The Effect of Irradiation With Different Wave-Lengths on the Oestrous Cycle of the Ferret, With Remarks on the Factors Controlling Sexual Periodicity

Bissonette, as a result of an investigation in Cambridge, made the important discovery that with increased illumination by electric light (200 watt bulbs) in midwinter and therefore at the time of the normal anoestrus, ferrets came into full oestrus with typical vulval swelling in from 38 to 64 days. This work was suggested to him by Rowan’s experiments with the junco finch and his own experiments with the starling, in which sexual activity was also brought about by artificial illumination in mid-winter (Rowan, 1926, 1929, 1930; Bissonette, 1930, 1931a, b, 1932, 1933, Bissonette and Wadlund, 1931). At the same time Baker and Ranson (1932) showed that the oestrous cycle in the vole could be modified by varying the rations of artificial light, though other factors were concerned. Furthermore, the effect of illumination with ordinary electric light in producing oestrus in the ferret has been fully confirmed by Hill and Parkes (1934).

The present paper describes experiments which were made in order to determine whether this acceleration of the recurrence of the oestrous cycle is brought about by general heat and light radiation, or whether the effect is due to stimulation by definite portions of the spectrum. The ferrets were subjected to irradiation by monochromatic, or by narrow bands of light, in different regions of the spectrum extending from the near infra-red to the near ultra-violet.

It was intended that these experiments should begin in the early autumn of 1932 but, owing to an outbreak of distemper which necessitated the killing of all the stock and acquiring a new lot of animals, there was considerable delay in starting. The ferrets have since been kept under observation, being again utilised in the past autumn, winter and spring, and others have been added to the stock. In addition to the experiments with irradiation, some observations have been made on the effect of darkness, and, in one instance, of blindness, on the recurrence of the cycle.

The general arrangement of cage, light source and filters is shown diagrammatically in Fig. 1. Light from the source S passed through the filters F and fell on the cages, the fronts of which were open except for wide mesh netting. The backs of the cages were formed by glass mirrors M inclined at an angle of ca. 25° to the normal, so that additional radiation was reflected downwards. The layer of wood shavings on the floor was shallow, so that the ferrets could not escape from the radiation. The fronts of the cages were 43 cm. wide by 35 cm. high and two ferrets were placed in each cage. The cages were placed symmetrically round three sides of the source, which was carefully screened so that no stray artificial light could escape. The filters were square, 15 × 15 cm. and were placed at a distance of from 15 to 23 cm. from the source. The front of the cage, the whole of which could be illuminated, was another 40 cm. in front of the filter. The room was well lit by windows and, during the daytime, the cages were in reasonably bright diffused daylight. Artificial illumination was given after dusk for a period of 8 hours. The irradiation was started at dusk each day ; the time of switching on varied of course with the time of year, from 4 o’clock in January to 8 o’clock (Greenwich time) in May. The room was well ventilated with a high ceiling and partition walls open at the top, and the lamps caused no noticeable change in the temperature of the room. The ferrets were fed once a day on milk with a small portion of meat occasionally. The cages, mirrors and filters were cleaned every other day. The air temperature in the cages was not appreciably altered by the radiation.

Two sources of radiation were used. A large Sollux filament lamp of 1000 watts for the infra-red, red and green, and an Hanovia A.c. quartz lamp for the yellow, violet and ultra-violet. Both lamps were run for a considerable time before use, so that the change in intensity due to “ageing “would be reduced to a minimum. The intensity given is that of the direct radiation at the centre of the floor of the cage at the end of the year. It was slightly greater at the beginning. The intensity in the very front and at the very back of the cage would be 60 per cent, greater and 30 per cent, less than that in the middle. The intensity and spectral distribution of the various radiations were measured by means of a large quartz monochromator (Bowden and Snow, 1934), and a thermopile. The following radiations were used.

Source: filament lamp. Filter: ilford infra-red. This filter transmits radiation of wave-length longer than 7350 Å. which is on the limit of visibility. The intensity at 7500 Å. (which is invisible to the human eye) was 2640 ergs cm.²/sec. at the centre of the cage. The intensity at somewhat longer wave-length (λ 8000-Aλ 10,000 Å.) was 33,000 ergs cm.²/sec.

Source: filament lamp. Filter: a 1 cm. thickness of 10 per cent, copper sulphate solution (to remove the infra-red) plus an Hford “Spectrum Red” gelatine filter. This transmitted a band in the red extending from λ 6200 to λ 6700 with a maximum at λ 6500. Total red intensity at centre of cage 99 ergs cm.²/sec.

