1. Unlike any other species of flea which has been cultured successfully Spilopsyllus cuniculi (Dale) is entirely dependent for its own reproduction upon its host becoming pregnant.

  2. The ovaries of fleas kept on male or non-pregnant female rabbits remain immature whereas those on a pregnant host are mature at the time of parturition. Vitellogenesis commences at a critical point approximately 7 days pre-partum, irrespective of how long the fleas have been on the host. It is postulated that a factor required by the flea for ovarian development is only available during the final week of pregnancy, and not at all in male or non-pregnant rabbits. The factor disappears from the adult after parturition but is present in her nestlings for at least 7 days.

  3. Most fleas desert the adult doe shortly after the young are born and enter her nest, where copulation and oviposition occur.

  4. The factor does not act solely as a trigger initiating in the flea developmental processes which then continue in its absence; ovarian regression occurs among maturing fleas transferred to hosts that do not supply the factor.

  5. The processes of vitellogenesis, not those of oogenesis, fail in the absence of the factor. It is suggested that the ‘yolk-forming hormone’ normally secreted by the corpus allatum of the adult insect may only be produced by rabbit fleas when they can obtain the postulated factor. Variations in the quantity of this factor available to fleas on different hosts could explain the observed variations in ovarian activity. Some parallels between the factor and steroid hormone levels in the host’s blood are discussed.

After several attempts to culture the European rabbit flea Spilopsyllus cuniculi (Dale) in the laboratory had failed when using standard methods, a fuller investigation of the reproductive processes of this flea was undertaken. The normal technique for breeding fleas (Leeson, 1932) is to release young adults into a suitably proofed vessel with a small mammal to act as a source of blood for the fleas. The vessel also contains a supply of food and drinking water for the host and a diet which usually includes dried blood and yeast for the flea larvae. Granules of mica (vermiculite) may be added to absorb the urine of the host, which can otherwise be injurious to the flea eggs and larvae. After a number of days the mammal is removed and the cultures are kept in a warm, humid atmosphere for the duration of the pre-adult stages. This culture technique, with minor modifications, has proved successful for several species of rodent fleas, especially those of rats, and fleas of some domesticated animals such as cats (Edney, 1945; Linduska & Cochran, 1946; Elbel, 1951, 1952; Burroughs, 1953; Hudson & Prince, 1958 a, b).

When specimens of S. cuniculi were released on caged, non-pregnant European rabbits, both the wild and the domestic strains of Oryctolagus cuniculus (L.) being used, the fleas showed no signs of reproduction. No eggs or larvae were found, and the female fleas never attained a gravid appearance. Similar negative results were obtained when the hosts used were a white rat Rattus norvegiens (Berk.), a white mouse Mus musculus L., a field mouse Apodemos sylvaticus (L.), and a guinea-pig Cavia porcellus L.

During April 1958 150 female fleas, gravid in appearance, were removed from a wild rabbit shot in Kent and were released on the head and ears of a domestic rabbit housed in a standard wire-mesh cage supported over a tray of water. It was anticipated that eggs would be shed by the fleas and that they could be collected from the surface of the water. However, no eggs were found and after 7 days some of the fleas were removed from the rabbit and dissected. It appeared that the larger oocytes in the ovaries were undergoing yolk resorption and subsequent destruction.

No further attempts were made to culture the flea until i960. By then a study of samples of fleas removed from wild rabbits collected at weekly intervals had shown that there was a peak of reproductive activity of the fleas in February and March which coincided with the beginning of the peak breeding season of the wild rabbit.

The experimental breeding of the rabbit flea was first accomplished in May 1960. Similar groups of previously unfed, virgin fleas were released on each of three female rabbits, which were pregnant wild, pregnant domestic and non-pregnant domestic specimens. Each rabbit was housed in a two-compartment breeding hutch and for the non-pregnant one an old, but sterilized, rabbit breeding nest was placed in the dark compartment. The nests made by the pregnant rabbits were examined 5-8 days after parturition and contained flea eggs and larvae. Dissection of a number of the female fleas found on the rabbits at this time confirmed that the ovaries had matured, and that copulation and impregnation had occurred. By contrast, no eggs or larvae were found in the ‘nest’ of the non-pregnant rabbit after 16 days and the fleas had immature ovaries and had not been impregnated. These observations formed the starting point for the present investigations, parts of which have been briefly reported elsewhere (Mead-Briggs & Rudge, 1960).

Most of the fleas used were from wild rabbit nests collected in Kent by Mr J. W. Sears. The parental generation of fleas was removed from suitably infested nests and discarded, and these nests were then kept in closed Polythene bags until the filial generation emerged. As an alternative, larvae were collected from some nests and reared in isolation in glass phials containing a medium composed of 100 parts of washed sand, 20 parts of rabbit blood (vacuum-dried) and 1 part of de-bittered yeast, and kept at approximately 80% R.H. The rest of the fleas used were reared from eggs and larvae produced in some of the more successful of the breeding experiments. Experiments could thus be commenced using fleas that had not previously taken a blood meal. They were also all virgin fleas as copulation in this species never takes place on emergence from the cocoons.

When specimens of S. cuniculi are placed on a rabbit most of them quickly move into the pelage and disappear from sight. A proportion, often from 5 to 15%, do not pass through the fur to the body surface, but move towards the sides of the head or body via the tips of the hairs and finally drop from the rabbit without attempting to feed, and these fleas usually make no attempts to return even when they are near death by starvation. The fleas that ‘take’ to the rabbit may remain on the host for weeks or months, provided other rabbits are not kept in the same hutch. This tendency to remain on the host’s body, as seen in rabbit fleas, is the exception rather than the rule among fleas.

Batches of fleas were always released on a test rabbit, held firmly on a large enamelled tray, by inserting its ears into 3x1 in, glass specimen tubes containing the fleas. Any that left the rabbit were easily collected with an aspirator. As the fleas in the tube came into contact with the ear most of them moved down its length towards the head, often probing the skin and feeding in the process. After about 10 min. the tubes were removed and any fleas loose on the ears were encouraged to move towards the head, where they were afforded greater initial protection from the rabbit, which once it was returned to its hutch normally spent some time flicking and scratching its ears in an attempt to rid itself of the fleas. After a few hours the rabbit settled down and no longer appeared to be troubled by the fleas, even when a hundred or more had been released. Rabbit fleas tend to congregate in certain regions on their host and the most favoured are the ears, the eyelids and the top of the head from between the ears to the nostrils. Fleas which settle on the ears or the eyelids are usually firmly attached by their mouth-parts whereas those on the head are frequently unattached. On the ears the outer, hairy surface of the antero-lateral flap of the pinna is preferred, but groups of fleas may be found on the inner, semi-glabrous surface of this flap slightly distal of the midpoint. Fleas could be recovered from the rabbit by combing the head or picking them from the ears with fine forceps. The latter method was used whenever possible as it allowed the removal of the exact number of fleas required with the minimum of disturbance to the rabbit and other fleas.

Usually there was no danger of any unwanted fleas straying on to a test rabbit but occasionally groups of marked fleas were used so that individuals could be recognized on subsequent recovery as belonging to a particular group. Attempts to mark fleas semi-permanently with paints or dyes proved unsuccessful, although such a method has been reported by Kosminsky & Soloviova (1959). The technique used involved the removal of the terminal subsegment from the tarsus of one of the six legs; no regeneration of the amputated part is possible and similar disfigurements occur only very rarely under natural conditions. The reproductive physiology and general behaviour on caged rabbits of fleas marked in this manner were unaffected.

Most of the information for this paper was obtained by the examination of the reproductive organs of female fleas removed from test rabbits at intervals after their original release. The reproductive system was dissected from each flea in Ringer’s solution and Mead-Briggs (1962) describes the method of dissection and the structure of the reproductive organs. The female rabbit flea usually has five or six panoistic ovarioles in each ovary and during preliminary observations measurements were made of the length and breadth of the largest oocyte follicle in each ovariole (i.e. the follicle nearest the oviduct and referred to in this paper as proximal). It was found that the ovarioles could be arranged into two groups, of approximately equal numbers, one group having rather larger and more mature proximal oocytes (Fig. 1), and that the production of chorionated oocytes alternated between the two groups of ovarioles. The length of one of the larger proximal oocyte follicles (Fig. 2) gave a satisfactory indication of the state of maturity of the ovaries, and this criterion is adopted for comparative purposes throughout this paper. When several fleas were taken from a host at the same time their ovaries were all in a very similar developmental condition; this allowed a sampling technique to be used, and it is assumed that each sample of fleas dissected was representative of the condition of the fleas that remained on the rabbit at the time of sampling. Whenever possible the sample consisted of three fleas.

Fig. 1.

Diagram of the internal reproductive organs of a female rabbit flea with ovaries near maturity. (The terminal filaments of the ovarioles disconnected during dissection and the material arranged to lie in one plane.)

Fig. 1.

Diagram of the internal reproductive organs of a female rabbit flea with ovaries near maturity. (The terminal filaments of the ovarioles disconnected during dissection and the material arranged to lie in one plane.)

Fig. 2.

Diagrammatic optical section of the posterior (proximal) region of an ovariole to show the measurement taken as a criterion of the degree of ovarian maturity, ep. pl., epithelial plug; fol. ep., follicular epithelium; g.ves., germinal vesicle; l., length of proximal oocyte follicle; 00.1oo.2, oo.3, oocytes1−3 respectively; ped., pedicel; y.g., yolk granule.

Fig. 2.

Diagrammatic optical section of the posterior (proximal) region of an ovariole to show the measurement taken as a criterion of the degree of ovarian maturity, ep. pl., epithelial plug; fol. ep., follicular epithelium; g.ves., germinal vesicle; l., length of proximal oocyte follicle; 00.1oo.2, oo.3, oocytes1−3 respectively; ped., pedicel; y.g., yolk granule.

