 4.65 gms. ± 0.19 for the fœtuses in operated rabbits and of 4.00 gms. ± 0.15 for those in controls. Thus, although it would appear that overcrowding causes increased fœtal atrophy, the size of the individual fœtus is not affected adversely; it shows, in fact, the increase usually associated with a somewhat smaller litter. The increase in fœtal atrophy then cannot be due to nutritive conditions, nor to differences in the inherited constitution of the fœtuses, for care was taken to keep external factors uniform for each series, and the rabbits were drawn haphazard from, as far as possible, similar genetic stock and mated with the same bucks. The increased weight borne in one horn may increase the traumatic dangers to which the fœtus is exposed; for example, it may cause strangulation of the blood-vessels which would lead to fœtal atrophy. Strahl and Henneberg ¹⁶  show that ligaturing the maternal vessels produces atrophy and absorption of the fœtus.

In 1906 Bond ²  showed that hypertrophy of the ovary takes place in the rabbit after unilateral ovariotomy. He considered that the corpora lutea were smaller in the ovaries of his operated rabbits than those in the ovaries of his controls, and concluded that there is some hypertrophy of the stroma. Carmichael and Marshall ⁸  showed that an ovary which is left untouched can, in the absence of the other, double in weight, while if only one-fifth of an ovary remains it can hypertrophy till it exceeds the weight of a normal ovary. This work was also performed on the rabbit.

Doncaster and Marshall ⁸  find hypertrophy to occur in the rat, but King,¹¹  in discussing her work on semi-spayed rats, says: “The remaining ovary appeared normal in every female, and there was no marked increase in size to compensate for the loss of the other ovary as Doncaster and Marshall find in the case of each of the two rats upon which they operated. As these investigators operated upon adult rats and killed them about two months after the operation, it seems probable that the noticeable increase in size of the remaining ovary must have been due to some pathological cause and not to a normal compensatory hypertrophy.” King seems to be the only worker who has failed to obtain this effect, and the hypertrophy cannot take place only in the rabbit, for Arai ¹  working on the rat finds that the weight of the remaining ovary is one and a half to two times that of the corresponding ovary in unoperated animals. He thus confirms the finding of Doncaster and Marshall. He finds that the increase in weight is due to an increase in the total number of well-developed normal and degenerate follicles, and to an excess of corpora lutea above the usual value. He finds little, if any, change in the stroma. It would appear that King’s rats were pathological rather than those of Doncaster and Marshall.

Lipschutz, Wagner, and Tamm ¹⁵  find that the remaining ovary of unilaterally ovariotomotised rabbits increases in weight to the extent of 55 to 80 per cent. as compared with the ovaries of controls.

The figures set forth in Table II. in this paper confirm the view put forward by Arai that there is a larger number of developed follicles in the unilateral ovary than in the normal organ (taking the number of corpora lutea as the criterion). In the case of the Z rabbits in which the number of corpora lutea was appreciably lower in the operated rabbits than in the controls, there was a smaller degree of hypertrophy. A varying amount of undeveloped follicular material and stroma had been removed from these ovaries however.

The effect of removal of varying amount of material is shown in Table III., in which the hypertrophy in Group Z is shown to be 20 per cent. of the normal. Here, however, from one-third to two-thirds of the remaining ovary was removed in individual cases, yet on hypertrophy the remaining tissue tended to reach a constant weight, more so, in fact, than the controls, the deviation from the mean being in each case 0. 23 and 0.39. This supports the view that the animal tends to build up a definite quantity of mature material whatever the reserve of this material may be.

An examination of stained sections of operated and control ovaries failed to reveal any signs of hypertrophy of the interstitial material, but in view of the scattered nature and irregular shape of this tissue no exact comparisons could be undertaken.

It has long been known that an egg shed from one ovary may find its way to the opposite uterine horn and there develop. Hammond ⁸  has shown this to be the case in swine and rabbits; his method being to count the corpora lutea and fœtuses on each side. On comparison of these figures more fœtuses were found on one side than eggs shed, showing, unless we consider this in every case to be due to the occurrence of polyovular follicles, that ova have crossed from one side to the other.

This is confirmed by the following records taken from the controls in this experiment :—

It would appear that the phenomenon is not by any means uncommon in the rabbit as it took place in 3 of 16 rabbits examined. It may have taken place in others, but where all the ova cannot be accounted for there is no means of telling.

Leopold ¹²  showed that this crossing may take place intra-peritoneally by removing one ovary and ligaturing the opposite Fallopian tube, when embryos developed in the uterine horn without a corresponding ovary.

That the cross-over does not occur in the rabbit by the passage of the ovum down one horn and up the other is evident from the present work in which the Fallopian tube corresponding to the ovary removed was ligatured. In no case of the 14 examined was an embryo found in the horn attached to the ligatured tube. It must be remembered that the uterus of the rabbit is bipartite and that internal migration would involve passage across the vagina and not merely from one horn to the other within the body of the uterus. The intraperitoneal route then is the only one possible in the rabbit for the passage of the ovum across the middle line, but this does not exclude the possibility of the intrauterine route in animals with bicornuate uteri.

Von Seligson ¹⁸ quoted by King ¹¹  removed the left ovary from two rabbits and bred them, on recovery, with a normal male, obtaining male offspring only. On removing the right ovary and breeding, females only were produced. Goenner ⁷  repeated this work and found that offspring of both sexes were produced in each case. He did not, however, prove by autopsy that no ovarian tissue remained on the operated side.

Dawson ⁵  suggested that this relation between sex of offspring and originating ovary holds in man.

Doncaster and Marshall,⁶  King ¹¹  and Copeman ⁴  showed that unilateral ovariotomy in the rat did not prevent the birth of young of both sexes in the same litter. They confirmed their findings by autopsy.

The following figures obtained in this work confirm Goenner’s results. No ovarian tissue was found on the operated side in any of the animals which were killed. Further, when sufficient numbers are compared, the sex ratio is similar to that of the controls.

I have to thank Mr J. Hammond, M.A., who kindly permitted me to experiment with the rabbits which he had operated on in connection with his experiments on the reproductive life of rabbits. His help and advice during the course of the work were unfailing and greatly appreciated. (The operations were performed at the Field Laboratories, Milton Road, Cambridge.)

My thanks are also due to Dr F. H. A. Marshall, F.R.S., for his advice and encouragement at all stages of the work.