1. Methods and experiments are described for testing the length of fertility of the sperms in the male and female tracts of rabbits.

  2. The sperms may retain their fertility in the male tract up to 38 days, but in the female tract only up to 30 hours.

  3. In cases where the sperms had remained in the female tract for 24-30 hours before ovulation small litters were frequently produced; it is suggested that this is due to the end point of the vitality of the sperms falling within the limits of time during which the process of ovulation occurs.

  4. The bearing of these results on problems of fertility and sterility is discussed.

The fertility of a mammal depends first, on the number of ova which are shed at each heat period, secondly on the number of these which become fertilised, and thirdly on the number of fertilised ova which develop to birth. The second of these factors is (except for pathological conditions, such as occlusion of the Fallopian tubes, etc.) mainly one of time relations between the life of the ova and of the sperms. The duration of fertility of the ova has already been dealt with(i), it being shown that they are only capable of being fertilised for about 2–4 hours after they leave the follicle. The present paper is confined to an account of the life of the sperm.

A determination of the length of fertile life of the sperms, both in the male and female tracts of the rabbit, has been made. It must be stated at once that the main point treated is that of length of fertility rather than that of motility. Work with Mr A. Walton on the length of life of the sperm outside the body has shown that although sperms under the conditions of these experiments may still be motile, yet their fertility has been lost, a point which was indicated by Wolf’s (2) results, and accordingly a distinction has been made for the purposes of this paper. This distinction appears to be justified, for further evidence to be described indicates that under certain circumstances the sperms drawn from the epididymis of a buck, although motile, are infertile. While much work has been done on the length of time of motility the practical aspect of fertility appears to have been neglected.

A study of the length of fertility of the sperms in the reproductive passages is a necessary preliminary in any consideration of the conditions under which artificial insemination may be practised. The length of life in the male tract gives an indication of the order of longevity which must be aimed at under the best conditions outside the body ; while, as ovulation in most species occurs at a definite time in the reproductive cycle, an important factor in success must be the capability of the sperms to live until ovulation takes place. In other words, we require to know the length of time in which successful insemination may occur ; thus Anderson (3) states that owing to the short life of the sperms the time relations between insemination and ovulation is important in mares, and Ivanoff (4) finds that conception as a result of a mare being inseminated outside “heat” is exceptional. In most species copulation is allowed only at the time of oestrus, immediately before the occurrence of ovulation. In those cases in which there is a prolonged oestrus, or in which, as in man, copulation occurs at any time in the cycle, fertility as measured by the percentage of copulations which are successful is low. This may be explained on the assumption that, if the sperms live in the female tract for a limited period only, copulations which occur earlier than this with reference to ovulation are sterile.

The rabbit has been chosen as the experimental animal in this investigation for several reasons. The most important is that the doe remains on heat for long periods and usually ovulates only on coitus, ovulation succeeding coitus by an average of 10 hours (Heape(5)); ova may consequently be produced at any definite time as required. Also as the process of reproduction in the rabbit has been worked out in detail (1) results could be obtained with greater certainty than if a fresh species were used. In addition, the rabbit has the advantage that it breeds rapidly, mates under observation and is comparatively cheap to maintain.

Methods

In order to determine for how long the spermatozoa remain capable of fertilisation in the male tract, ligatures and division of the middle part of the epididymis between the upper and lower poles were made (see Diagram I) on each side through an abdominal incision along the linea alba. In a few of the earlier experiments no special precaution was taken to withdraw the testes to the scrotum from the abdominal cavity after the operation and in most cases they returned naturally, but as one case occurred (F 215) where the testis was retained in the abdomen by adhesion to the fatty tissue surrounding the bladder, and especially as in this particular case the buck was sterile when mated, the testes were replaced in the scrotum after the operation in all subsequent experiments. Care was taken not to injure or ligature any blood vessels supplying the testes or epididymis during the operation, for damage to these soon causes marked degeneration of these organs.

Diagram I.

Showing method of operation on bucks. L, L, ligatures; C, cut; T, testis; EH, head of epididymis; ET, tail of epididymis; V, vas deferens.

Diagram I.

Showing method of operation on bucks. L, L, ligatures; C, cut; T, testis; EH, head of epididymis; ET, tail of epididymis; V, vas deferens.

By this operation the supply of newly-formed sperms from the testes was cut off and the length of time that those already formed at the date of operation, and stored in the lower pole of the epididymis, could remain capable of fertilisation was tested by mating such bucks at definite intervals after the operation with does on heat. The does used for this purpose were carefully selected and for the most part were those which had produced a litter a few days previously and the young of which had been removed as soon as they were born, or those which were at the end (after the 16th day) of a pseudo-pregnant period. Occasional cases of sterility from adherent Fallopian tubes do, however, sometimes occur, but does so suspected have been mated to normal bucks afterwards and where they failed to produce a litter a post-mortem was made, the animal being marked as probably sterile back to the time of the last known litter. No animal already under suspicion of sterility was used in any of the experiments. The tattooing of numbers in the ears of all rabbits and the systematic keeping of all breeding records enabled the writers to trace the history of all animals, in most cases since birth.

The bucks used were tested for fertility by mating with does before they were operated on. The majority were bucks which had been reared by the writers and were about 9–11 months old, well over the age of puberty, which ranges from about 4 (strain F) to 6 (strain E) months. In addition at the critical periods, older bucks which had been successfully used as stud animals were also operated on. On the whole the duration of life of the sperm in these older bucks was rather less than in the younger ones.

In order to ensure a good supply of sperms in the epididymis the bucks were usually not allowed to serve any does for 3 days to a week before the operation, while during this period and between the time of operation and mating they were especially well fed and kept in good condition to lower the chances of absorption of sperm occurring.

In most cases the buck was mated once with the doe on the first occasion on which he was used after the operation, on the second occasion twice and on the third occasion thrice. The purpose of the two copulations on the second occasion was to ensure if possible that the does should not be sterile through lack of sufficient semen, the total amount available in these bucks being limited by the supply stored up in the lower poles of the epididymis before the operation. In this way an attempt was made as far as possible to equalise the amounts of semen ejaculated in each of the first three does with which the buck was mated.

In a few cases (E 370, σ) where trouble was experienced in getting the buck to mate at the appointed time it was killed and does were inseminated from the fluid contained in the lower poles of the epididymes diluted with Ringer’s or Walton’s fluid (buffered isotonic salt solution mixed with sugar).