The region of the ovariole adjacent to the lateral oviduct was examined in every dissected specimen. This region may contain evidence of a recent ovulation in the form of a collapsing follicular epithelium, which is subsequently converted into a zone of diffuse tissue with prominent yellow pigmentation constituting the familiar ‘corpus luteum’, or follicular relic.

The spermatheca of each female flea was separated from the ovarian tissue and checked for the presence of spermatozoa but because of the opacity of the spermathecal bulga spermatozoa were frequently difficult to see. In view of this the spermatheca was always ruptured, by draining the saline from beneath the coverslip with absorbent paper until sufficient pressure was exerted, when any spermatozoa present flowed out.

1. Fleas on a pregnant rabbit (normal ovarian maturation)

In one experiment 100 female and forty male, unfed, virgin fleas were released on a pregnant rabbit when it was 2612 days pre-partum. A smaller number (40♀ and 20♂) of similar fleas was released on a non-pregnant doe, which served as a control. Nine samples of three female fleas were removed from the pregnant rabbit during its pregnancy, the final sample being taken 12 hr. pre-partum. Eight similar samples were taken from the non-pregnant rabbit over a period of 31 days from the original release. The ovaries of each flea were examined and the degree of development assessed by measurement of the length of a proximal oocyte follicle. The results are shown in Fig. 3, where the mean length of the proximal oocyte follicles for each sample of three fleas is plotted against the number of days that the fleas had been on the rabbit. A second ordinate scale relates to the pregnant rabbit and indicates the number of days from parturition when the samples were taken. The results show that there was only slight ovarian development among the fleas fed on the non-pregnant rabbit during the period of 31 days, whereas the fleas fed on the pregnant rabbit underwent rapid ovarian growth during the last week of the host’s pregnancy, after an initial period of slow development. The transition in rates of growth occurred about 712 days pre-partum when the length of the proximal oocyte follicle was 130μ.

Fig. 3.

Graph to contrast the development of the proximal oocyte follicles of fleas kept on a pregnant rabbit and on a non-pregnant rabbit. Each point is a mean based upon the dissection of three fleas.

Fig. 3.

Graph to contrast the development of the proximal oocyte follicles of fleas kept on a pregnant rabbit and on a non-pregnant rabbit. Each point is a mean based upon the dissection of three fleas.

In another experiment thirty-five female and twenty-five male, unfed, virgin fleas were released on each of four rabbits. One rabbit was a non-pregnant control (rabbit A) and the others (rabbits B, C and D) were pregnant, respectively 10 days, 19 days and 29 days pre-partum at the time the fleas were released. No fleas were removed from the rabbits until the tenth day, which was the time when rabbit B gave birth to its young. Samples of three female fleas were then taken from each rabbit for dissection. Further samples of fleas were taken subsequently and the results are plotted on Fig. 4. Again there was no marked ovarian development among fleas on the nonpregnant rabbit, whilst there was development on the pregnant hosts. The results also show that the ovarian development was not controlled by the number of days the fleas had been on the pregnant host. At the end of 10 days fleas from rabbits C and D were immature, whereas those which had been on rabbit B were fully mature. Similarly, 9 days later fleas on rabbit C were mature, but those on rabbit D were still undeveloped. The ovarian growth curves for fleas on rabbits C and D were of a similar shape to the one shown on Fig. 3, suggesting an abrupt change in the rate of development at a critical point. This type of growth curve was obtained each time a progressive series of samples of fleas was taken from a pregnant rabbit. The approximate critical point was obtained for each series as the point of intersection of two straight lines drawn to provide the best fit to the data and the results are given in Table 1.

Table 1.

The time in the host’s pregnancy, and the corresponding size of the flea ovaries, when the transition in ovarian growth rate occurred

The time in the host’s pregnancy, and the corresponding size of the flea ovaries, when the transition in ovarian growth rate occurred
The time in the host’s pregnancy, and the corresponding size of the flea ovaries, when the transition in ovarian growth rate occurred
Fig. 4.

Graph to show that the development of the proximal oocyte follicles of fleas kept on pregnant rabbits (B, C and D) is dependent upon the stage reached in the pregnancy, and not upon the number of days they have been on the host. At the time of release of the fleas, rabbits B, C and D were respectively 10, 19 and 29 days pre-partum. Line A relates to fleas from a nonpregnant rabbit. Each point is a mean obtained from the dissection of three fleas.

Fig. 4.

Graph to show that the development of the proximal oocyte follicles of fleas kept on pregnant rabbits (B, C and D) is dependent upon the stage reached in the pregnancy, and not upon the number of days they have been on the host. At the time of release of the fleas, rabbits B, C and D were respectively 10, 19 and 29 days pre-partum. Line A relates to fleas from a nonpregnant rabbit. Each point is a mean obtained from the dissection of three fleas.

These results suggest that pregnant rabbits supply a factor required for the initiation of ovarian development in the flea, and that this factor is absent, or below a necessary threshold level, until the last week of the pregnancy. (An alternative explanation would appear to be that an inhibitory factor is present at all times except during the last week of the pregnancy.)

The final rate of ovarian development of the fleas kept on a pregnant rabbit was such that they were mature, or nearly so, at the time of parturition. Some of the fleas dissected 12 hr. pre-partum had a chorion developing on the proximal oocyte in each ovariole of the more mature type. Chorionated oocytes were seldom found earlier in the pregnancy than this.

Every female flea taken from a pregnant host right up to the time of parturition was still virgin, despite the fact that male and female fleas were frequently attached to the host in adjacent positions. The behaviour of both male and female fleas changed abruptly at the time of parturition. Instead of remaining on the adult rabbit, most detached themselves and entered the nest within a few hours of the birth. Fleas were seen to leave the head of a rabbit during this period simply by passing to the surface of the fur, moving over the tips of the hairs until they reached the cheeks and then dropping. As the rabbit was usually cleaning the new-born young, or rearranging the nest at this time many fleas dropped among the nestlings, upon which they fed avidly. Every female flea removed from a nest, even within 12 hr. of parturition, had been impregnated, and had either ovulated from one or both groups of ovarioles, or else had fully developed, chorionated oocytes on the point of ovulation in the more mature group of ovarioles.

As a consequence of this ovarian maturation pattern on the pregnant adult, and of the change of behaviour of the fleas at parturition, fertilized eggs were actually being laid in the nest almost from the time the young rabbits were born. The presence of suitable larval food and microclimatic conditions in the nest, compared with situations where eggs might be dropped by fleas remaining attached to the host, indicates some obvious advantages for these reproductive mechanisms.

In another experiment fifty unfed, virgin, female fleas were released on a pregnant rabbit which was 7 days pre-partum to determine whether normal ovarian maturation and desertion of the doe at parturition would occur in the absence of male fleas. On examination of the doe shortly after the birth of the young only three fleas, gravid in appearance, could be found; each had nearly mature ovaries with chorions forming on the proximal oocytes in the group of larger ovarioles. At the time the doe was examined a quick inspection of the nest and nestlings showed that at least thirteen gravid females had left the doe and reached the nest. Three of the latter fleas were placed in separate specimen tubes for 2 hr. before dissection, and during this period two of them deposited batches of eggs, normal in appearance although, of course, not fertilized. Each of these three fleas had its stomach well filled with fresh blood, very probably taken from the nestlings. It is concluded that the presence of male fleas is not essential for normal ovarian maturation, movement into the breeding nest of the host, or subsequent oviposition, although unfertilized eggs do not develop.

Table 2 provides a scheme for the recognition of four stages of ovarian development, defined according to the appearance and size of the proximal oocyte follicles. The most pronounced visible change in the developing ovaries occurs when sufficient yolk has been deposited in the proximal oocytes to make them opaque; this transition occurs at a follicle length of 150μ. On the above classification, ovaries which appear to be in stage 1 may have slightly larger proximal oocytes than some specimens judged on the grounds of slight coloration of the oocyte by yolk to be in stage 2. Also, there is no sharp change in ovarian appearance corresponding to the size of the proximal oocyte at the time the critical point is reached.

Table 2.

The stage of ovarian maturation related to the appearance and size

The stage of ovarian maturation related to the appearance and size
The stage of ovarian maturation related to the appearance and size

During stage 3 of ovarian development, usually when the proximal oocyte follicle is 200-250μ long, a secretion appears in the cavity of the pedicels, and later also in the lateral oviducts. The secretion is granular, opaque and white, not unlike yolk in appearance. It must be produced by the generalized epithelial cells of the oviducts and pedicels as there is no specialized glandular tissue present.

The external appearance of the female flea alters during stage 3. The abdomen becomes more and more swollen and the colour changes from the normal dark brown to a pale orange-brown. These effects are produced by the growth of the ovaries owing to deposition of yolk, which is light-reflective, and the greater transparency of the abdominal wall as the inter-segmental membranes are stretched.

2. Fleas on a post-par turn rabbit

Most fleas left the adult doe within a few hours of parturition, but the changes in the ovaries of those that did not do so is of interest. It might have been expected that the eggs in their ovaries would continue to mature and that these would be ovulated by the flea whilst still on the doe. This did not occur, and instead the ovaries commenced to regress shortly after the parturition of the host. Yolk was resorbed from all the larger, yolky oocytes, of which there are from three to five in each ovariole in the mature ovary, and the secretion in the pedicels and oviducts disappeared. The follicular cells of each regressing oocyte increased in length and invaded the oocyte, which was gradually destroyed. During this process the muscles of the ovariolar sheath contracted and relaxed rhythmically causing a piston-like action by the more immature follicles on the regressing ones, which were finally compressed into a short length of diffuse tissue. The regression was completed, so that the ovaries returned to stage 1 or ‘early’ stage 2, within about 5 days.

The processes of regression commenced at the largest (i.e. proximal) oocyte in each ovariole and gradually spread to the adjacent yolky oocytes so that a definite sequence of oocyte destruction was set up. This could be determined by measuring the lengths of the largest oocytes, which still appeared to have a normal shape and follicular epithelium, present in the ovaries of fleas removed and dissected at various intervals after the parturition of their host. As an example Table 3 indicates the lengths, after various intervals, of the largest, healthy (non-resorbing) oocyte follicles present in the ovaries of fleas that did not leave their two respective hosts at parturition.