Source: quartz lamp. Filter: a copper sulphate gelatine film plus Hford “Spectrum Yellow.” This transmitted the mercury yellow lines at λ 5769 and λ 5790. The arrangement of this cage differed somewhat from the others since it was placed underneath the quartz lamp. The cylindrical shape of the horizontal lamp made this the only possible way of getting uniform illumination on a third cage. No mirror was used in this case, and the intensity given is that falling on the bottom of the cage where the ferrets necessarily lived. Intensity of yellow light 13·5 ergs cm.²/sec.

Source: filament lamp. Filter: 1 cm. of 10 per cent. CuSO₄ plus Hford “Spectrum Green.” This transmitted a band in the green from λ 5050 to λ 5350 with a maximum at λ5200. Intensity at centre of cage 124 ergs cm.²/sec.

Source: mercury lamp. Filter: 1 cm. of 10 per cent. CuSO₄ plus Hford “Spectrum Blue.” Transmitted the mercury violet A 4368. The intensity was low (cu. 1 erg cm.²/sec.) at the centre of the cage.

Source: mercury lamp. Filter: 1 cm. io per cent. CuSO₄, plus Chances u.v. glass. Transmitted the ultra-violet light λ 3650 Å. Intensity at centre of cage 350 ergs cm.²/sec.

Fourteen ferrets were used, and these were grouped in pairs in seven separate cages. In the experiments of 1932−3 four pairs were kept as controls (including one blind ferret), and the remaining pairs irradiated with infra-red, red and green light respectively. Apart from the irradiation, the housing and treatment of all the ferrets were identical. The ferrets were examined weekly and the size of the vulva observed.

The irradiation was begun on December 23rd, 1932, and the results are shown in Table I. The date recorded with the + sign is that at which the vulva first showed a definite swelling which continued to increase.

The controls all came on heat in April and May in the normal way except No. 8, which is blind and did not come on at all. Column 4 of Table I shows the number of days which elapsed before the ferrets came on heat reckoned from the date on which radiation was begun (December 23rd). The first control came on after 111 days, the last after 144, and the mean for the seven unirradiated ferrets was 124 days. Both the ferrets in the infra-red came on after 111 days. The first ferret in the red came on after 66 days. The second, however (No. 12), did not come on until the normal time. Both ferrets in the green light came on in 66 days. The last column shows the acceleration in the oestrous cycle produced by the different radiations. If we neglect ferret No. 12 the experiments indicate that both red and green light cause a marked acceleration and infra-red radiation has only a slight effect, if any.

Before discussing this further we may consider the more detailed results obtained in 1933−4. These experiments were begun earlier (November 1st), and the radiations used extended from the near infra-red to the ultra-violet. The same ferrets were used as in the previous year but they were interchanged so as to eliminate, as far as possible, their individual characteristics. (See p. 414.)

Nearly all the ferrets were required for irradiation, but the information from the last year made it unnecessary again to keep a large number of unirradiated controls, since it is known from this and from other work at what time the normal ferret comes on. The one unirradiated ferret came on on April 11th, which is within a few days of the mean obtained in the previous year. The results are shown in Table II.

It is again apparent that irradiation with light of different wave-lengths has a very marked accelerating effect. This is brought out clearly in Fig. 2, where the date at which+occurred for all the ferrets is plotted on a graph. The abscissae are quire arbitrary and merely indicate the “colour” of the radiation. Each point represents a single ferret.

The points plotted as X (Fig. 2) under “no irradiation” were obtained for the controls in 1932-3 ; the point • was for the single unirradiated ferret in 1933−4. The rest of the points were obtained in the latter year. The ultra-violet light was the most effective, the ferrets coming on in 28 and 42 days respectively instead of in 176 days (a mean acceleration of 141 days). The green, red and yellow were also effective, causing a mean acceleration of 74, 68 and 67 days respectively. The infrared again appears to have had a slight effect (mean acceleration, 26 days), and the violet a very small effect, if any at all. Before discussing these results further it is advisable to consider the reproducibility of the results.

It will be remembered that one ferret (No. 12) showed no appreciable acceleration with red light, whereas three others did so, and it seemed desirable to investigate how far this might be due to the individuality of the animal. The ferret in the following year came on about the usual time. The way in which the ferrets were interchanged for the two sets of experiments and their behaviour under irradiation is shown in Table III.