Fertility

Table I shows the detailed results obtained from a number of such experiments. When the first results of the spermatozoa living as long as 17 days after the operation were obtained it was at first doubted whether the connection between the epididymis and testis had been effectively cut off (see bucks C 224 and F 214); the buck rabbits were therefore again bred to normal does to test this point and having produced no litters were killed and examined by inspecting the fluids in the vas and lower pole of the epididymis for spermatozoa. There could be no doubt, however, that the operation was successful, for while there were no sperms in the lower pole of the epididymis (at 103 and 126 days after operation) the upper pole was packed with motile sperm. Subsequently when the motility of the sperm was being tested (see below) this fact was repeatedly confirmed.

Table I.

Life of spermatozoa in the male tract. Size of litter obtained from does with which mated.

Life of spermatozoa in the male tract. Size of litter obtained from does with which mated.
Life of spermatozoa in the male tract. Size of litter obtained from does with which mated.

The cases (bucks C 321 and C 272) where the buck was not fertile at.the first or second mating but was so subsequently are possibly to be explained by lack of ovulation by the doe (which may occasionally occur) or to ejaculation of semen mainly composed of vesicular fluid, which may happen in an easy ejaculation taking place after a prolonged sexual rest where a lot of this secretion has accumulated (see F 375, Table III). In the summary of results given in Table II these cases have been classed as potentially fertile at the first mating.

Table II.

Life of sperm in male tract. Summary of Table I.

Life of sperm in male tract. Summary of Table I.
Life of sperm in male tract. Summary of Table I.
Table III.

Observations on motility of sperm in male tract.

Observations on motility of sperm in male tract.
Observations on motility of sperm in male tract.

The summary shows that normal fertility is maintained up to about the 20th day after the operation, 93 per cent, of the bucks (excluding one in which the testes remained in the abdominal cavity) tested being fertile up to the 20th day. After this time the fertility begins to fall off so that between 21 and 30 days only 36 per cent, of the bucks tested were fertile, while between 31 and 40 days only 20 per cent, gave litters. The detailed account given in Table I shows that there is considerable individual variation between bucks as to the length of time the sperms remain fertile, the causes for this, however, have not been specially investigated, although suggestions have been made in the discussion below.

Since, as will be shown below, the life of the sperms in the female tract is very short and the sperms after their life in the male tract have to exist in that of the female for 10 hours (until ovulation occurs), it was thought that it might be possible to obtain litters from these operated bucks beyond the time normally possible by reducing the time the sperms have to exist in the female tract (see Diagram II). This was tried in three cases (E 505, X 544 and E 559, see Table I) by mating the does with a sterile buck 8 or 612 hours before they were mated with the operated buck, i.e. the sperms were inserted 2 or 312 hours before ovulation, the latter normally occurring in the rabbit 10 hours after coitus. In the first two cases, at 2 hours before ovulation, no result was obtained, but possibly the interval was too short for such attenuated sperms to reach the ova in the time. In the third case, however, at 312 hours before ovulation the attempt was successful and a litter of five was obtained 40 days after the buck had been operated on, although he was sterile when mated normally to does both before and after this time. Such a single result wants confirmation but it suggests that attenuated sperms have more chance of fertilising ova when the interval between insemination and ovulation is short, provided sufficient time is given for them to reach the upper ends of the tubes.

Motility

In the case of some of the bucks which were killed in order to inseminate does (see Table I) from the fluids in the lower poles of the epididymis motile sperms were found (E 370 at the 25th day and B 1 at the 29th day) although the does so inseminated failed to produce young, the does of course having been mated to a sterile buck just before insemination so as to cause ovulation. After this fact was ascertained the motility of the sperms was watched by collecting a drop of fluid from the vulva of the doe after each mating made by the bucks and afterwards by carefully examining the fluids of the lower epididymis when the bucks were killed. Table III gives the results obtained, the size of the litter produced (when mated) being shown by the figure in brackets. The last observation in the case of each buck was obtained from examination of the lower epididymis at death while the others represent observations on drops obtained from the vulva of the does with which he was mated before this time.

The facts given in this table show that the sterility of the bucks which begins to appear after the 20th day following the operation was not due to exhaustion of the supply of sperms in the lower pole of the epididymis, as in every case up to 60 days some sperms were found in this position when the buck was killed. In a few cases sperms could not be observed in the drop obtained from the vulva of some of the does after service (bucks F 375, E433, E 480, see Table III) and it is possible that in these cases the vesicular fluid formed the large bulk of the ejaculation ; in these same bucks, however, sperms were seen in other ejaculations and were present in the lower epididymis when the bucks were killed.

The fact that the motility of the sperms persists long after they have ceased to be capable of fertilising ova is also shown conclusively by Table III. It has been shown above that fertility diminishes greatly from 20 to 40 days after the operation, the motility of the sperms, however, does not begin to decline perceptibly until after about 40 days, as will be seen from the summarised account shown in Table IV; after this time (40 days) the proportion of the bucks containing only dead (immotile) sperms rises considerably, although in one case motility persisted up to 60 days. In bucks which had copulated frequently the sperms had all disappeared from the lower epididymis from 91 to 357 days after the operation.

Table IV.

Motility in male tract. Summary of Table III.

Motility in male tract. Summary of Table III.
Motility in male tract. Summary of Table III.

The writers are indebted to Mr A. Walton for many observations on the histology of the sperms during this period which will be published by him, but it may be said here that many were found in the bent form before their motility was lost completely, the tail being bent over and lying almost parallel to the head so that they tend to wander round in circles instead of progressing in a straight line.

As Table III shows, there is great variation between different bucks as to the time at which the motility of the sperms is lost, just as there is with fertility. It would appear that the sperms produced by some bucks are deficient in vigour as compared with those of others and that in extreme cases this may lead to sterility in normal matings, although motile sperms may have been observed in the semen of such animals.

Size of litter

A point of interest in connection with the life of the sperms, as it passes from full fertility to sterility, is whether on the border-line between these two phases a condition exists in which small litters are produced by does mated with these bucks and whether foetal degeneration may be caused in this way.

Owing to the way in which the experiments were planned it is necessary first to consider the effect of the order in which the does were mated with the buck; as already mentioned the doe first mated usually copulated once, the second twice, and the third thrice, to ensure as far as possible equal numbers of sperms being injected into each doe. As in the first few cases (see Table I) investigated the bucks did not produce young when mated subsequently to the first six matings, few attempts were made to continue mating beyond this and six matings to three does made in this way were arbitrarily taken as a standard number in each case. The last column in Table V shows that there is no reduction in size of litter with the first three does so mated. As the does to which the bucks were mated belonged to several different strains, which varied in fertility, each strain has been averaged separately.

