FLEA, SPILOPSYLLUS CUNICULI (DALE)

J. Exp. Bil. (969), 5, 95-5 Printed in Great Britain SOME REQUIREMENTS FOR MATING IN THE RABBIT FLEA, SPILOPSYLLUS CUNICULI (DALE) BY A. R. MEAD-BRIGGS AND J. A. VAUGHAN Ministry f Agriculture, Fisheries and Fd, Infestatin Cntrl Labratry, Tangley Place, Wrplesdn, Guild-frd, Surrey {Received ij February 969) INTRODUCTION Whilst mst species f fleas breed in the nests f their hsts, few appear t be as dependent n the sexual cnditin f their hst as is the Eurpean rabbit flea Spilpsyllus cunicuu (Dale). Mead-Briggs & Rudge (i96) reprted that maturatin f the varies ccurred nly in fleas which Were allwed t feed n rabbits in the final stages f pregnancy. It was pstulated that such rabbits prvided a 'factr' which was required befre vitellgenesis culd ccur in the fleas. It was fund subsequently (Mead-Briggs, 96 a) that nestlings up t at least 7 days f age als prvided the 'factr' and Rthschild & Frd (96a, b) demnstrated that hrmnes f the hst were invlved. This mechanism ensures that fleas have mature varies whenever the female rabbit prepares and uses a maternal nest, which is the nly nest made by wild rabbits. Mst f the fleas n the de detach themselves at the time f parturitin and pass int the nest. Cpulatin f the virgin fleas then ccurs in the nest, typically whilst the female fleas are feeding avidly n the hindquarters f the new-brn nestlings. The Hn. Miriam Rthschild had realized that it was imprtant t determine whether male fleas als underwent maturatin n the pregnant hst befre they culd mate. She suggested studying whether cpulatin wuld ccur in the nest between gravid female fleas and males taken frm nn-pregnant r male hsts. The results f experiments t investigate this questin are reprted here, with sme bservatins n the reprductive system f the male flea and n mating behaviur. MATERIALS AND TECHNIQUES The basic experimental prcedure was t put small numbers f virgin male and female fleas tgether in a mating arena, a warmed bx which wuld als accmmdate a recently brn nestling rabbit if required, fr at least 6 hr. Subsequently the spermatheca (receptaculum seminis) f each female flea was excised and examined fr the presence f spermatza as the criterin f successful mating and inseminatin. The undamaged spermatheca was readily teased frm the psterir prtin f a flea which had been divided, in a drp f saline n a glass slide, by a cut directed bliquely thrugh tergites 7-8 and sternites -5. An estimate f the number f spermatza present was made n a scale f nne, -2, -, -, and mre than. The bulga f the spermatheca can be partially paque and if nly a few spermatza are present they 2 Exp. BiL 5, a

96 A. R. MEAD-BRIGGS AND J. A. VAUGHAN may be bscured and the specimen incrrectly classified as nn-impregnated. T ensure that this did nt ccur the bulga was always ruptured, by remving saline frm beneath the cverslip with bltting-paper until sufficient pressure was exerted, when any spermatza present flwed ut Initially the fleas were taken frm cultures shrtly after they emerged frm their ccns and the sexes were separated, rughly by eye and then checked under CO S anaesthesia with a lw-pwer dissecting micrscpe. Batches f 2 fleas f the same sex were then released n individually caged rabbits, and allwed t remain fr up t days. Fr the male fleas the hsts selected were male, nn-pregnant, and pregnant female dmestic rabbits s that fleas given different maintenance regimes wuld be available. In sme cases these fleas were put fr a further 2 hr. with a newbrn rabbit. Gravid female fleas were usually required and were btained by releasing females n recently mated des and recvering them at the time f parturitin, when they all had mature varies. Small brwn rabbits, frm a stck which had been backcrssed ccasinally with a wild buck, were used in the first three series f experiments. In subsequent series New Zealand Whites were used as the mrtality f fleas n them was lwer, their pregnancies were mre reliable, and their litters larger. This change f rabbit stck had n apparent effect n the results. The sex f the nestlings used in the preliminary maintenance re'gime and in the mating arena was recrded in all the series f experiments in which New Zealand Whites were used as hsts. Ten f these nestlings were used in maintenance and thirty-nine in the arenas. As we were unable reliably t sex new-brn rabbits by external characters they were either killed and their uringenital systems dissected, r they were marked and re-examined when - weeks ld. The bxes used as mating arenas were glass-tpped, cardbard museum (display) bxes sized 6 x x 2 in. Althugh bxes with well-fitting lids were always selected ne r tw fleas ccasinally escaped. During an experiment the bxes were maintained in the dark in an incubatr at 2-25 C, 7% R.H. A yung nestling rabbit culd be kept in such cnditins fr up t 2 hr. withut becming chilled and if returned t its nest was always accepted by the de. It was necessary t prevent certain batches f fleas frm feeding n the nestling rabbit which was frequently present in the mating arena. The methd used was t amputate part f the piercing muthparts. T accmplish this the flea was lightly anaesthetized with CO 2 and rientated n a slide n the stage f a dissecting micrscpe s that the head was t the right and the muthparts pinted away frm the peratr. The flea culd be held in this psitin by light pressure frm a fine, bent, munted needle laid acrss the prntal cmb and behind cxa. The laciniae and epipharynx were then gently drawn frwards, clear f the ther appendages, using the back f a Brradaile needle, and when clear were pressed firmly against the slide and the terminal 5 ft, r s, cleanly remved with the blade. Particular care was taken t ensure that bth labial palps, which are very thin and transparent, were free frm the laciniae. Sme direct bservatins were made f the behaviur f small numbers f male and female fleas put tgether in an arena with a nestling. At first all the fleas were virgin, but mature, and taken frm different pregnant rabbits at the time f parturitin. Fr these bservatins the glass-tpped museum bxes used as arenas were kept n a