It is clear from the table that the effects are due to the action of the light and not to any peculiarity of the ferret. The agreement among the irradiated ferret pairs is good (with the single exception of No. 12 in 1932–3 and of the violet pair in 1933•4), and in every case the ferrets behaved similarly when treated similarly in the successive years or changed their behaviour when the irradiation was changed.

Originally it was intended that the intensity of all the radiations falling on the ferrets should be the same, but the limited size of the rooms available made this impossible. We thus have two variables, wave-length and intensity.

In Fig. 3 the mean acceleration (for each pair of ferrets during 1933−4) produced is plotted against the intensity for all the radiations used. This curve shows that both wave-length and intensity are important. Although the intensity of the near infrared at λ 7500 Å. was at least twenty times as great as that of the red (λ 6500) it produced a much smaller acceleration (26 days as compared with 68 days for the red in 1933−4, and 13 days compared with 58 days for the red in 1932−3). This result is important, since it shows that the accelerating effect cannot be due merely to the absorption of heat rays by the animal. It will be noticed that the wave-length limit is quite sharp ; λ 6500 caused a marked effect, λ 7500 a much smaller one. Had the intensities been equal the difference would probably have been more striking still.

Over the rest of the spectrum it will be seen that the acceleration produced follows in the same order as the intensities of the radiations and not as their wavelengths. It would seem probable that the effective radiation extends from the red to the near ultra-violet and that over this region of the spectrum intensity is more important than wave-length. This conclusion is of course only qualitative and further experiments with more ferrets and varied intensities would be necessary to determine the degree to which it holds true. There is possibly a difference in “intensity of radiation” considered physically and considered biologically, that is to say, radiations outside the area red to near ultra-violet which do not stimulate the retina of the eye (or such parts as may be stimulated by visible light) are distinct biologically from radiations to which the eye is sensitive, so that the intensity only becomes important when the radiations are within the above mentioned region of the spectrum.

The further history of the irradiated and normal ferrets remains to be recorded up to August 1st, but it is doubtful whether any significance should be attached to the results. The irradiations were discontinued in 1933 on May 1st and in 1934 on May 14th. On August 1st the two irradiated with ultra-violet light (Nos. 2 and 14) were still fully on heat. No. 3 (violet light) which was late in coming on and No. 11 (infra-red) were also still on heat. No. 4 (violet light) died while still on heat on June 30th. No. 10 (infra-red) also died while still on heat on July 10th. The ovaries showed “blood follicles.” Both of the ferrets which died were rather fat but showed no sign of disease. The other ferrets began to go into anoestrum about the following dates: No. 1 (yellow), June 13th; No. 5 (green), May 16th; No. 6 (green), May 23rd; No. 7 (red), April 25th; No. 9 (red), May 16th—this ferret’s vulva never subsided to the complete anoestrous state and it began to swell again, indicating a second heat period, about June 27th. No. 12, which was not irradiated in 1933−4, and came on heat about April nth, began to go into anoestrum on June 20th, but started to come on heat a second time on July 4th and is still on (August 1st). Of the normal (control) ferrets in 1932-3 two came on heat a second time, the dates of the periods being as follows: first ferret on heat from May 16th to June 27th and from July 18th to September 5th ; second ferret on heat from April 17th to May 20th and from July 18th to September 5th. The irradiated ferrets did not come on heat a second time in one year, except No. 9 recorded above.

Seven female ferrets which were subjected to incomplete darkness from April 21st to July 23rd to see if this had any inhibiting effect on the recurrence of oestrus. Of these four were beginning to come on heat before they were put into the dark room. The darkened conditions did not prevent them from coming fully on and remaining on until the end of June. The other three which were anoestrous at the commencement remained so throughout, showing no swelling of the vulva at all.

It may be again mentioned that the blind ferret (with cataract) has never shown any swelling of the vulva but has remained perfectly healthy and well nourished (August 1st).

The experiments show that the activating radiation is not confined to a narrow range of wave-lengths but extends from the red (λ 6500) to the near ultra-violet (λ 3650). There appears to be a fairly sharp threshold on the long wave-length end of the spectrum in that λ 7500 is barely effective although its intensity is high. This long wave-length threshold corresponds approximately to the limit of visibility of the human eye. At the other end of the spectrum, however, the ultra-violet radiation, which is beyond the limit of most human vision, is still very effective. It is possible of course that the ferret may be able to see by this radiation¹.