Table 3.

To show the decline in ovarian maturity among fleas remaining on rabbits after parturition

To show the decline in ovarian maturity among fleas remaining on rabbits after parturition
To show the decline in ovarian maturity among fleas remaining on rabbits after parturition

These observations that ovarian regression occurred in fleas remaining on the postpartum rabbit were necessarily based on the dissection of only a few specimens, as most passed into the nest. It was arguable that as these fleas did not leave the host at parturition they were abnormal and hence no valid conclusions could be made. However, it did appear possible that the factor postulated to be present in the pregnant rabbit disappeared soon after parturition, and that the maintenance of this factor was essential if the ovaries were not to regress.

The hypothesis that the factor was absent from the adult after parturition was tested by releasing twenty-five female and eighteen male fleas on to a rabbit which was a few hours post-partum. The fleas used had previously fed for 41 days on a non-pregnant rabbit and three females were dissected prior to releasing the others on the post-partum rabbit; they were non-impregnated and had a mean proximal oocyte follicle length of 78/4, i.e. the fleas were still quite immature. Previously fed fleas were used as it had been found from other work (see §3) that the ovaries of such fleas could respond more quickly than those of unfed fleas to the presence of the maturation factor. Samples of female fleas were recovered from the post-partum rabbit at intervals and dissected; none had been impregnated. The mean lengths of the proximal oocyte follicles are given in Table 4. These results show that there was only a very slight growth of the ovaries up to 3 days from release and then a decline to the original dimensions. It thus appears that the maturation factor is effectively absent from the post-partum rabbit.

Table 4.

To show the lack of any substantial ovarian development among immature fleas which were released on a rabbit shortly after parturition

To show the lack of any substantial ovarian development among immature fleas which were released on a rabbit shortly after parturition
To show the lack of any substantial ovarian development among immature fleas which were released on a rabbit shortly after parturition

3. Fleas in the rabbit breeding nest

Fleas were always to be found in the nest and on the nestlings within a few hours of the parturition of a rabbit on which these and other fleas had been released during or before its pregnancy, but the proportion of fleas leaving the host after parturition has not been determined. Although the number of fleas originally established on any test rabbit was known, the numbers that got dislodged subsequently could not be determined easily. In many experiments a definite number of fleas was released on a pregnant host and samples were removed subsequently for dissection. The handling of the rabbit each time these samples were being obtained often led to the rabbit scratching itself vigorously when freed in its hutch. Scratching was done with the claws of a fore paw and after passing over an ear or the forehead the paw was taken quickly to the open mouth. Whilst it was unlikely that many fleas were eaten it did seem probable that some were dislodged by this behaviour and may then have been removed during the routine cleaning of the hutch.

Whereas several fleas were often visible, fixed on the ears, up to the time of parturition, most had disappeared soon afterwards. Those remaining had sometimes detached their mouth parts from the rabbit and were possibly also about to desert the host. The time at which a pregnant doe commenced to construct its nest was variable, ranging from five days to a few hours pre-partum; a quick examination of these nests before parturition occurred invariably failed to detect any fleas.

The data in Table 5 provide some indication of the distribution of female fleas after the parturition of their original host. The behaviour of the male fleas is believed to be generally similar. The number of fleas recorded as being on the rabbit at the time of parturition is the maximum number that could have been present assuming none was dislodged and lost during the pregnancy. It is obvious that this number exceeds the total known either to have remained on the post-partum doe, or to have entered the nest. The numbers in the nest are minimum numbers and refer to fleas actually removed for dissection. No search was made for fleas that died in the nest after an exhaustive spell of egg laying.

Table 5.

To show that most fleas originally on a pregnant host enter the nest after parturition

To show that most fleas originally on a pregnant host enter the nest after parturition
To show that most fleas originally on a pregnant host enter the nest after parturition

For 1-2 days after the birth of the young the fleas that entered the nest appeared to spend more time on the nestlings than they did subsequently. The flanks and sacral regions of the nestlings were the locations most favoured by feeding fleas and these parts became thickly spattered with flea excrement. Whilst feeding, gravid females were frequently seen in copula. Female fleas were impregnated soon after entering the nest; every specimen examined had spermatozoa in its spermatheca, even if it had been in the nest for only a few hours.

The fleas recorded in the final column of Table 5 had almost certainly returned tb the doe, after some days of oviposition in the nest, because they had not been found when the doe was examined 1-2 days post-partum. They were also invariably impregnated, and copulation is only known to occur in the nest. The condition of the ovaries of these returning fleas was variable and presumably depended upon how soon they were dissected after they had left the nest. Occasional specimens had healthy, well-developed yolky oocytes and had probably only recently left the nest. The majority of those dissected had ovaries with obviously resorbing oocytes; the final stage in this regression of previously mature ovaries was when the proximal oocytes were in stage 1 or ‘early’ stage 2 and there was a typical corpus luteum at the base of each ovariole.

Ovulation of the batch of eggs that was mature, or nearly so, at the time the fleas left the adult rabbit and entered the nest was followed by the maturation of further eggs. In two experiments several female fleas were recovered from the nest 14 days after parturition and all contained maturing eggs.

The number of eggs that can be laid by a rabbit flea has not been determined. It could be some hundreds as each of the ten to twelve ovarioles in fleas that have never ovulated usually contain at least fifteen oocyte rudiments arranged as a single row (end to end) in the vitellarium and there are more, irregularly arranged, in the region adjacent to the germarium. It was noticeable that female fleas removed from the nest had fewer oocyte rudiments after 1-2 days of ovulation than prior to its onset, presumably indicating that any replacement of oocytes from the ‘irregular zone ‘or the germarium was at first a slower process than maturation and ovulation. The example given in Table 6 shows that the numbers of oocyte rudiments did not continue to fall although ovulation proceeded for at least 14 days. This may have been due to a reduction in the rate of egg production after the first few days.

Table 6.

To show that the numbers of oocyte rudiments (arranged as a single row) decrease during the first few days of ovulation in the nest but then remain at a fairly constant level

To show that the numbers of oocyte rudiments (arranged as a single row) decrease during the first few days of ovulation in the nest but then remain at a fairly constant level
To show that the numbers of oocyte rudiments (arranged as a single row) decrease during the first few days of ovulation in the nest but then remain at a fairly constant level

The appearance of the germaria after periods of ovulation was not always similar in every specimen dissected; the germarium might remain narrow and pointed anteriorly, or become shorter and broadly spatulate, but it is not known why this was so. Occasionally specimens were seen in which the number of oocyte rudiments in the germarium was so reduced as to suggest that they were nearly spent and could not be continually replaced in adult life (see also §6 on male fleas).

One flea that was taken from a nest 212 days after parturition had one abnormal ovariole in which there were the remains of five eggs. Most of the yolk had disappeared from them but the rather flattened chorions clearly indicated that five oocytes had matured and then failed to pass through the pedicel into the lateral oviduct. This observation suggested that in each of the other ovarioles at least five eggs had matured and had been successfully ovulated during the first 212 days in the nest.

It is unlikely that such a high rate of egg production could be maintained for long. The mature ovaries of female fleas taken from a pregnant host shortly before parturition usually have from three to five yolk-opaque oocytes (i.e. oocytes with length > 150μ.) in each ovariole. Typical examples of the lengths of the larger oocyte follicles in the two groups of ovarioles are given in Table 7. (These data show how the production of mature eggs alternates between the two groups of ovarioles.) After a day or so in the nest the numbers of yolky oocytes have fallen to two in each ovariole of the one group, and one in each of the other group. This arrangement persists as long as the fleas remain in the nest, because as the larger of the two yolky oocytes in the ovarioles in the one group reaches maturity and is ovulated a second oocyte in each ovariole of the other group has reached the stage of yolk opacity. Thus it appears that the speed at which development is completed among the three to five largest oocytes present at parturition is markedly greater than the rate of growth of the oocytes destined to replace them as ovulation occurs. This tends to confirm the conclusion, already drawn from observations on the numbers of oocyte rudiments present, that there is a falling off in the rate of egg production after the first few days in the nest.

Table 7.

The lengths of the larger oocyte follicles of two fleas taken from a pregnant rabbit shortly before parturition and of two fleas from the nest after about 2 days ovulation to show that there are more oocytes at an advanced stage of development among the former fleas

The lengths of the larger oocyte follicles of two fleas taken from a pregnant rabbit shortly before parturition and of two fleas from the nest after about 2 days ovulation to show that there are more oocytes at an advanced stage of development among the former fleas
The lengths of the larger oocyte follicles of two fleas taken from a pregnant rabbit shortly before parturition and of two fleas from the nest after about 2 days ovulation to show that there are more oocytes at an advanced stage of development among the former fleas

The results already described have shown that there was a major difference in the functioning of the ovaries of fleas which entered the nest as compared with those which remained on the post-partum rabbit, or returned to it after a period in the nest. In the first case eggs were produced, and continued to be produced, by those fleas still in the nest for up to at least 14 days, whereas those staying on the post-partum rabbit resorbed their larger oocytes and did not produce any eggs. At this stage it appeared that either the nestlings provided the necessary maturation factor, or else once mature (and impregnated) fleas could continue to produce further eggs while in the nest, although they could not if they returned to the doe. Experiments were therefore done to determine if the maturation factor was present in nestling rabbits.

Each experiment involved releasing a number of marked fleas into a nest among the nestlings, at a known time after the rabbits were born, and then subsequently attempting to recover samples of the fleas for dissection. In most of the experiments the fleas used had previously been kept for a known period on a non-pregnant rabbit. The female fleas were still virgin and had ovaries in stage 1, but the ovaries were slightly larger and the oocyte rudiments were better organized than they had been when the fleas emerged from their cocoons and were released on the non-pregnant rabbit. The fat-body reserves had also been increased. Previously fed fleas were usually chosen as results indicated that they could respond to the presence of the maturation factor more rapidly than unfed fleas. The recovery of female fleas from the nests for dissection and examination for any ovarian development and impregnation was frequently difficult, especially as the age of the nestlings increased. If the nestlings were more than 7 days old when the fleas were released most of the fleas passed on to the adult rabbit instead of remaining in the nest to feed on the nestlings. Thus in one experiment when thirtythree female and fifteen male fleas were placed in the nest of 12-day-old rabbits only one female could be found in the nest 1 day later whereas twenty-three female and eleven male fleas were combed from the parent doe. The experimental results are summarized in Table 8.