Table V.

Fertility of does mated with bucks which had been operated on. Size of litter produced.

Fertility of does mated with bucks which had been operated on. Size of litter produced.
Fertility of does mated with bucks which had been operated on. Size of litter produced.

Having shown that the method of mating adopted did not produce small litters, the original problem may now be considered. From the facts given in Table V it would appear that there is some indication that small litters are produced on the border-line between full fertility and sterility, due to the attenuation of the sperm as a result of age. The number of litters obtained at this time, however, are from the nature of the case few, but the results are similar in the case of each strain, although naturally more evident in the more fertile strains. That the small litters which occur at this time are not due to lack of ova produced by the does is shown by the following two cases in which the doe was killed during pregnancy :

  • Buck E 429 was mated on the 26th day after the operation with doe E 335 ; this doe, killed during pregnancy, contained two normal foetuses and ten corpora lutea of pregnancy, showing that although ten follicles had ruptured, only two ova had developed.

  • Buck E 429 was mated on the 28th day after the operation with doe E 416; this doe, killed during pregnancy, contained six normal foetuses and eleven corpora lútea of pregnancy.

In the rabbit in normal matings there are often more corpora lutea than foetuses during pregnancy, but the difference is rarely so wide as this, the average of missing ova being under 24 per cent. (6). The possible causes of small litters are referred to below.

Sex ratio

The sex ratio of the young produced as the result of matings made after the sperm had existed for different lengths of time in the epididymis is given in Table VI. After the 20th day, when the fertility was falling off, the percentage of females was marked, but the proportion of males born in any one litter is notably very variable, so that data based on a small number of litters cannot be very reliable.

Table VI.

Life of sperm in male tract. Sex of young produced.

Life of sperm in male tract. Sex of young produced.
Life of sperm in male tract. Sex of young produced.

Methods

It has been shown in a previous publication (1) that spermatozoa in the rabbit do not retain their fertilising power for 4 days in the female passages. This was determined by making a mating before the 16th day of pseudo-pregnancy (for the rabbit will, if well nourished, often copulate during this time) when large corpora lutea are present in the ovary which prevent ovulation occurring after coitus ; a second mating was then made with a (sterile) vasectomised buck at the 20th day, at which time the corpora lutea have atrophied sufficiently to allow of ovulation occurring. Owing to variations in the exact day (between 15 and 19) on which the corpus luteum atrophied in different individuals (just as the duration of the oestrous cycle varies in other species) it was not possible by this method to obtain reliable data for the life of the sperm in the female passages for periods of under 4 days, the reason being that ovulation might either occur at the first coitus with a fertile buck or at the second coitus with a sterile one.

A further attempt to solve the problem was therefore made by the use of rabbits showing dominant and recessive Mendelian colour differences. Does of a recessive (albino) colour were used; these were made pseudo-pregnant by copulation with vasectomised bucks and at various times from the 13th to 16th day of pseudopregnancy were mated with homozygous bucks of a dominant colour (agouti) and again on the 20th day of pseudo-pregnancy with bucks of the recessive (albino) colour (see Table VII). Should ovulation occur at the first coitus the young would all be agouti, but should it not occur until the second coitus, then, if the sperms from the first mating lived, there would be a chance of getting young of both colours in the litter, or if they did not live, then only albino young would be produced.

Table VII.

Life of sperm in female by means of double matings (with colour differences’) during and at the end of pseudo-pregnancy.

Life of sperm in female by means of double matings (with colour differences’) during and at the end of pseudo-pregnancy.
Life of sperm in female by means of double matings (with colour differences’) during and at the end of pseudo-pregnancy.

Consideration of the duration of pregnancy furnished a further check on the date of ovulation, normal does of this strain going 31–32 days from the time of mating (see Controls, Table VII).

In the first series of animals (see Table VII) the time of the second coitus was kept constant at 20 days and the first coitus was varied from 13 to 16 days, but no litters of mixed colours were produced. In the second series the time of the first coitus was kept constant at 15 days and the time of the second coitus was varied from 16 to 17 days ; here again no litters of mixed colours were produced. A number of control animals in which no second mating was made gave data as to the average duration of pregnancy in this strain and also showed that ovulation might occur as early as the 13th day or fail to occur as late as the 17th day. This variation at which the time of ovulation might occur therefore gave a lower limit of 2 days, below which it was impossible to determine the duration of fertility of the sperm in the female tract, doe No. D 103 at 2 days’ interval between the copulations failing to give a litter of mixed colour. It was evident from these experiments, however, that the length of life of the sperms in the female tract was to be measured in hours rather than in days.

The method of artificially inseminating does on heat was finally adopted and for this purpose does which had recently produced a litter or which had completed a pseudo-pregnant period were selected. The time of insemination was noted and the does were then mated to sterile (vasectomised) bucks at various intervals after this. Since ovulation would occur 10 hours after the mating the length of life of the sperm could be calculated as the time between insemination and copulation +10 hours. At first (Table VIII, dates 20. viii. 24, 23. viii. 24 and 16. ix. 24) in order to avoid any stimulation of the vaginal passages by the insemination, which might possibly cause ovulation, a few does were inseminated into the uterus direct via an abdominal incision. Previous investigation (1), however, had shown that ovulation after coitus occurs as a consequence of the orgasm attendant on it rather than as a result of the mechanical intromission of the penis, so that direct insemination into the vagina, being a much simpler and easier method, was adopted in all subsequent experiments. However, since an orgasm might be produced on insemination if the doe was not too frightened by the method of control adopted, it was considered advisable to have a number of controls which were not mated after the insemination. Incidentally also a number of does which were inseminated refused to mate when tried with the buck, some probably because they were not on heat (although every effort was made to select does which were so) and others because they were often tried under rather unusual circumstances, i.e. at night with the aid of a flash lamp, which had to be done, however, in order to get the required interval of time between insemination and copulation ; others again were of a nervous temperament and difficult to mate.

Examination of the literature also showed that there is evidence that ovulation is but rarely caused by artificial insemination in the rabbit. Heape(7) inseminated does and found that sperms reached the tubes but no ovulation occurred. Yamane and Egashira(8) obtained three litters from 36 does (8·3 per cent.) artificially inseminated without coitus, whereas over 62 per cent, were fertile when artificially inseminated in association with a sterile coitus, figures which are not very different from the results given below. Lush (9), from insemination of does, after copulation with vasectomised bucks, only obtained a 14·3 per cent, fertility, but since his normal matings only gave a 19·7 per cent, fertility the low figure is probably due to matings made when the does were not on heat and to the frequence of sterility among his does.