The irradiated ferrets did not appear to be more physically active than the nonirradiated ones. Also it is scarcely conceivable that they could have had more physical exercise than normal ferrets which are employed for “rabbiting,” and which come on heat at the normal time. The onset of sexual activity cannot be due simply to increased physical exercise and movement. The irradiated ferrets did not consume any appreciably greater quantity of food.

The wide range of effective wave-lengths argues against the photochemical synthesis of some simple chemical substance, since this is usually brought about by irradiation over a much narrower region of the spectrum. On the other hand Warburg has shown that the respiratory function of ferments—probably a more complicated process—is stimulated by all wave-lengths extending from the red to the ultra-violet. (See Anson and Mirsky, 1930.)

The intensity of the radiation is important and unless it exceeds ca. 1 erg cm.²/sec. it does not have any marked effect. This latter conclusion is, however, based on the “violet” results, and needs further examination to see if it is true for other wave-lengths. It should be noted that in no case did the radiation produce an inhibitor effect such as Bissonette (1933) found for starlings in green light.

It would seem most probable that the light acts through the ferret’s eyes stimulating the receptors in the retina.

Evidence that the stimulus set up by irradiation acts through the intermediation of the pituitary gland has been obtained by Hill and Parkes (1933) who have demonstrated that after hypophysectomy ferrets do not come on heat even though they be artificially irradiated, and this is in conformity with what is known of the function of the pituitary in other animals (Smith, 1927 ; Collip, Selye and Thomson, 1933). The probability is therefore that the stimulus is transferred from the retina to the pituitary. In this connection Hogben’s observations on the effect of radiations on the pituitary in the Cape clawed toad (Xenopus) may be quoted, although in this case the effect was shown in another function of the gland. Hogben found that in Xenopus the secretion of the melanophore hormones of the pituitary is stimulated through colour vision. “In the eyeless animal the melanophores are neither fully expanded nor fully contracted. The dark background response (fully expanded melanophores) is caused by photoreceptors in the fundus of the eye which reflexly stimulate the pars intermedia (of the pituitary). These receptors are most sensitive to light from the red end of the spectrum and are hardly affected by blue-green rays. On the other hand, the photoreceptors of the white background response are located in the periphery of the retina ; they are not sensitive to red light and appear to control the activity of the pars tuberalis. On this view, the two responses are separate entities and the secretion of two different hormones of the pituitary is differentially controlled by different wave-lengths” (Hogben, private communication to Miss Whetham (Whetham, 1933)). These results are interesting as showing that in some animals at any rate the activity of the pituitary gland is directly affected by irradiation through the retina. Moreover, it was found that in blinded toads (Xenopus) the ovary was under-developed (Hogben, Charles and Slome, 1931). This condition may be compared with that of the blind ferret (still alive) which never experienced oestrus. The fact that Ceni (1928) found that cocks which had been blinded underwent degenerative changes in the testes is perhaps a comparable phenomenon.

Hill and Parkes, however, have expressed the view that the oestrous cycle is normally due to some inherent rhythm of the anterior pituitary and occurs independently of external factors ; in other words, the problem of the causation of oestrus in the ferret is analogous to the problem of the causation of the five-day cycle of the unmated mouse. This conclusion is based largely on the results of experiments in which ferrets were subjected to darkness for several weeks, and nevertheless subsequently experienced oestrus at about the usual time. It is to be noted, however, that on comparing the dates of vulval swelling with those observed by Hammond and Marshall (1930) there was on the average a definite lag in the times of the onset of oestrus. This is born out by the results recorded in this paper. Bissonette also (1933) found that with two hooded ferrets there was a delay in the onset of oestrus.

If there were no external factor in the recurrence of heat and the problem were merely like that of the dioestrous cycle of the mouse, as suggested by Hill and Parkes, there is no apparent reason why the periodicity of sexual activity in the ferret, as in most animals, should so generally conform to that of the seasons, this relation between seasonal change and the recurrence of breeding being a matter of common knowledge. Concerning the precise factors which determine the relation, there is much left to discover but the present paper presents evidence to show that light or irradiation of a certain kind is an important one. The recurrence of the short cycle within the sexual season in polyoestrous animals is another problem and in no way comparable. It is not a matter that concerns the ferret which is monoestrous and does not experience short dioestrous cycles.