Table 8.

Summary of experiments in which immature fleas were released into rabbit nests and subsequently examined to determine whether any ovarian development or impregnation had occurred

Summary of experiments in which immature fleas were released into rabbit nests and subsequently examined to determine whether any ovarian development or impregnation had occurred
Summary of experiments in which immature fleas were released into rabbit nests and subsequently examined to determine whether any ovarian development or impregnation had occurred

The results obtained for fleas recovered from the nest in Expts. 1, 2 and 8 show that substantial ovarian development (from stage 1 to at least stage 3) had occurred within 24 hr. of their release among new-born young. In Expt. 8 no male fleas were released and it was apparent that their absence in no way affected ovarian development or oviposition as shrivelled, unfertilized eggs were subsequently found in the nest. Similar rapid development was found when the nestling rabbits were initially 5 days old (Expt. 3), 612 days old (Expt. 4) and perhaps also when 7 days old (Expt. 5), but this result depends upon one flea. By contrast, only one out of four fleas released in the nest of 812-day-old rabbits (Expt. 6) had developed to stage 3 within 24 hr., although the other three had commenced to mature. The single flea recovered from the nest of 12-day-old rabbits (Expt. 7) was undeveloped. Even after 2 days from their release in the nest of new-born young (Expt. 9) only one out of three previously unfed fleas had developed to stage 3—a slow response compared with that obtained when using fleas previously fed on a non-pregnant animal. It is concluded that a factor involved in the development of the flea ovaries is provided by recently born rabbits.

Only a few of the fleas recovered from the parent does, after release in their nests, had signs of ovarian development. This development doubtless occurred whilst the fleas were in the nest, evidence for this being that in several cases the fleas were also impregnated and some fleas had clearly resorbing oocytes whereas the ovaries of those remaining in the nest never showed any signs of regression.

The rate of ovarian development of previously fed, but immature fleas when placed among new-born rabbits was much more rapid than expected from a consideration of the data on the normal maturation of fleas on a pregnant host (Table 9). In the latter case obvious development commenced at a ‘critical point’ about 6 or 7 days prepartum and was normally more or less completed in just under a week so that the first ovulations occurred in the nest soon after parturition. In each of the three female fleas of nestling Expt. 2, dissected after 2 days in the nest, mature eggs had been ovulated from at least one group of their ovarioles. In nestling Expt. 8 one flea had ovulated a batch of eggs within 1 day. In nestling Expt. 9 when previously unfed fleas were placed in the nest, the development was slower and even after 5 days one flea had not reached maturity and had not ovulated any eggs. This latter result probably merely indicates that some days of feeding, even in the presence of the necessary maturation factor, are required by previously unfed fleas before their ovaries, fat bodies and other organs are in a condition that allows the required mobilizations for vitellogenesis to occur. However, the difference between the steady development of the ovaries of fleas on pregnant rabbits during the final 7 days of the pregnancy, and the mere 1-2 days required on new-born young suggests that either a different mechanism is involved (perhaps a different maturation factor), or that the response to the factor is quantitative and that the factor is at a relatively high concentration in nestlings as compared with the pregnant adult.

Table 9.

To contrast the rates of ovarian development up to the time of ovulation between fleas on a pregnant host and on new-born nestlings. In nestling Expts. 2 and 8 the fleas had previously been fed on a non-pregnant adult rabbit but in Expt. 9 they were unfed

To contrast the rates of ovarian development up to the time of ovulation between fleas on a pregnant host and on new-born nestlings. In nestling Expts. 2 and 8 the fleas had previously been fed on a non-pregnant adult rabbit but in Expt. 9 they were unfed
To contrast the rates of ovarian development up to the time of ovulation between fleas on a pregnant host and on new-born nestlings. In nestling Expts. 2 and 8 the fleas had previously been fed on a non-pregnant adult rabbit but in Expt. 9 they were unfed

4. Fleas on a non-pregnant female rabbit, and on a male rabbit

As described in § 1, and illustrated in Figs. 3 and 4, there was little change in the size of the ovaries of female fleas kept for up to a month on non-pregnant doe rabbits. In another experiment one hundred female and seventy-five male, unfed, virgin fleas were released on a non-pregnant doe and samples were only removed for dissection after long intervals. The results shown in Table 10 indicate that the ovaries did not develop beyond stage 1 even after a long period on the host. The absence of corpora lutea and lack of impregna-tion in all the specimens dissected were further evidence that fleas kept on a non-pregnant host never reproduce.

Table 10.

To show the absence of any substantial ovarian development among fleas kept on a non-pregnant doe rabbit for a long period

To show the absence of any substantial ovarian development among fleas kept on a non-pregnant doe rabbit for a long period
To show the absence of any substantial ovarian development among fleas kept on a non-pregnant doe rabbit for a long period

In another experiment similar batches of forty-five female and twenty male, unfed, virgin fleas were released on a young adult male rabbit and on a pregnant rabbit 1412 days pre-partum. Samples of female fleas were taken from both rabbits for dissection at the intervals after release indicated in Table 11. Whereas the female fleas on the pregnant host had nearly mature ovaries by the time of parturition and then proceeded to undergo normal reproduction in the nest, the ovaries of the fleas on the male rabbit did not develop beyond stage 1; as was the case with immature fleas released on a non-pregnant doe, or a recently post-partum doe, no corpora lutea were produced and the spermathecae were always devoid of spermatozoa. Identical results were obtained in two further experiments in each of which similar groups of previously unfed, virgin fleas were released on male, non-pregnant and pregnant rabbits. It is concluded that the factor necessary for ovarian development is absent in male rabbits and non-pregnant female rabbits.

Table 11.

To contrast the lack of substantial ovarian development of fleas kept on a male rabbit with the development among fleas on a pregnant rabbit and after parturition in its nest

To contrast the lack of substantial ovarian development of fleas kept on a male rabbit with the development among fleas on a pregnant rabbit and after parturition in its nest
To contrast the lack of substantial ovarian development of fleas kept on a male rabbit with the development among fleas on a pregnant rabbit and after parturition in its nest

5. The transfer of fleas from a pregnant rabbit, a few days pre-partum, to other rabbits

A set of experiments was carried out to investigate more fully whether the factor apparently provided by a pregnant rabbit acted as a trigger setting in motion in the flea developmental processes that would then continue in its absence, or whether its presence was required throughout the period of ovarian development. Each experiment was commenced by releasing 100 female and forty male, unfed, virgin fleas on a pregnant host. Samples of the female fleas were removed after various intervals and dissected to determine their stage of ovarian development. When stage 3 had been reached as many fleas as possible were removed from the rabbit and batches were released on other host rabbits selected to represent various reproductive states, including male, non-pregnant female and females at several different stages of pregnancy. Samples of three female fleas were taken from each of the second hosts after various intervals to see what changes were occurring in the reproductive organs. The results of three series of experiments are summarized graphically in Figs. 5-7.

Fig. 5.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to a male rabbit, to a different pregnant rabbit, and by starvation. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed for dissection. Each point is a mean based upon the dissection of three fleas.

Fig. 5.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to a male rabbit, to a different pregnant rabbit, and by starvation. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed for dissection. Each point is a mean based upon the dissection of three fleas.

Fig. 5 (curve A) shows that maturing female fleas transferred from a 512-day prepartum host to a male rabbit underwent rapid ovarian regression. The resorption of the larger oocytes was completed within 5 days, so that after this period there was no significant difference in the state of development of the ovaries of the fleas on the male rabbit and of the fleas taken from the non-pregnant control rabbit. The oviducal secretion disappeared within 3 days.

There was a generally similar ovarian regression when fleas were transferred to a non-pregnant doe rabbit (Fig. 6(A)), but the last sample of specimens was dissected only 4 days after the transfer and the resorption was not fully completed by then. Resorption had commenced in under 1 day from the time of transfer; it involved all the proximal oocytes in both groups of ovarioles after 1 day, but did not extend to the second oocytes until after the second day. It appeared likely that the pattern of resorption was the exact reversal of that of maturation; thus resorption of the group of larger proximal oocytes commenced first, to be followed by the other proximal oocytes and then in due course back to the second oocytes in the first group.

Fig. 6.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to a non-pregnant rabbit, and to a different pregnant rabbit. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed. Each point is a mean based upon the dissection of three fleas.

Fig. 6.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to a non-pregnant rabbit, and to a different pregnant rabbit. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed. Each point is a mean based upon the dissection of three fleas.

Fleas transferred to a pregnant doe which was 17 days pre-partum (Fig. 7(A)) underwent initial ovarian regression. Oocyte resorption reduced their state of development within 4 days to a level roughly corresponding to that which the fleas would have had on the original host when it was at 13 days pre-partum.

Fig. 7.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to rabbits at different stages of pregnancy, and by starvation. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed. Each point is a mean based upon the dissection of three fleas, except where indicated (2) when only two specimens were available.

Fig. 7.

Graph to show the changes in the ovarian development of maturing fleas produced after their transfer to rabbits at different stages of pregnancy, and by starvation. The arabic numerals on the graph indicate the number of days the subsequent pregnant host was pre-partum when fleas were removed. Each point is a mean based upon the dissection of three fleas, except where indicated (2) when only two specimens were available.