The semen used for the purpose of insemination was derived from the vagina of a doe which had just copulated with two or three bucks. More than one buck was used in almost every case, in order to increase the bulk of the semen obtained and also to avoid, as far as possible, testing samples which might be deficient in vitality due to lack of vigour in any particular buck.

The collecting doe was killed immediately after copulation by breaking its neck and was hung up by the hind legs. A clip was put on the vulva immediately, so that peristaltic contractions did not drive the semen out; the abdomen was cut onen and the bladder nicked with a sharp pair of scissors so that the urine fell out on the floor ; if this was not done the bladder contracted when it was touched and drove the urine into the vagina and spoilt the sample of semen. After nicking the bladder another clip was placed at its base to prevent the semen running back into the bladder when the vagina was touched ; a small nick was made in the vaginal wall below this point and the end of the inseminator—a glass tube about 312 inches long with a rubber teat at the end (like a fountain-pen filler)—was inserted and the semen was sucked up into it. When this was removed another clip was placed over the nick in the vagina to prevent the remaining semen running out. The does to be inseminated were at hand and were held by one of the writers, sitting down, with the head and fore legs of the doe held between the knees and the hind legs held uppermost, one in each hand ; the other then inserted the inseminator into the vulva, taking care to pass it over the pelvic brim before squeezing the bulb and ejecting the contents. By using a glass tube it was possible to see how much semen had been collected and how much was given to the doe ; the tube tapered from the point, with a diameter of about 18 inch, to the base with a diameter of about 14 inch and care was taken to have the point well rounded so as to avoid danger of damaging the vaginal walls. The sterilised instrument was kept before use within a boiling tube which was immersed in a water bath at body temperature. In this way it was thought that the transference of the semen from the collecting doe to those being inseminated was as near as possible to natural conditions. The amount of semen obtained from the collecting doe varied and consequently the number of does which could be inseminated from one animal also varied; in Table VIII the lots inseminated from each collecting doe (and consequently bucks) have been shown separately; this has been done so that the variations with different samples can be seen.

Table VIII.

Life of sperm in female tract. Size of litters produced by does inseminated at different intervals (N) of time before mating with a sterile buck.

Life of sperm in female tract. Size of litters produced by does inseminated at different intervals (N) of time before mating with a sterile buck.
Life of sperm in female tract. Size of litters produced by does inseminated at different intervals (N) of time before mating with a sterile buck.

During the course of the investigation several other methods of collecting the semen were tried but all shown in Table VIII were collected in this way except where otherwise stated. The disadvantage of the above method is that a doe has to be killed each time a set of rabbits is to be inseminated and in times of shortage of does (4. i. 25 and 8. i. 25) bucks were killed and the semen from the epididymis diluted with Ringer’s fluid was used ; the results did not differ materially from those in which the semen had been obtained from the vagina of the doe, but the does injected with the semen so obtained were not mated at critical periods.

Another method used to collect the semen from a doe was to insert a small compressed net-sponge into the vagina with a pair of forceps before the doe was mated, leaving the string or thread hanging from the vulva. It was, however, a tedious business getting the doe to copulate under these circumstances as the thread was continually getting in the way of the penis and without the thread the sponge could not be easily recovered. After copulation the sponge was squeezed out in Ringer’s or Walton’s fluid and does were inseminated with this; the method was used on two occasions (25. iii. 25 and 4. iv. 25). In several other cases where there was difficulty in obtaining sufficient semen for all the does available it was slightly diluted or the vagina of the collecting doe was rinsed out with Ringer’s or Walton’s fluid, either for the whole set or for one or two does in a set. The results obtained in this way (see Table VIII) did not differ materially from those in which the semen had not been so diluted.

The method of inserting the inseminator into the vagina after copulation and withdrawing the semen is not very safe and has not been used in these experiments, for the doe frequently urinates when this is being done and that this occurs cannot always be detected easily unless the amount of urine is large. Urination is also liable to occur in the sponge method, but here it takes place as the sponge is being inserted before copulation and so can be detected and another sponge used after the bladder has been emptied.

Table VIII gives a picture of the way in which the experiments were conducted —first a few does were inseminated at each consecutive time interval to find the approximate length of life, then larger numbers at certain moderately wide intervals to put these results on a statistical basis, this being followed by filling in the gaps between critical periods and obtaining larger numbers at critical times.

The incidence of sterility in the does as shown in this table appears larger than it really is owing to the fact that the same does were used over and over again in the course of the experiments. The bulk of the sterile animals, which had adherent Fallopian tubes, could be traced to one or two lots which had been inseminated with a rather long and pointed tube and it is possible that this penetrated the vaginal wall and set up peritonitis, the ciliated ends of the tubes naturally collecting the débris from the abdominal cavity, this leading to adhesions. Where a doe had not produced a litter for about 4 months she was tested by a normal mating and if she proved sterile the results obtained during the previous 4 months had to be questioned back to the time the last litter was produced. The advantage obtained by using the does over and over again was that of noting the fertility and size of the litter produced at the different times in the same doe ; unfortunately, owing to lack of space it has been found impossible to incorporate the numbers of the does in Table VIII.

Fertility

Table VIII shows the results obtained in each animal inseminated; a few of these does were suspected of sterility and have been so marked in the table; these results, however, are not included in the summary of results shown in Table IX. Of the 28 control animals which were inseminated but not tried with the buck only 1 or 3·6 per cent, proved fertile, showing that where does are kept in separate cages, as these were, there is very little chance of spontaneous ovulation or of ovulation as a result of inserting a tube into the vagina. Of the 30 does which were inseminated and tried with the buck afterwards but which refused to copulate 4 or 13·3 per cent, produced litters, the chances of ovulation being increased, presumably by the increased liability to an orgasm where the doe was jumped by the buck, although not actually mated. The former are the true controls, for coitus was not attempted in the experimental animals until the interval required after insemination had elapsed.

Table IX.