Sme requirements fr mating in the rabbit flea 97 warmed surface n the labratry bench. In sme cases the fleas were individually marked with spts f differently clured cellulse paints. It was apparent frm the results f the mating experiments that the sexual cnditin f the rabbit n which the male fleas were maintained prir t being put with the female fleas partly determined whether cpulatin and impregnatin ccurred. As a cnsequence, the reprductive systems f male fleas which had been kept fr different lengths f time n male, nn-pregnant and pregnant female rabbits were examined fr any grss differences in appearance. The reprductive system f the male flea is relatively simple and easily dissected frm a freshly killed specimen (Mead- Briggs, 962). Each f the paired pyrifrm testes cntains, within a thin cntractile uter cnnective tissue sheath, many bundles f spermatza, and als prximally (psteriad) a cnvluted tube f much greater diameter than the vas deferens. This tube has been referred t as the epididymis (Minchin, 95) and may functin as a vesicula seminalis. There are tw pairs f tubular accessry glands, thse f each side pening by separate pres clse t where the vas deferens and ductus ejaculatrius unite. At first there are paired ducti ejaculatrii bund tgether in a cmmn muscular sheath but where they enter the penis they unite int a cmmn duct terminating at the gnpre. The phallsme, r intrmittent rgan, is a cmplex structure and as it was f n cncern in the present bservatins it was dissected away frm the penis prper befre a cver slip was put ver the preparatin. Previus experience with the male system had indicated that there was little variatin in the verall size f the testes f different virgin fleas. Sme f the mst imprtant criteria f variatin between different males cncern the lcatin f spermatza in the carefully remved system, and the pssibility f frcing spermatza thrugh the ducts t the exterir. The prcedure fr applying pressure was t remve saline very gradually frm beneath the cverslip with bltting-paper, if necessary until the sheaths f the testes finally ruptured. The midgut epithelium was als examined since a crrelatin between its height and varian develpment has been reprted in the case f the female flea (Mead- Briggs, 966). Measurements f the heights f the epithelial cells were necessarily very apprximate, and were btained frm fresh, unfixed preparatins in which the epithelium had been everted by cntractin f the circular muscle fibres after the muscular gut wall has been slit with a fine munted needle. RESULTS, OBSERVATIONS AND INFERENCES () The influence n mating success f different maintenance regimes fr the male flea The scheme f experiments and summary f results are indicated in Table ; the results are given in a mre detailed frm in Table 2. There were eighteen basic experiments each repeated tw r mre times t give a ttal f sixty-seven individual experiments. Each experiment was started by placing ten gravid female and eight male fleas in a mating arena and terminated 6 2 hr. later when the spermatheca f every female recvered was examined fr the presence f spermatza. Seventeen f the 67 females either escaped during the experiments, r their spermathecae were lst during dissectin, leaving 65 definite recrds. -2

98 A. R. MEAD-BRIGGS AND J. A. VAUGHAN The male fleas culd be gruped int six categries accrding t their preliminary maintenance regimes, viz. kept fr 26- days n male, nn-pregnant and pregnant female rabbits and similar fleas subsequently kept 2 hr. alne with a new-brn rabbit. The fleas maintained n pregnant rabbits were always cllected within a few hurs f the parturitin f their hst and befre they had mved int the nest. Table. Summary f the prprtins f gravid female fleas impregnated during cnfinement in an arena vrith male fleas that had been given varius preliminary maintenance rigimes and in the presence r absence f a new-brn rabbit (In ne grup f experiments the muthparts f the malefleaswere damaged t prevent their feeding n the nestling in the arena.) Maintenance regime fr (J fleas prir t experiment On pregnant rabbit up t parturitin Ditt and then 2 hr. n nestling On male rabbit Ditt and then 2 hr. n nestling On nn-pregnant female rabbit Ditt and then 2 hr. n nestling Prprtin f gravid? fleas impregnated during experiment N nestling L trcofl /8* /8 /9 /2 /8 /2 (J fleas intact S9/6 6/5 /6 25/27 /6 27/28 Nestling in arena /8 indicates nne ut f eighteen females was impregnated. <? fleas with damaged muthparts 29 / / 5/5 In mst experiments a new-brn nestling was placed in the arena immediately befre the fleas were released, but in ne grup f experiments the nestlings were absent. The results shw that, whatever type f male flea was used, there was n successful mating when there was n nestling even thugh several f the experiments were run fr 2 hr. By cntrast, when a nestling was present mating and sperm transfer ccurred in many cases. These experiments fell int tw grups, thse in which the malefleaswere intact and thus able t prbe and feed at will n the nestling in the arena and parallel nes in which the muthparts had been damaged, immediately prir t release in the arena, t prevent their piercing the skin f the nestling. The results in Table shw that a very high prprtin f female fleas (59/6) was impregnated when intact male fleas frm pregnant hsts were used. The partial amputatin f the laciniae and epipharynx f similar malefleas reduced the prprtin f females impregnated (29) but clearly did nt s upset the males as t prevent all mating. A higher impregnatin rate (/) was recrded when the males had had a preliminary 2 hr. n new-brn rabbits befre their muthparts were damaged. The results with male fleas prepared n male rabbits and n nn-pregnant female rabbits were remarkably similar t ne anther, and yet differed in several respects /9 8/7