In one transfer, shown as curve B on Fig. 7, fleas were taken from the original pregnant host when it was 212 days pre-partum and placed on a 10-day pre-partum host. The three fleas dissected 2 days after transfer had ovaries undergoing regression. In one flea all the oviducal secretion had disappeared and in the other two a little remained, but only in the pedicels. The yolky proximal oocytes present in both groups of ovarioles at the time of transfer had been destroyed, and the only evidence of their previous existence was a patch of diffuse tissue in each ovariole just anterior to the pedicel. A typical resorbing yolky oocyte was present in half the ovarioles between the patch of diffuse tissue and the first healthy oocyte. This latter oocyte was smaller than the proximal oocyte in the other ovarioles and provides another indication that resorption alternates between each group of ovarioles. The sample dissected 4 days after transfer had no oviducal secretion remaining, and the proximal oocytes were smaller than those in the sample dissected 2 days previously. However, regression now appeared to be completed as no further oocytes were being actively resorbed, although the ovaries were still markedly larger than those of fleas on the non-pregnant control. Three days later, when the host was 3 days pre-partum, the fleas dissected had signs of renewed ovarian development and two of them had traces of secretion reappearing in the pedicels. That ovarian regression had occurred could be detected from the patch of diffuse tissue, now slightly yellowish, persisting between the pedicel and the proximal oocyte in each ovariole. Only two female fleas could be found on the rabbit 9 days after transfer and both had ovaries that were nearly mature, and there was abundant oviducal secretion. The diffuse tissue at the base of the ovariole had been compressed further by the growth of the proximal oocytes and now appeared as a narrow band of slightly yellowish tissue. These results indicated that ovarian regression occurred after the transfer up to the time when the second pregnant host would be expected to have passed the ‘critical stage ‘and could then itself provide the factor for ovarian maturation.

That initial ovarian regression can be followed by renewed development was confirmed by two further transfers to pregnant rabbits which were respectively 6 days and 312 days pre-partum. In the first transfer (Fig. 5 (B)) fleas dissected after 3 days showed signs that ovarian regression had occurred by the absence of oviducal secretion and the presence, proximally in each ovariole, of patches of diffuse tissue. However, the resorption already appeared to have stopped in two of the specimens and to be nearly completed in the third. In two samples taken subsequently rapid development had occurred and mature chorionated eggs were present in two of the three fleas removed from the host actually during parturition. In the second experiment (Fig. 6(B)) the host to which the fleas were transferred would already have passed the ‘critical stage’. In this case the growth of the ovaries was only very briefly inhibited. The oviducal secretion was never withdrawn and in five of the nine specimens examined there was no evidence that any oocyte resorption occurred. In the other four specimens the proximal oocytes in just one group of ovarioles had been resorbed.

These results indicate that the maturation factor does not act solely as a trigger mechanism. The part-mature ovaries of transferred fleas regressed unless, or until, the second host could itself be expected to provide the factor.

Two experiments were carried out in each of which nine female fleas were taken from the first host when their ovaries were maturing and were then allowed to starve for periods of up to 7 days. The fleas were placed in groups of three in corked 3 x 1 in. glass phials each provided with a slip of paper on which the fleas usually rested. The tubes were kept at room temperature in an empty rabbit hutch in the rabbit house used for all the work on rabbit fleas. The fleas were dissected after various intervals and the changes in the ovarian development are shown in Figs. 5(C) and 7(C). On the seventh day of starvation the fleas were alive but rather weak. The results show that there was no immediate regression in the ovaries, such as occurred in nearly every case when similar fleas were transferred to a second host; rather there was an indication that slight ovarian growth occurred at some stage during the first 2-3 days of starvation. The quantity of oviducal secretion was little altered throughout. Only the fleas dissected after 7 days had definite signs of ovarian degeneration; this was more extensive than any regression seen previously, it simultaneously involved one to three oocytes in each ovariole and there was an impression that all the oocytes were about to be resorbed to provide nutriment for the starving insect. Presumably the maturation factor did not disappear as quickly from starved fleas as from fleas which were transferred to a second host, possibly because the fleas were kept at a somewhat lower temperature than they would have been on a rabbit. Alternatively, the fact of being on another host, and perhaps feeding on it, may normally lead to the rapid elimination of the factor.

6. Impregnation and associated phenomena

The spermatheca of every female flea dissected was examined for spermatozoa. Before a group of fleas taken from a culture was released on any test rabbit a sample of three females was removed and dissected. None was impregnated and it was assumed that all the fleas were virgin when the experiments were begun. They remained virgin unless they entered the breeding nest of a doe. Thus all female fleas taken after varying periods from male rabbits or nonpregnant female rabbits were non-impregnated although there had been male fleas in close proximity throughout. Similarly there was no impregnation among fleas taken from a doe during pregnancy, or immediately after parturition. However, every female flea removed from a rabbit nest, after the parturition of the original host, had spermatozoa in its spermatheca. Some impregnated fleas were occasionally found on does a few days after parturition and it is considered certain that these were fleas which had returned to the doe from the nest.

It appeared that the environment of an occupied nest was required for mating but fully mature ovaries were not essential. In an experiment when previously unfed, virgin fleas were released on a rabbit 3 days pre-partum, fleas could be found only in the nest after parturition. Three female fleas were examined only 2 hr. after the birth and each had its spermatheca filled with spermatozoa although the ovaries of two of them had reached only stage 2 of development.

These findings were in agreement with the fact that copulation was never observed among fleas on an adult rabbit, whereas it was seen commonly among fleas in the nest. Copulation appeared to occur most frequently while the female flea was biting a newly born rabbit somewhere on the hindquarters, although pairs in copula were sometimes found free among the fur lining the nest. The position adopted by rabbit fleas during copulation is similar to that described for other species of fleas (Holland, 1955). The posterior segments of the abdomen of the male are arched dorsad so that although the male is positioned beneath the female the complex male terminalia can come into apposition with the female genital chamber.

Data given in Table 8 show that when virgin male and female fleas which had previously been fed for several weeks on a non-pregnant rabbit were placed in a nest with new-born young the females were impregnated within 24 hr. A similar result was obtained when the nest contained 5-day-old young, but not when they were 7, 812 or 12 days old. When unfed male and female fleas were placed in a nest with new-born young the females were not impregnated after 2 days, but they were after 5 days. There was no impregnation within 3 days of the release of similar fleas among 7-day-old rabbits.

The male flea has a pair of pyriform, saccular testes each normally containing many cysts of spermatozoa. If the contractile outer wall of one of the testes is deliberately ruptured all the spermatozoa present flow out and appear to be active and mature, even with freshly emerged fleas, presumably indicating that maturation of the spermatozoa occurs during the pupal stage. Posterior to the cysts of spermatozoa the wall of the testis also surrounds a convoluted tube, reminiscent in appearance of the mammalian epididymis. This tube is continuous with, but mostly of greater diameter than, the vas deferens. In the newly emerged flea this ‘epididymal tube’ is empty but spermatozoa enter the tube if the testis is gently compressed. Further compression of the testis fails to force any spermatozoa into the vas deferens and instead the sheath of the testis ruptures. The impression gained is that there is a blockage in the duct at the point where it passes through the sheath of the testis. Male fleas removed from adult rabbits may have spermatozoa in the epididymal tube but still the blockage remains; however, fleas taken from a pregnant host shortly before parturition or from the nest after parturition appear to have no blockage as spermatozoa can frequently be forced right through the vas deferens and ductus ejaculatorius and out of the gonopore.

Apparently there is no replacement of the spermatozoa transferred during copulation and the testes gradually become depleted, and may even be entirely emptied. The spermatheca of every female flea taken from a nest during the first week after parturition was found to be tightly packed with spermatozoa. After some 10 days in the nest, there were far fewer spermatozoa in the spermathecae and this fact was correlated with the depleted appearance of the testes of the male fleas. Individuals in which the testes had become spent probably did not survive long, but they were occasionally found on the adult doe from 8 days post-partum onwards. Table 12 summarizes the findings for impregnation and male reproductive condition for nestling Expt. 2 in which thirty female and twelve male, virgin fleas, previously fed on a non-pregnant doe, were placed among new-born rabbits.

Table 12.

Indicating the decline in the fecundity of male fleas during the days immediately after their entry into the rabbit nest, and the consequential decrease in the numbers of spermatozoa in the spermathecae of ovulating female fleas

Indicating the decline in the fecundity of male fleas during the days immediately after their entry into the rabbit nest, and the consequential decrease in the numbers of spermatozoa in the spermathecae of ovulating female fleas
Indicating the decline in the fecundity of male fleas during the days immediately after their entry into the rabbit nest, and the consequential decrease in the numbers of spermatozoa in the spermathecae of ovulating female fleas

An experiment was carried out to find how long unused spermatozoa persisted in the spermatheca. Forty female fleas were removed from a nest 5 days after the parturition of their original host. Three were dissected and each had maturing eggs and abundant spermatozoa in its spermatheca. The remaining fleas were released on a non-pregnant doe and samples of three were recovered after 14, 21 and 28 days. In every case the ovaries had undergone a complete regression to stage 1 and contained typical corpora lutea as evidence of this regression. The contents of the spermatheca of each flea were as shown in Table 13. It was apparent that the number of spermatozoa in the spermatheca decreased rapidly after the fleas were placed on the adult rabbit despite the fact that ovulation would have stopped almost immediately after the transfer.

Table 13.

To show the rapid decrease in the numbers of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked

To show the rapid decrease in the numbers of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked
To show the rapid decrease in the numbers of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked

A somewhat similar result was obtained in another experiment after seventeen female and seven male fleas were transferred from a nest containing 5-day-old rabbits to a rabbit 5 days pregnant (not the parent of the nestlings). The experiment was intended to simulate the conditions of reproduction of the wild rabbit, in which parturition is normally quickly followed by another pregnancy, but unfortunately the last host had an abnormal pregnancy which terminated in an abortion after 33 days. The results given in Table 14 for the spermathecal contents indicate that spermatozoa did not persist in the spermatheca for much more than 30 days. Although male fleas, which had presumably mated successfully whilst in the nest before transfer, were present there was no suggestion that they mated when on the adult host.