Life of sperm in female tract

Life of sperm in female tract
Life of sperm in female tract

In contrast to the low degree of fertility where no mating occurred after insemination were the cases in which the does were mated with vasectomised bugks from −2 to + 8 hours after insemination ; of the 22 does which were mated at this time 20 (or 91 per cent.) produced a litter, a degree of fertility only slightly below that obtained in normal matings extending throughout the year. After this time (8 hours after insemination, which means a life of 18 hours in the female passages before fertilisation) the chances of fertility began to decrease, thus at 10–14 hours (or 20–24 hours in the female tract) of 40 does which were mated with vasectomised bucks 24 (60 per cent.) were fertile, at 16 hours 10 out of 18 (55 per cent.), and at 18 hours 8 out of 20 (40 per cent.) were fertile. After sterile coitus at the 20th hour after insemination, or a life in the female tract of 30 hours, only 4 out of 31 (13 per cent.) were fertile, while complete sterility was obtained in the case of 29 does mated at 22 and 24 hours and after.

Thus it may be concluded that the fertility of the sperms begins to fall off after about 20 hours in the female tract to 32 hours when complete sterility is reached. While this may be taken as a general statement, the results shown in Table VIII indicate that there may be considerable variation in different samples, in all probability due to differences in the vigour of sperms obtained from different bucks and from the same bucks under different conditions of nutrition.

Size of litter

Table X gives the size of the litter produced from sterile matings made at different times after insemination. The stock of rabbits consisted of three pure strains (and some crosses between these) which are being inbred to produce homozygosity for determining physiological differences as regards fertility and growth; consequently the normal average size of the litter in the different strains varied considerably, so that in judging the effect of the length of the survival of the sperms on the size of litter it is necessary to consider each of these strains separately. For this purpose the average litter sizes of each strain obtained from normal matings have been given for comparison.

Table X.

Life of sperm in female tract. Size of litters produced.

Life of sperm in female tract. Size of litters produced.
Life of sperm in female tract. Size of litters produced.

It will be seen that litters below normal size begin to be produced at and after the 12th hour (or 22 hours after insemination), the time at which the decrease in fertility of the does, or loss in vitality of the sperms begins to show itself, while at 16–20 hours (or 26–30 hours after insemination) this decline is very marked. Two litters of 7 and 8 at 20 hours are exceptional and may possibly be explained by an earlier ovulation than usual ; statistically they lie within the range of possibility that ovulation was induced in these cases at the time of insemination.

By the above-mentioned means it has thus been found possible to produce experimentally small litters from does which normally give large litters.

In order to show that these small litters were not chance results, such as may occur in any strain, Table XI has been drawn up. This shows the distribution of litter size in normal matings of the different strains as well as those produced after inseminations where the sperm had remained for extended periods in the female tract before ovulation occurred. From this table it will be seen that the distribution in litter size is nearly similar to that of normal matings where the sterile coitus has occurred from −2 to + 8 hours after insemination, but that the distribution gets progressively shifted to a smaller size of litter as the interval at which the sterile coitus was made is raised to 10–14 hours and 16–20 hours.

Table XI.

Life of sperm in female tract. Distribution of litter size.

Life of sperm in female tract. Distribution of litter size.
Life of sperm in female tract. Distribution of litter size.

There can be little doubt, therefore, that as the end point of the vitality of the sperms in the female tract is being reached small litters instead of large ones are produced.

In order to make quite certain that the small size of litter obtained at the I4th-2oth hour was not due to a smaller number of ova being shed, does which had been inseminated at intervals of 14–24 hours before a sterile coitus were killed on or after the 20th day after coitus and the number of corpora lutea and foetuses were counted. Table XII, which gives the results of this examination, shows quite definitely that the small-sized litters and the sterility which commences from the 14th hour is due to lack of fertilisation, 41 per cent, of the ova shed being “missing” at the 14th hour (or after the sperm had remained 24 hours in the female tract before ovulation), 50 per cent, at the 16th hour and 100 per cent, at the 18th, 20th and 24th hour. In the case of doe No. C 312, where the sperms had remained 26 hours in the female tract before ovulation, only one foetus was present, although 9 ova had been shed (as shown by the number of corpora lutea present). It will be noted also that the percentage of atrophic foetuses found (3–4 per cent.) is below normal, which suggests that attenuation of the sperms results in non-fertilisation or failure of the ovum to develop rather than in lack of vigour in development.

Table XII.

Life of sperm in female tract. Post-mortems on does killed after insemination at different times before sterile coitus.

Life of sperm in female tract. Post-mortems on does killed after insemination at different times before sterile coitus.
Life of sperm in female tract. Post-mortems on does killed after insemination at different times before sterile coitus.

Diagram II has been prepared in order to offer an explanation of the causes for the gradual decline in the size of litter from inseminations occurring from 24 to 30 hours before ovulation. It is suggested that the number of sperms available (in the upper ends of the Fallopian tubes) to fertilise ova varies at different times after insemination in the form of the curve X ; the first sperms probably reach the upper tube 2–3 hours after insemination and the number increases rapidly to a maximum, this being followed by a slow decline as the sperms die off due to the conditions found in the female tract. When this end point of the life of the sperms therefore just overlaps the variability curve of the time of ovulation (A) a small litter will result while the litter will be increased in size as the time of insemination begins to approach that of ovulation.

Diagram II.

Illustrating the possible causes of small litters due to lengths of vitality of the sperm and ova.

K = Curve of available (upper tube) ova, due to variability in time of ovulation and to their short life.

X, Y and Z = Curves of available (upper tube) sperm, due to the time taken to reach the upper tubes and to the short life in the female tract.

X = From insemination (A) at 18 hours before sterile coitus (B) and 28 hours before the mean time of ovulation–small litter produced.

Y = From a normal fertile coitus at B 10 hours before the mean time of ovulation–normal large litter produced.

Z = From a fertile coitus at C made 10 hours after a sterile coitus at B–small litter produced.

Diagram II.

Illustrating the possible causes of small litters due to lengths of vitality of the sperm and ova.

K = Curve of available (upper tube) ova, due to variability in time of ovulation and to their short life.

X, Y and Z = Curves of available (upper tube) sperm, due to the time taken to reach the upper tubes and to the short life in the female tract.

X = From insemination (A) at 18 hours before sterile coitus (B) and 28 hours before the mean time of ovulation–small litter produced.

Y = From a normal fertile coitus at B 10 hours before the mean time of ovulation–normal large litter produced.

Z = From a fertile coitus at C made 10 hours after a sterile coitus at B–small litter produced.

In a normal mating at B the sperm curve Y rises well above the ovulation curve and persists well over it so that there is very little chance of a small litter being produced in a normal mating through lack of sperms, except in the rare cases in which the form of the sperm curve differs considerably from that given in Diagram II, for example, in cases of deficient vigour of the sperms from a buck in the incipient stages of sterility or experimentally through lowered vitality due to the length of time the sperms remained in the male tract (see above).