As pointed out previously the principle that the action of light rays may be a factor in the recurrence of oestrus is probably of very wide application, as shown by the general tendency for animals to breed in the spring and early summer when the days are lengthening (Marshall, 1932). On the other hand, in tropical countries where the amount of daylight and other correlated environmental conditions are similar throughout the year, there is often no restricted season for breeding among the animals. This fact is remarked on and stressed by Semper (1881) in his account of the fauna of the Philippine Islands. More recently Bates (1908), in a paper on the breeding of birds in Southern Kamerun, makes a similar observation for that country but states that there are exceptions to the general rate. Thus, woodpeckers, barbets and starlings are hindered from breeding by the rains, while colies and thrushes prefer the rainy season. In a later paper (1927) he makes confirmatory statements. Other observers (Lynes, 1925; Paget-Wilkes, 1928 ; Vaughan, 1929,1930 ; and Moreau, 1931) on tropical birds affirm that many species at any rate have restricted breeding seasons dependent upon ecological factors—food supply, growth of suitable vegetation, the right kind of grass for nests (weavers and reed warblers), soft mud, consequent upon the rains, for nests (swallows), etc. Further examples of seasonal breeding in tropical birds are to be found in Bannerman’s Birds of Western Africa (1930, 1931, 1933). It would appear then that where there is little or no variation in the daylight, other factors may come into play and control the breeding times for certain species. These factors probably act through the nervous system by the intermediation of the anterior pituitary on the organs of reproduction.

Reference should also be made to Miss Whetham’s statistical researches on egg production in fowls in relation to seasonal change (1933)-These show that there is some correspondence between variation in egg production and variation in daylight in different latitudes, and in view of the further fact that artificial light is utilised to increase egg production there can be little doubt that we have here another case of activity of the reproductive organs depending upon radiation. Cole’s recent paper on his experiments on the relation of light to the reproductive cycle in the mourning dove provides additional confirmation (1933).

At the same time it must be freely admitted that species of animals differ widely in their reactions to the factors that control periodic breeding, as shown, for example, by the different times of year at which they come “on heat.” Thus, some species of deer such as the red deer in this country and certain varieties of sheep have their breeding season in the autumn when the daily duration of light is decreasing. It may be that in such animals the light rays act differently from what they do in the majority. Nocturnal animals such as the bats and animals which spend much of their lives underground such as the mole are also exceptional. Moreover, there can be no doubt that food supply is for all animals an important factor in sexual and reproductive activity and that the vitamin E, so plentiful in green food, is specific in its effect upon the organs of reproduction. In this connection the evidence adduced by Heape (1931) as to food supplies in relation to excessive reproduction as shown in certain years by voles and lemmings and other animals must not be overlooked.

Furthermore, apart altogether from light rays or other seasonal environmental factors there can be no question that there is in all animals an inherent tendency towards an internal reproductive rhythm. Thus, in the case of a pony belonging to the late Prof. Ewart, brought to Scotland from Timor which is in the southern hemisphere, the animal when first imported came “on heat” in the autumn (or at the same time as spring in Timor) but subsequently adjusted its sexual periodicity to the conditions of the northern hemisphere and foaled in the spring (Marshall, 1910). Again, it is well known that in some species the gonads begin to undergo their normal development preparatory to breeding in mid-winter or even earlier and so before the days commence to lengthen. That a physiological process is conditioned by many factors is only in keeping with what we know about metabolism generally and the view adopted in this paper that in the ferret and probably in many other animals light radiations of a particular wave-length and sufficient intensity are an efficient cause in the periodicity of reproduction is in no way inconsistent with the assumption that other factors exist.

The recurrence of oestrus in the female ferret is greatly accelerated by irradiation with light of various wave-lengths. Heat rays and the near infra-red (A 7500) are comparatively inactive. The effect begins with the red radiation (A 6500) and extends throughout the visible to the near ultra-violet (A 3650). Over this range of the spectrum intensity appears to be more important than wave-length.

None of the wave-lengths employed produced retardation in the recurrence of oestrus.

Female ferrets subjected to incomplete darkness did not come on heat but individuals which had already begun to come on heat entered into full oestrus and remained in that state for a normal period.

It is concluded that in the ferret and so probably in many other animals light radiations of particular wave-length and sufficient intensity are an efficient cause of reproductive activity, but that the recurrence of the oestrous cycle is conditioned also by other factors which in the absence of variation in the daily duration of light may play an important part.

Our thanks are due to the Hanovia Quartz Lamp Company for the trouble they took in providing us with the lamps used in the investigation. We desire also to express out indebtedness to Mr J. Pike for his valuable help in superintending the experiments and looking after the animals.