Table 14.

To show the limited period of persistence of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked

To show the limited period of persistence of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked
To show the limited period of persistence of spermatozoa in the spermathecae of fleas in which a cessation of egg production had been provoked

7. Follicular relics; their formation, appearance and pigmentation

In some species of insects the follicular relics persist for a considerable period after ovulation and they often become pigmented. In recent years considerable attention has been given to the recognition of these follicular relics (corpora lutea), especially among the blood-sucking Díptera, as they can provide an indication of the number of gonotrophic cycles completed and consequently of the physiological, and sometimes calendar, age of the individual insect.

The region of the ovariole between the pedicel and the proximal oocyte has been examined in every female flea dissected. In newly emerged and unfed fleas the first oocyte was separated from the pedicel and lateral oviduct by a small plug of distinct cells, the epithelial plug. When the fleas were allowed to feed on a non-pregnant doe, or a male rabbit, the epithelial plugs persisted unchanged for about 2 weeks, during which time there was slight growth of the oocyte rudiments. After this period the group of larger proximal oocytes regressed, the oocytes being invaded by the enlarged follicular epithelial cells. At this stage half the ovarioles in the ovaries contained degenerating follicles proximally and the other half had typical epithelial plugs. In the former ovarioles the original oocyte2 took on the role of proximal oocyte, and was smaller than the proximal oocyte in the latter group, whose oocytes underwent 2-3 days of further growth before they died and were resorbed. After this, ovarioles in neither group contained recognizable epithelial plugs, and the follicles which had regressed first appeared as a patch of diffuse tissue, very slightly coloured and compressed between the pedicel and the new proximal oocyte. The slight growth of the proximal oocyte follicles and their subsequent resorption occurred alternately in the two groups of ovarioles as long as the fleas remained on the rabbits, but it was not possible to detect from the appearance of the area of diffuse tissue how many oocytes had been resorbed in any one ovariole. Thus the ovaries of fleas placed on nonpregnant or male rabbits did not just develop slightly and then enter a resting state; instead they were continually developing oocytes up to a certain size, after which the oocytes died and their position was taken by more immature rudiments.

The ovaries of fleas which were placed on a pregnant host soon after it had been mated showed similar oocyte resorptions after about 2 weeks and these continued until the critical point was reached, after which the proximal oocytes normally developed to maturity. This development compressed the remnants of the resorbed oocytes to such an extent that they were no longer visible in whole-mount preparations. Among fleas placed on pregnant rabbits at about mid-term the first oocyte resorption leading to the disappearance of the epithelial plugs was more rapid and occurred between 7 and 9 days from their release.

Soon after parturition most fleas entered the nest and after any final maturation, and normally fecundation, the females ovulated a batch of eggs, leaving the large, almost colourless remains of the follicular epithelia at the base of the appropriate ovarioles. By the time the second group of ovarioles had developed its proximal oocytes to the stage for ovulation the follicular remnants (relics) in the other ovarioles had shrunk considerably and gradual autolysis converted them into rather diffuse structures, still more or less colourless. The occurrence of subsequent ovulations was recognizable in whole-mount preparations if the dissection was made soon after the ovulation when the follicular epithelium was still visible and before autolysis and compression merged the relics into one. The composite follicular relic in each ovariole of fleas which had been ovulating for several days was pigmented a bright yellow, the typical corpus luteum. This pigmentation appeared after about 4 days in the nest and was fully developed from about 6 days onwards. Similar corpora lutea were developed by non-impregnated fleas when males were deliberately excluded from the nest, and by fleas whose ovaries had matured entirely on the nestlings instead of initially on the parent doe. The experiment detailed in §6, concerning the survival time of spermatozoa in the spermatheca of female fleas transferred to a non-pregnant rabbit after laying eggs for 5 days, indicated that the corpora lutea persisted without any obvious changes for at least 30 days.

As described in §2 any female fleas which remained on a rabbit after parturition underwent a rapid ovarian regression. This regression was normally evident within 1 day; the largest oocytes took on an irregular shape, yolk was withdrawn and the follicular epithelia thickened and invaded the oocytes, which were destroyed. All the larger oocytes were destroyed in turn so that the ovary returned to stage 1 of development. A patch of yellow tissue developed in the basal region of the ovarioles and was generally similar to the corpora lutea present in fleas which had repeatedly ovulated. The only obvious difference was that the latter fleas were normally impregnated. This criterion allowed fleas which had remained on a post-partum doe and developed corpora lutea by resorption of mature oocytes to be distinguished from those that had returned to the doe after laying eggs in the nest.

The experiments involving the transfer of partly mature fleas to other hosts (§5) showed that little pigmentation resulted when fairly small yolky oocytes were resorbed, but there was more coloration when the oocytes had been larger at the time resorption occurred.

It has been shown that although individuals of the rabbit flea usually remain on a host, thus having an abundant source of food available at all times, they do not necessarily produce any mature eggs. Indeed, fleas even when kept for long periods on unmated female rabbits or on male rabbits never produce any eggs. This is in contrast to the findings for several other species of fleas which have been cultured under laboratory conditions. Owing to their ready availability and ease of handling white rats and mice are frequently used as hosts, and with them Leeson (1932), Hollenbeck (1946), Elbel (1951, 1952), Burroughs (1953) and Stark & Kartman (1957) have successfully bred Xenopsylla cheopis Roths, and at least five other species of rodent fleas. Presumably any individual host used in these breeding experiments was selected indiscriminately of its sex, and there has been no suggestion of any greater fecundity among these species of fleas according to whether they were fed on one sex of host rather than the other. The use of suckling mice was preferred by Hecht (1943) and Edney (1945) as these hosts are unable to catch and eat the fleas and thereby reduce the numbers in the culture. Buxton (1948) kept X. cheopis both on adult and on young mice and he likewise noted the considerable losses of fleas resulting from predation by the adult mice. He also found that fleas fed on young mice tended to produce far fewer eggs per day than those fed on adult mice, and that there was a greater delay before fleas started to lay eggs when the mice used were very young. This is the only published work of which I am aware that indicates the fecundity of fleas may vary according to the age of their host, although Poole & Underhill (1953) and Hudson & Prince (1958 a) have shown that fecundity can differ when using alternative species as hosts for the fleas. Buxton (1948) suggested that mammalian sex hormones might somehow be involved in rat flea reproduction as he presumed that the blood of young mice was relatively deficient in these hormones. Rothschild (1961) has suggested that sex hormones of the host may be important in the case of the rabbit flea.

The reproductive mechanisms exhibited by the rabbit flea have definite advantages for the species. Wild rabbits normally live in warrens, and in contrast to the burrows of many mammals these warrens do not contain any nest material or bedding, so that the rabbits rest on bare soil. The only nests that are made are those of the breeding does and in Britain these are usually away from the main warren and at the blind end of a short burrow. These burrows are probably often dug only a few days prior to parturition and the nests are made with an outer covering mostly of grass and an inner lining of hair. The young are suckled in the nest for about 4 weeks, after which the mother deserts them, but the young rabbits may use the nest burrow for a few more weeks. The micro-habitat provided in the nest is far more suitable for the development of the larval stages of the rabbit flea than would be the bare earthen floor of a normal burrow. For completion of their development larvae of many species of fleas require a dietary component present in dried blood, and normally obtained from the faeces of adult fleas as the larvae are themselves incapable of piercing a host and sucking blood (Bacot, 1914; Sikes, 1931; Edney, 1947). If eggs were matured and laid indiscriminately by rabbit fleas while they were still attached to a rabbit these nonadhesive eggs would normally fall on to the surface of the ground or the floor of a burrow; larvae hatching from these eggs would appear to have only slight chances of finding particles of dried blood in such situations. This difficulty is obviated by the fleas only producing mature eggs at the end of their host’s pregnancy when they are likely to be carried to a nest. Their behaviour in respect of attachment to the adult rabbit changes abruptly at the time of parturition and leads to a concentration of males and gravid females loose in the nest. The female fleas in particular feed avidly on the nestling rabbits and the latter become thickly spattered with faeces of part-digested blood which subsequently probably mb off into the nest. Some blood is likely to be lost by the parent rabbit during parturition and the net result is that there is an abundance of dried blood in the nest for any developing flea larvae. Flea larvae also require vitamins of the B complex and these are probably obtained by eating fungi (Sharif, 1948). Fungal growth is likely to be far more prevalent in warm, moist rabbit nests than in normal, well-used burrows.

During the past 25 years much research has been directed towards a fuller understanding of the physiology of insect reproduction, and the importance of the cerebral neurosecretory and endocrine systems and of correct nutrition are now well appreciated (Johansson, 1958). Wigglesworth (1936a) demonstrated that a functional corpus allatum was essential for the deposition of yolk in the oocytes of Rhodnius prolixus Stål by experiments involving decapitation, implantation and parabiosis. It was concluded that a ‘yolk-forming hormone’ is secreted by the corpus allatum of the adult R. prolixus, and it was shown later (Wigglesworth, 1948, 1961) that this could be the same substance as the ‘juvenile hormone’ secreted during the first four nymphal instars. The proximal oocytes in each ovariole of a specimen deprived of its corpus allatum develop normally up to the stage when yolk deposition should occur and then die and are gradually resorbed, whilst the adjacent, living oocytes take their place and undergo partial development. This sequence of partial development followed by degeneration after allatectomy resembles the course of events when intact rabbit fleas are maintained on rabbits other than nestlings or pregnant does which are less than 7 days pre-partum. This, in turn, suggests that for some reason the corpora allata of the rabbit flea do not secrete ‘yolk-forming hormone’ except when kept on rabbits of the latter types. If this is so it would be interesting to determine how the secretory activity of the flea’s corpora allata is controlled. Clearly an insect has to be suitably nourished for the required hormone synthesis and Wigglesworth (1936a) found that when adult R. prolixus were starved their corpora allata had a shrunken, inactive appearance and the ovaries underwent changes similar to those seen in allatectomized specimens. Similarly, Johansson (1954) found with the milkweed bug Oncopeltus fasciatus (Dallas) that both the ovaries and the corpus allatum were affected by starvation, but implantation of corpora allata from sexually active individuals into starved ones led to egg development by the latter. Thus inactivation of the ovaries by starvation was indirect, and resulted from insufficient activity of the corpus allatum, as in this species the basic materials required for egg production must have been present at the time the active corpora allata were implanted. In the present study the gut of each dissected flea was inspected cursorily and a subjective appraisal made of the state of digestion of the contents. There was no particular indication that fleas kept on male or non-pregnant female rabbits were undernourished.