The shape of the curve of the available ova (K) is due firstly to individual variation from one doe to another as to the exact time of ovulation as well as to variation in the time at which the different follicles of the same doe rupture, of the latter evidence is available from experiments in conjunction with Mr A. Walton (to be published shortly) in which the process of ovulation has been watched under the microscope in the living animal. Secondly, the shape of this curve is due to the short length of time during which the ova remain capable of fertilisation(i); other experiments now in progress (which will be published in detail later) show that a fertile mating (C) made from 9 to 12 hours after a sterile one at B also results in a small-sized litter, this being due to the variability in the ovulation curve combined with the shape of the sperm curve or time taken by the sperm to reach the upper ends of the tubes, only the last-shed ova being fertilised by the first sperms to arrive in the upper parts of the tubes.

Incidentally the production of small litters in either of the above ways is now being used by one of the writers to obtain evidence on foetal nutrition as affected by the size of the litter ; each young of the litters of one or two from does of high fertility born under these circumstances weighs up to 90 gm. or more at birth as compared with about 55 gm. for a litter of 7 or about 40 gm. for a litter of 12. They are also specially well developed at birth and mostly have long hair like a rabbit of a week or 10 days old.

The exact shapes of the curves shown in Diagram II require further investigation, but the experiments now in progress will probably enable a more accurate estimate to be made. As drawn they represent average curves, but individual variability, whether due to breed or feeding conditions, will affect, them. There is evidence for this in the data presented in Table VIII, the point of death of the spenns varying in different samples of semen possibly according to the vigour of the buck from which it was obtained. In the same way the ovulation curve varies when a large or small number of ova are shed by the doe, for (as is partly shown in Table X) it is easier to obtain small litters from does normally producing a large number of ova than from those shedding a smaller number ; in the latter case there is a greater tendency to give a full litter or none at all.

Sex ratio

The sex ratio of the young produced from litters born as a result of sterile matings made at different periods after insemination are shown in Table XIII. There is a slight excess of males in some of the litters produced from the matings made near the end point of the vitality of the sperm in the female tract, but in view of the great variability found in sex ratios where small numbers are being dealt with it is concluded, since no very striking result is shown, that little or no difference in the sex ratio is caused by the length of time the sperm remains in the female tract before ovulation occurs.

Table XIII.

Life of sperm in female tract. Sex ratio of young born.

Life of sperm in female tract. Sex ratio of young born.
Life of sperm in female tract. Sex ratio of young born.

Since much of the literature on this subject has been referred to in a previous publication (1) it is not proposed to quote it here in any detail.

The first point to consider is how far the short life of the sperms in the female tract is peculiar to the rabbit and whether this species is typical of mammals in general.

It is well known that in the case of insects (bees) the sperm may live for a year or more in the female tract. Among warm-blooded animals, the sperm of the bat, which mates in the autumn, lives throughout the winter in the uterus and is said to fertilise the ova which are shed in the spring (Courrier (10)), although mating has been seen to occur again in the spring ; the bat is peculiar, however, in that it hibernates during the winter, and in that very large quantities of semen are stored in the uterus during this period. In the turkey, one mating is sufficient to fertilise the whole clutch of eggs laid up to, perhaps, 30 days afterwards. In the fowl, numerous investigations have been made, most authors agreeing that the fertility of the eggs laid gradually falls off after the cock is removed from the pen until absolute sterility obtains about the 16th day (Gilbert(11), Frateur(12), Laurie (13), Chappellier(14), Payne (15), Elford(16), Gray (17), Philips (18) and Kaupp(19), although as in rabbits some variation in the actual time of absolute sterility was found on different occasions. In the duck Chappellier(i4) found absolute sterility from the 7th to nth day. In the case of birds, however, the testis is abdominal and the natural temperature at which the sperms are formed is similar to that of the oviduct, which is not the case in most mammals where the scrotal temperature is lower than tlSibdominal (see Table XIV).

Table XIV.

Temperatures (° F.) of scrotum and of abdomen in the rabbit.

Temperatures (° F.) of scrotum and of abdomen in the rabbit.
Temperatures (° F.) of scrotum and of abdomen in the rabbit.

Among the domestic animals there is no evidence, from frequency distributions of pregnancy duration records, that the sperms may live for 3 weeks the normal interval between heats) in the female tract and the experiments in artificial insemination that have been made have shown while it is rarely possible to fertilise an animal outside the heat period. Those cases where the finding of motile sperms in the female tract have been reported several days after mating also are not evidence of their fertility, as has been shown in the experiments described above. Sobotta (20) states that in the mouse the sperms live in the uterus only for a few hours and are then absorbed, while Hoehne and Behne(21) found that the sperms lived only a short time in the uterus of the guinea-pig.

Robinson (22) found in the ferret that the shortest time after coitus that ovulation occurred was 3012 hours, while in one case it had not occurred after 9012 hours, which would indicate that the sperm may live longer in this species.

Benoît (23), who studied the motility of the sperms in the male tract, found in the guinea-pig and mouse that after unilateral castration the sperms in the epididymis from the side from which the testis was removed were motile up to 1 month and 25 days and 2 months and 4 days respectively, times which correspond very closely with the maximum length of survival in the rabbit.

With regard to man ovulation is usually held to occur about 14 days after the beginning of the menstrual flow, and Siegel (24) has found, from the records of men on leave during the war, that the result of a single mating made at different times during the cycle is attended by very different degrees of fertility, according to the period of the cycle at which it was made. Such a finding brings man into line with the majority of other species in which conception is only possible for a short period (oestrus) during the cycle. By his statistical methods, compiled from the averages of different individuals, Siegel found that the percentage of fertility increases from the beginning of menstruation, reaches the high point 6 days later and remains about the same height to the 13 th day and then declines to the 22nd day, after which time there is absolute sterility. It should be pointed out, however, that while this may express the average probability in the mass, it is not necessarily true of the individual, for example, Giles (25) found that the length of the cycle may vary from 21 to 35 days and so the time of ovulation, if all were grouped on the 28-day cycle, would vary from −7 to + 7 days from the normal, thereby causing a variation in degree of fertility at different times in the cycle such as found by Siegel for woman. It would, therefore, seem possible that the time in the cycle during which any individual woman is capable of conception may be more limited than is usually supposed, and the case of woman cannot, on the present evidence, be cited as evidence for a lengthy survival of the fertility of the sperm in the female tract.

This theory could, however, be tested by experiment on the monkey on the lines of a single mating made at the most unlikely time for conception first, viz. the 26th day of the cycle, and should this fail to be fertile, working back by 2-day periods to the 24th, 22nd day, and so on, until the period was found at which conception occurred; afterwards the other directions could be proceeded with, 28th, 2nd, 4th, etc. until the other point in the cycle at which fertility occurs is found. In such a way it might be found possible to obtain for monkeys much real information as to the course of the cycle and the time of oestrus of which so little is known as compared with that of the lower species of animals, since it is obscured by the fact that coitus may be allowed at any time during the cycle or during pregnancy. The rabbit also allows coitus during pregnancy.

With regard to the time differences observed between the loss of motility of the sperms and the loss of their power to fertilise ova, the writers are not inclined to think that there is any abstruse mechanism involved in these differences whereby the fertilising power is destroyed without affecting the motility. The results given above suggest rather that difficult conditions which exist for the sperm in the female passages, which allow of a life of 30 hours instead of 33 days, are sufficient to kill off sperms of low vitality under 10 hours, the time at which ovulation normally occurs after coitus in the rabbit. Although the sperms from the operated bucks may appear motile when examined under the microscope before insemination, yet they may not be able to exist in a fertile state for another 10 hours under the conditions they meet in the female tract, even though they might be able to do so under the more favourable conditions found in the male tract. An indication of this was obtained (see above, p. 160) when the time in the female tract was reduced and a litter was produced 40 days after the buck had been operated on. Benoît(23), who found that sperms taken from the testes were immotile while their motility increased as they passed down the epididymis, suggests that the secretions of the epididymal epithelium are necessary for the acquisition of motility by the sperms ; the writers are inclined to believe, however, that the increase in their motility is more probably a part of their life-cycle, just as in the failing motility in old age, rather than any specific function of the secretion, for some motile sperms have been seen in the testis itself.

The main fact which stands out in this investigation is the difference in length of time that the sperms can retain their fertility in the male tract (38 days) as compared with that in the female tract (30 hours). At present one can only speculate as to the cause for this ; in all probability, however, it is dependent not on one factor alone but on a number of contributory causes of which some may be mentioned. Crew (26) first suggested, and Moore (27) has published experimental evidence, to show that high temperatures cause the seminiferous tubules to become atrophic. Since the female organs are abdominal while the male are scrotal the possibility presents itself that the difference in length of life of the sperm in the two tracts is due to the temperature differences so produced. The actual temperature differences between the abdomen and the scrotum in a number of bucks were taken during the course of the operations described above and were found to be on the average about 3°F. (see Table XIV). The actual scrotal temperature of different individuals, however, varied as much as this without any apparent ill-effects on the sperm, although in all cases it was below the abdominal temperature.

Another possible cause of the short fertility of the sperm in the female tract is the presence of leucocytes in these organs. In semen obtained from does a few days after parturition one of the writers has, in conjunction with Mr A. Walton, observed the sperms clustering round leucocytes and apparently attempting to fertilise them ; in this way the majority of the sperms would become flocculated after remaining for some time in the female tract. There is, as is well known, difficulty in breeding from females which have an excessive discharge of leucocytes from the tract.

Benoît (23) found that although the sperms in the epididymis may remain motile for 2 months in the operated side of a unilaterally castrated mouse or guinea-pig, yet after bilateral castration they became immotile about the 8th-14th day after the operation, as had also been found by Marshall and Jolly(28) in the rabbit. Benoît observed that under these latter conditions the secretion of the epithelium of the epididymis ceased, whereas in the unilaterally castrated animals it continued. This suggests that the sperms are nourished as well..as stored by the epididymis and it might be taken that the lack of proper nourishment was a cause for their early loss of fertility in the female tract.

The results given above suggest that in spite of the large numbers of spermatozoa usually produced in the semen of normal males it is quite possible that smallsized litters may be produced by males whose sperms are deficient in vigour, although still numerous and motile.

Scientific opinion has hitherto been against the idea of a male being a good or bad “getter” of young, it being held that since the number of sperms is usually so large it is a question of all or none of the ova being fertilised. The results given above rather support the opinion often held among farmers that a boar may influence the size of a litter produced by a sow, or that a cow will not “hold” to one bull (which may be fertile with other cows) whereas she may if mated to another bull, especially if the latter is a hardy rough scrub or mongrel animal, which often has more vigour than highly-bred pedigree stock. Robertson (29) states that in Kansas it is the practice to put the mare to a horse a few minutes before she is mated with the ass stallion and that in the usual way mules only are produced, which suggests that the spermatozoa of the ass have some advantage over that of the horse stallion. Marshall (30) describes an experiment in which he inseminated a Dandie Dinmont bitch with mixed semen from a Dandie Dinmont and a mongrel dog. The resulting young were all mongrels. The sperms of the unrelated type were superior in fertilising power to those of the related type. Crew (31) has just recently shown that in double matings where the sperm of one individual is low in vigour, caused by the length of time it had remained in the female tract or by the age and lack of vigour of the cock from which it came, the newer or more vigorous sperm had a greater chance of fertilising the eggs of the hen. Lewis (32) has also obtained evidence that the sperm of some boars will live longer outside the body than that of others. In many of the results from the life of the sperm in the female tract, where the semen from dominant and recessive coloured bucks was mixed, and recessive coloured does inseminated, the young produced were often of one colour only, suggesting a more vigorous condition of the sperm from some males than from others.

In the literature of sterility among domestic animals many cases are quoted in which no pathological or anatomical defect could be found to account for the fact ; these cases are more numerous where intensive systems of agriculture are practised and where matings are made “by hand,” i.e. the male only being allowed access to the female once during the heat period for service. Under natural conditions the female would continue to be served by the male until she went off heat and so, even though the sperms from the first service might not live until the time of ovulation, those from the last service would probably do so. The results given above suggest that many of these cases may be due to delayed ovulation in the female, alone or together with a low vigour of the sperms. It seems possible that the very low fertility of travelling stallions (only 50–60 per cent, of the mares served each year produce foals(33)), where the mare is served only once in a heat period lasting about 4–7 days, may be due to this cause—the short time during which the sperms remain fertile in the female tract. No such low percentage of fertility exists in any other class of farm stock, in all of which the length of the heat period is much shorter ; nor does it occur in the wild horse or among ponies where the stallion is run with the mares.