Some insects, particularly discontinuous feeders, apparently require a meal of a certain minimum size before ovarian maturation is possible and in such cases it is likely that initiation of activity in the corpus allatum is mediated via the central or stomatogastric nervous systems after stimulation of nerve endings by a stretching of the abdomen. Fleas of many species probably normally leave their host after each meal and could thus be considered as discontinuous feeders, but the rabbit flea is more difficult to classify as it tends to stay on a host, often with its mouth-parts inserted in the skin, although not continuously sucking blood. No obvious changes were noticed in the feeding behaviour of rabbit fleas kept on pregnant hosts at a time that corresponded to the ‘critical point ‘in ovarian development, and it is unlikely that a simple alteration in feeding rate is responsible for the change in ovarian function although it would be expected that once fleas commenced to develop their ovaries they would require additional nourishment. Female fleas were more often seen on the cartilaginous swellings at the base and back of the ears towards the end of the pregnancy than at other times. Whether this observation has any significance is uncertain as the ovarian development of these fleas was no different from that of those feeding elsewhere on the ears.

Among insects which are more continuous in their feeding and varied in their diet distension of the abdomen is probably of less importance than is the biochemical composition of the foods taken. For example, female Calliphora erythrocephala (Meigen) will live on a diet of sugar and water but the ovaries only develop when protein is added, although protein itself will not maintain life (Fraenkel, 1940). Strangways-Dixon (1961) has shown that during each cycle of egg production C. erythrocephala selects, first, a protein-rich diet and subsequently a carbohydraterich one. The corpus allatum also undergoes cyclical changes in size and its activity appears to be controlled humorally by the concentration of yolk precursors, derived from the ‘protein’ diet, present in the haemolymph (Strangways-Dixon, 1962). The ‘protein ‘diet used was Mannite and milk, and which of the possible metabolites was involved in the activity cycle of the corpus allatum is at present unknown. Harlow (1956) working with another blowfly found that Marmite provided a non-protein accessory substance, possibly a B vitamin, vital for yolk formation. It is well known that blood does not provide haemophagous insects with a sufficiency of vitamins of the B complex and they are dependent upon the presence of symbionts in the gut unless their larval stages are not restricted to blood feeding (Wigglesworth, 1936b). It is also of interest that Baines (1956) working on the development of symbiont-free Rhodnius prolixus noted that, compared with mouse blood, the blood of rabbits was relatively deficient in some unidentified accessory substance normally provided by the symbionts.

In the case of the rabbit flea vitellogenesis only occurs in fleas present on pregnant does during the final 7 days of their pregnancy or in those released among newly born rabbits, and it is postulated that such hosts provide the fleas with an essential factor. The factor may act in a quantitative manner as complete vitellogenesis can occur in 1-2 days on new-born rabbits but takes about 7 days on the pregnant adult. (No experiments have been done in which previously fed, but immature, fleas are released on pregnant rabbits at differing intervals after the ‘critical stage ‘has been reached to investigate whether the ovarian response is relatively more rapid among those released last. If this occurred it could indicate a quantitative response to an increasing factor stimulus.) The transfer type of experiment showed that the factor must be maintained throughout the period of vitellogenesis, and resorption of previously developing yolky oocytes is apparent within 1 day of transferring fleas to a host that does not provide the requisite stimulus. Impregnation of the female flea has no bearing upon whether or not yolk deposition takes place, although impregnation is important in some insects, e.g. Cimex lectularius L. (Mellanby, 1939).

Some of the more likely paths of action of the factor are as follows. (1) As an external stimulus to the central nervous system. This might explain the rapid initiation of resorption after transfer to an unsuitable host, but if we accept that a graded, quantitative response can occur it requires the existence of a graded external stimulus, which appears less likely. (2) From the haemolymph, either via the nervous system or humorally, to the neurosecretory system and/or the corpus allatum and leading to a release of stored hormone. (3) From the haemolymph to the neurosecretory system and/or the corpus allatum, which is accordingly activated to produce and secrete hormone. (4) From the haemolymph to the ovaries. This might be the case if the factor is an essential metabolite or precursor for yolk formation.

If a ‘yolk-forming hormone’ is produced only when the factor is present in the haemolymph it is possible to envisage that variations in the factor concentration could lead to variations in the rate of hormone production and in turn to variations in the rate of yolk deposition. There is evidence that in several insects response to corpus allatum implantation varies with the quantity of this tissue implanted (Joly, 1945; Johansson, 1958). A simple variation in the quantity of the factor available to fleas kept on different types of hosts would then explain the observed differences in ovarian activity. The transfer experiments imply that, to prevent oocyte resorption, hormone must be maintained at an appropriate level throughout vitellogenesis and that release on an unsuitable host rapidly results in a reduction of hormone concentration below this threshold, whereas enforced cessation of feeding does not. Likewise, Johansson (1958) found with Oncopeltus fasciatus that once vitellogenesis had commenced allatectomy (complete removal of further supplies of hormone) normally led to oocyte degeneration whereas starvation (initially probably only reducing production of hormone) did not.

Fleas which acquire the maturation factor most probably obtain it in their blood meals, although as yet there is no experimental evidence for this. Substantial changes are known to occur in the composition of mammalian blood during pregnancy and the most obvious of these changes concern the concentrations of the steroid hormones and their metabolites, which are generally at a very low level in non-pregnant animals. Using a bioassay based on in vivo myométrial response Schofield (1957, 1960) investigated the interaction of endogenous oestrogen and progesterone in the control of pregnancy and parturition in the rabbit. She concluded that progesterone was the principal hormone determining the course of pregnancy and that its influence on the myometrium was dominant from 1 day after mating until 112 days before parturition. The source of the progesterone in the rabbit, which is a species that cannot maintain pregnancy after ovariectomy, was shown to be principally, if not entirely, luteal although the presence of the placentae was necessary to maintain luteal activity. The level of oestrogen fell within I day of mating and remained low for the first half of the pregnancy, after which it rose to a maximum at the time of parturition. The possible sources of this oestrogen were not investigated. Using the Hooker-Forbes bioassay Zarrow & Neher (1955) found that blood serum taken from the marginal ear vein of female rabbits contained little progestin (progesterone) at the time of mating and again within 1 day after parturition. After mating the concentration rose to a high, plateau level maintained until about 1 week before parturition when a further rise occurred and this highest concentration did not fall until shortly after parturition.

Analytical methods for measuring concentrations of steroid hormones in the blood have been developed only recently, but as more data become available it may be possible to explain the apparent discrepancies which can result from the use of different bioassays. Unfortunately, a quantitative comparison of the hormones in the plasma of the foetus at the time of birth with that of its mother does not appear to be available for rabbits, but it has been shown for man that both progesterone and oestrogen (oestriol) are present at a much higher concentration in the foetal circulation (Aitken, Preedy, Eton & Short, 1958). The rate of metabolism of these hormones in the new-born child is also apparently much slower than in the adult as substantial quantities persist in the blood for some time. Much steroid hormone is produced by the placenta and its loss at parturition is a major cause of the sudden decrease in circulating hormone in the post-partum adult.

In several respects the changes in steroid hormone levels parallel those of the factor postulated to explain the variations in flea ovarian activity. For example, oestrogen levels are very low in males and non-pregnant females, rise in the pregnant female towards term, fall rapidly after parturition and may be very high in the new-born, whilst the factor is demonstrable only towards term and in the nestlings. Even if steroids are involved their influence may not be direct as there is evidence that increasing concentrations of steroid hormone in the blood may produce secondary effects also detectable in the blood and hence possibly by the feeding fleas. Thus Horger & Zarrow (1957) indicate that oestrogen and progesterone play important roles in the fluid metabolism of the pregnant animal and are partly responsible for the anaemia of pregnancy in the rabbit. Sabin, Miller, Smithburn, Thomas & Hummel (1936) showed that new-born rabbits had a macrocytic anaemia which was gradually reduced after the first week of life and that the red cell count characteristic of adult rabbits was reached by the third week. Consequently it would appear that, although from different causes, the blood both of recently born rabbits and of does during the final third of their pregnancies has a relatively high plasma volume/cell volume ratio.

It must be emphasized that the ‘juvenile hormone ‘of insects which appears to be similar to the ‘yolk-forming hormone’ is most probably an acyclic compound, such as a terpene alcohol or derivative (Schmialek, 1961; Wigglesworth, 1961) and the only steroid insect hormone described is one from the brain of Bornbyx mori (L.) (Kirimura, Saito & Kobayashi, 1962). Extracts with juvenile hormone activity have been obtained from a very wide range of living organisms and sources include mammalian ovary, corpus luteum and placenta and 1-day-old rats (Williams, Moorhead & Pulis, 1959); hence it is just possible that the rabbit flea is responding to an increase in substances similar to its own yolk-forming hormone and that this is being received directly in its diet.

Work has been carried out on Spilopsyllus cuniculi because of its economic importance as the principal vector of myxomatosis in Great Britain; I am indebted to the members of the Scientific Sub-Committee of the Myxomatosis Advisory Committee for their encouragement and in particular to the Hon. Miriam Rothschild. She had suspected for several years that the breeding cycle of this flea might be related in some way to the breeding cycle of the rabbit and has suggested several useful lines of investigation. It is a pleasure to record the interest of Mr A. J. B. Rudge, who was the first person to see the larvae of S. cuniculi in the nest of a laboratory rabbit, an observation from which the present study largely stemmed. I have received technical assistance from several other colleagues in the Infestation Control Laboratory, notably from Messrs A. Howard, R. J. C. Page, J. W. Sears and K. L. Turns. I wish to thank Dr I. Thomas and Mr H. V. Thompson for their advice and for reading and criticizing the manuscript.

Aitken
,
E. H.
,
Preedy
,
J. R. K.
,
Eton
,
B.
&
Short
,
R. V.
(
1958
).
Oestrogen and progesterone levels in foetal and maternal plasma at parturition
.
Lancet
,
ii
,
1096
9
.
Bacot
,
A. W.
(
1914
).
A study of the bionomics of the common rat fleas and other species associated with human habitations, with special reference to the influence of temperature and humidity at various periods of the life history of the insect
.
J. Hyg., Camb
.,
13
,
Plague Suppl. 3
,
447
654
.
Baines
,
S.
(
1956
).
The role of symbiotic bacteria in the nutrition of Rhodmus prolixus (Hemiptera)
.
J. Exp. Biol
.
33
,
533
41
.
Burroughs
,
A. L.
(
1953
).
Sylvatic plague studies. X. Survival of rodent fleas in the laboratory
.
Parasitology
,
43
,
35
48
.
Buxton
,
P. A.
(
1948
).
Experiments with mice and fleas. 1. The baby mouse
.
Parasitology
,
39
,
119
24
.
Edney
,
E. B.
(
1945
).
Laboratory studies of the bionomics of the rat fleas, Xenopsylla brasiliensis Baker and X. cheopis Roths. I. Certain effects of light, temperature and humidity on the rate of development and on adult longevity
.
Bull. Ent. Res
.
35
,
399
416
.
Edney
,
E. B.
(
1947
).
Laboratory studies on the bionomics of the rat fleas, Xenopsylla brasiliensis Baker and X. cheopis Roths. III. Further factors affecting adult longevity
.
Bull. Ent. Res
.
38
,
389
404
.
Elbel
,
R. E.
(
1951
).
Comparative studies on the larvae of certain species of fleas (Siphonaptera)
.
J. Parasit
.
37
,
119
28
.
Elbel
,
R. E.
. (
1952
).
Comparative morphology of some rat flea larvae (Siphonaptera)
.
J. Parasit
.
38
,
230
8
.
Fraenkel
,
G.
(
1940
).
Utilization and digestion of carbohydrates by the adult blowfly
.
J. Exp. Biol
.
17
,
18
29
.
Harlow
,
P. M.
(
1956
).
A study of ovarial development and its relation to adult nutrition in the blowfly Protophormia terrae-novae (R.D.)
.
J. Exp. Biol
.
33
,
777
97
.
Hecht
,
O.
(
1943
).
Algunos consejos técnicos para la cría de pulgas
.
Rev. Sanid. Asist. Soc
.
8
,
1163
5
.
Holland
,
G. P.
(
1955
).
Primary and secondary sexual characteristics of some Ceratophyllinae, with notes on the mechanism of copulation (Siphonaptera)
.
Trans. R. Ent. Soc. Lond
.
107
,
233
48
.
Hollenbeck
,
A. H.
(
1946
).
A practical method for mass production and transfer of Xenopsylla cheopis
.
J. Parasit
.
32
,
463
4
.
Horger
,
L. M.
&
Zarrow
,
M. X.
(
1957
).
Certain endocrine aspects of the anaemia of pregnancy in the rabbit
.
Amer. J. Physiol
.
189
,
407
11
.
Hudson
,
B. W.
&
Prince
,
F. M.
(
1958a
).
Culture methods for the fleas Pulex irritons (L.) and Pulex simulons Baker
.
Bull. World Hlth Org
.
19
,
1129
33
.
Hudson
,
B. W.
&
Prince
,
F. M.
(
1958b
).
A method for large-scale rearing of the cat flea, Cteno-cephalides fells fells (Bouché)
.
Bull. World Hlth Org
.
19
,
1126
9
.
Johansson
,
A. S.
(
1954
).
Corpus allatum and egg production in starved milkweed bugs
.
Nature, Land
.,
174
,
89
.
Johansson
,
A. S.
(
1958
).
Relation of nutrition to endocrine-reproductive functions in the milkweed bug Oncopeltus fascia tus (Dallas) (Heteroptera : Lygaeidae)
.
Nytt Mag. Zool
.
7
,
1
132
.
Joly
,
P.
(
1945
).
La fonction ovarienne et son contrôle humoral chez les Dytiscidés
.
Arch. Zool. Exp. Gen
.
84
,
49
164
.
Kirimura
,
J.
,
Saito
,
M.
&
Kobayashi
,
M.
(
1962
).
Steroid hormone in an insect, Bornbyx mori
.
Nature, Land
.,
195
,
729
30
.
Kosminsky
,
R. B.
&
Soloviova
,
N. T.
(
1959
).
A simple method to mark fleas
.
Med. Parazitol
.
28
,
203
5
. [In Russian with English summary.]
Lesson
,
H. S.
(
1932
).
Methods of rearing and maintaining large stocks of fleas and mosquitoes for experimental purposes
.
Bull. Ent. Res
.
23
,
25
31
.
Linduska
,
J. P.
&
Cochran
,
J. H.
(
1946
).
A laboratory method of flea culture
.
J. Econ. Ent
.
39
,
544
5
.
Mead-Briggs
,
A. R.
(
1962
).
The structure of the reproductive organs of the European rabbit flea, Spilopsyllus cuniculi (Dale) (Siphonaptera)
.
Proc. R. Ent. Soc. Land. (A)
,
37
,
79
88
.
Mead-Briggs
,
A. R.
&
Rudge
,
A. J. B.
(
1960
).
Breeding of the rabbit flea, Spilopsyllus cuniculi (Dale); requirement of a ‘factor ‘from a pregnant rabbit for ovarian maturation
.
Nature, Lond
.,
187
,
1136
7
.
Mellanby
,
K.
(
1939
).
Fertilization and egg production in the bed-bug, Cimex lectularius L
.
Parasitology
,
31
,
193
9
.
Poole
,
V. V.
&
Underhill
,
R. A.
(
1953
).
Biology and life history of Megabothris clantoni clantoni (Siphonaptera: Dolichopsyllidae)
.
Walla Walla Coll. Publ. Dept. Biol. Sci. and Biol. Stn
. (
9
),
1
19
.
Rothschild
,
M.
(
1961
).
Observations and speculations concerning the flea vector of myxomatosis in Britain
.
Ent. Mon. Mag
.
96
,
106
9
.
Sabin
,
F. R.
,
Miller
,
F. R.
,
Smithburn
,
K. C.
,
Thomas
,
R. M.
&
Hummel
,
L. E.
(
1936
).
Changes in the bone marrow and blood cells of developing rabbits
.
J. Exp. Med
.
64
,
97
120
.
Schmialek
,
P.
(
1961
).
Die Identifizierung zweier im Tenebriokot und in Hefe vorkommender Substanzen mit Juvenilhormonwirkung
.
Z. Naturf
.
16b
,
461
4
.
Schofield
,
B. M.
(
1957
).
The hormonal control of myométrial function during pregnancy
.
J. Physiol
.
138
,
I
IO
.
Schofield
,
B. M.
(
1960
).
Hormonal control of pregnancy by the ovary and placenta in the rabbit
.
J. Physiol
.
151
,
578
90
.
Sharif
,
M.
(
1948
).
Nutritional requirements of flea larvae, and their bearing on the specific distribution and host preferences of the three Indian species of Xenopsylla (Siphonaptera)
.
Parasitology
,
38
,
253
63
.
Sikes
,
E. K.
(
1931
).
Notes on breeding fleas, with reference to humidity and feeding
.
Parasitology
,
23
,
243
9
.
Stark
,
H. E.
&
Kartman
,
L.
(
1957
).
Studies on Pasteurellapestis in fleas. VI. The laboratory culture of Xenopsylla vexabilis hawaiiensis Jordan, 1932
.
Amer. J. Trop. Med. Hyg
.
6
,
707
11
.
Strangways-Dixon
,
J.
(
1961
).
The relationship between nutrition, hormones and reproduction in the blowfly Calliphora erythrocephala (Meig.). I. Selective feeding in relation to the reproductive cycle, the corpus allatum volume and fertilization
.
J. Exp. Biol
.
38
,
225
35
.
Strangways-Dixon
,
J.
(
1962
).
The relationship between nutrition, hormones and reproduction in the blowfly Calliphora erythrocephala (Meig.). III. The corpus allatum in relation to nutrition, the ovaries, innervation and the corpus cardiacum
.
J. Exp. Biol
.
39
,
293
306
.
Wigglesworth
,
V. B.
(
1936a
).
The function of the corpus allatum in the growth and reproduction of Rhodnius prolixus (Hemiptera)
.
Quart. J. Mier. Sci
.
79
,
91
121
.
Wigglesworth
,
V. B.
(
1936b
).
Symbiotic bacteria in a blood-sucking insect, Rhodnius prolixus Stàl. (Hemiptera, Triatomidae)
.
Parasitology
,
28
,
283
9
.
Wigglesworth
,
V. B.
(
1948
).
The functions of the corpus allatum in Rhodnius prolixus (Hemiptera)
.
J. Exp. Biol
.
25
,
1
14
.
Wigglesworth
,
V. B.
(
1961
).
Some observations on the juvenile hormone effect of famesol in Rhodnius prolixus Stàl (Hemiptera)
.
J. Ins. Physiol
.
7
,
73
8
.
Williams
,
C. M.
,
Moorhead
,
L. V.
&
Pulis
,
J. F.
(
1959
).
Juvenile hormone in thymus, human placenta and other mammalian organs
.
Nature, Land
.,
183
,
405
.
Zarrow
,
M. X.
&
Neher
,
G. M.
(
1955
).
Concentration of progestin in the serum of the rabbit during pregnancy, the puerperium and following castration
.
Endocrinology
,
56
,
1
8
.