With dogs it is the practice of some breeders to mate the bitch at 10 days after the onset of prooestrum and then again 2 days later if the bitch is still on heat in order to make certain of catching the ovulation.

It would appear that the determination of the time of ovulation in relation to the beginning and end of heat and the natural causes for variation in this are of the utmost importance to breeders of domestic animals. It has been found that in the cow ovulation normally occurs about 30 hours after the beginning, or 10 hours after the end of heat (34) and that in the sow ovulation occurs about 36 hours after the beginning of heat (32); but little or nothing is known of the causes which affect its variability. This knowledge is essential for the successful employment of artificial insemination, for where this method is being used, particularly under conditions in which the vigour of the sperms is likely to suffer (such as being kept for some hours outside the body), it would appear that the nearer the time of ovulation (allowing due time for ascent to the tubes) it is performed the greater will be the chances of success.

The methods described above for determining the life of the sperms in the male and female tracts appear to offer an experimental means of determining the conditions which affect the vigour of the body cell and of the sperm in particular, i.e. how far this can be influenced by feeding conditions (over-fatness, lack of vitamines, etc.), breeding conditions (hybrid vigour or inbreeding), or by management, Woodward (35) stating that the motility and longevity of bull sperms could be increased by daily exercise of the animal on a treadmill. These methods would appear to offer a more certain and refined test that that of mating the male directly with a number of females or of observing the length of life of the sperms outside the body—methods which, for example, were used by Lloyd-Jones and Hays (36) when determining the effect of excessive sexual activity by the male, as a result of which they found a slight decrease in the size of the litter from the 20th service made by the buck within 3 hours.

The experiments were performed at the Field Laboratories, Milton Road, Cambridge, in connection with the Institute of Animal Nutrition, and the expenses were largely defrayed out of a grant made to the Institute by the Ministry of Agriculture and Fisheries.

To Mr S. Tadman and Mr V. Swann the writers’ thanks are due for the care and feeding of the animals during the course of these experiments.

(1)
Hammond
,
J.
and
Marshall
,
F. H. A.
(
1925
).
Reproduction in the Rabbit
.
Edinburgh
.
(2)
Wolf
,
C. G. L.
(
1921
).
Journ. Agric. Sci
.
11
,
310
.
(3)
Anderson
(
1922
).
Kentucky Agric. Exp. Stat. Bull
.
244
.
(4)
Ivanoff
,
E. I.
(
1922
).
Journ. Agric. Sci
.
12
,
244
.
(5)
Heape
,
W.
(
1905
).
Proc. Roy. Soc. B
,
76
,
260
.
(6)
Hammond
,
J.
(
1921
).
Journ. Agric. Sci
.
11
,
337
.
(7)
Heape
,
W.
(
1897
).
Proc. Roy. Soc
.
61
,
52
.
(8)
Yamane and Egashira
(
1924
).
Dobutsugaku Zasshi, 36, No. 430 (Zool. Soc. of Japan). See also CASTLE (1925), Amer. Nat
.
59
,
280
.
(9)
Lush
,
J. L.
(
1925
).
Journ. Agric. Res
.
30
,
893
.
(10)
Courrier
(
1922
).
C.R. Soc. Biol
.
87
,
1365
; (1923) 88, 1163.
(11)
Gilbert
(
1904
).
Exp. Farms Report
,
Ottawa
, p.
300
.
(12)
Frateur
.
Univ, of Louvain, Inst, de Zootechnie, Bull
. No.
2
.
(13)
Laurie
(
1912
).
Joum. Dept. Agric., S. Australia
, No.
7
; (1919) No. 6.
(14)
Chappellier
,
M. A.
(
1914
).
C.r. l’ass. franç. l’av. sci
.
514
.
(15)
Payne
(
1914
).
Oklahoma Exp. Stat. Cir
.
(16)
Elford
(
1916
).
Exp. Farms Report
,
Ottawa
.
(17)
Gray
(
1916
).
North Carolina Exp. Stat. Report
.
(18)
Philips
(
1918
).
Journ. Amer. Ass., Ins. and Invest. Poultry Husbandry
,
4
, No.
4
.
(19)
Kaupp
(
1919
).
Journ. Amer. Ass., Inst, and Invest. Poultry Husbandry
,
5
, No.
7
.
(20)
Sobotta
,
J.
(
1895
).
Arch, mikros. Anat
.
45
,
15
.
(21)
Hoehne and Behne
(
1914
).
Zentbl. f. Gynaek
.
38
, No. 1 and Verhandl. d. Deut. Gesell. f.Gynaek. 1913, p.
514
.
(22)
Robinson
,
A.
(
1918
).
Trans. Roy. Soc. Edin
.
52
,
303
.
(23)
Benoît
(
1925
).
Travail de l’Institut d’Histologie de Strasbourg, and Bull. d’Hist. app. à la Phys, et la Path
.
1925
,
2
, 78.
(24)
Siegel
,
P. W.
(
1916
).
Münch. Med. Wochenschrt
.
63
,
748
.
(25)
Giles
,
A. E.
(
1901
).
Menstruation and its Disorders
.
London
.
(26)
Crew
,
F. A. E.
(
1922
).
Journ. Anat
.
56
,
98
.
(27)
Moore
,
C. R.
(
1924
).
Amer. Journ. Anat
. 34, 269, 337; Endocrinology (1924),
8
,
493
.
(28)
Marshall
,
F. H. A.
and
Jolly
,
W. A.
(
1905
).
Phil. Trans. Roy. Soc. B
,
198
,
99
.
(29)
Robertson
(
1917
).
Univ. Kansas Sci. Bull
.
10
,
293
.
(30)
Marshall
,
F. H. A.
(
1922
)
The Physiology of Reproduction
.
London
.
(31)
Crew
,
F. A. E.
(
1926
).
Proc. Roy. Soc. Edin
.
46
,
230
.
(32)
Lewis
(
1911
).
Oklahoma Exp. Stat. Bull
.
96
.
(33)
Marshall
,
F. H. A.
and
Crosland
,
W. P.
(
1918
).
Journ. Bd. Agric
.
24
,
1357
.
(34)
Hammond
,
J.
Reproduction in the Cow (in course of publication)
.
(35)
Woodward
(
1920
).
Washington Exp. Stat. Bull
.
158
,
20
.
(36)
Lloyd-Jones
,
O.
and
Hays
,
F. A.
(
1918
).
Journ. Exp. Zool
.
25
, No.
2
,
463
.