Sme requirements fr mating in the rabbit flea 99 frm thse just described. Cnsider first the cases in which males with damaged muthparts were used. There was n successful mating (/ and /9) if the males had nt had cntact with a nestling whilst still intact, whereas there was sme mating (5/5 and 8/7) when they had. There was sme cpulatin by each categry f undamaged male but the prprtin f females impregnated was very much higher (25/27 and 27/28 cmpared with /6 and /6) if the males had had preliminary access t a nestling. Sme additinal facts are illustrated by reference t the mre detailed results in Table 2. On tw ccasins the same grups f males were used fr tw experiments (a, a2 and bi, b2). There was n impregnatin f the females within 8 hr. in the absence f a nestling but when the males were placed with similar females in the presence f a nestling successful mating (6/8 and 8/8) ccurred during 6 hr. It culd be that the female flea refuses t mate until she had fed n, r been in cntact with, a new-brn rabbit. In three experiments (ci, C2 and C) the muthparts f gravid females were damaged t prevent their feeding n the nestlings during the experimental expsure and yet 29/ were impregnated. In tw ther experiments (di, d2) the intact, gravid females were left with a nestling fr 2 hr. befre being put, in the absence f a nestling, with males that had been prepared n pregnant rabbits and nestlings. Nne f the females was impregnated. It thus appears that feeding n a nestling by the already gravid femalefleasis nt the essential preliminary t cpulatin. Hwever, immediate cntact with a nestling may be vital fr the female and/r the male in the chain f events culminating in successful mating. In ne experiment (e) the skin f the nestling was washed in water, t remve water-sluble cmpnents left frm the amnitic fluid, befre the nestling was used in the mating arena. There was n indicatin that this treatment affected the mating f the fleas. Hwever, there was n mating in a parallel experiment in which a similar, but freshly killed nestling was used. Althugh the prprtin f females impregnated in the different experiments appears very variable a pattern can be distinguished. As a hypthesis let us assume that befre a malefleawill mate it requires t be ' primed' by prbing, and presumably feeding n, a pregnant rabbit r a new-brn nestling, and that male r nn-pregnant rabbits d nt prvide the necessary 'priming*. In thse experiments with intact males and a nestling in the arena there was always an pprtunity fr the males t prbe a nestling. With each categry f male there was sme successful mating but the prprtin f females impregnated was clearly much higher when the fleas had been prepared n a pregnant hst and/r n a nestling than merely n a male r nn-pregnant rabbit. The hypthesis is mre clearly supprted by the experiments in which the males had their muthparts damaged befre release in the arena. Althugh this methd f preventing the fleas frm piercing the skin f the nestling was far frm ideal, e.g. it did unknwn damage t the sensry receptr system and led t a lss f haemlymph which reduced the survival f mst individuals t under 8 hr., it did nt in itself entirely preclude mating. N female was impregnated by males straight frm the male r nn-pregnant hsts, but sme were when the males had been ' primed' n a nestling fr 2 hr. A higher rate f impregnatin was fund with males' primed' n the pregnant rabbits and the highest f all when a nestling was als included in the re'gime.

Sme requimnents fr mating in the rabbit jlea

52 A. R. MEAD-BRIGGS AND J. A. VAUGHAN Three experiments were carried ut t determine whether the male fleas had t remain n the pregnant hst until the litter was brn, r whether they were ' primed' sme days pre-partum. Grups f eight malefleaswere taken frm a pregnant rabbit 5, ^- and z\ days pre-partum, their muthparts were damaged, and then the males were placed with ten gravid females and a nestling in an arena. Impregnatin (5/) ccurred nly in the case in which the males had remained until 2\ days pre-partum. It was als fund that lng maintenance regimes n the pregnant hst were unnecessary. Male fleas were released n a pregnant rabbit nly 2 days pre-partum, recvered at parturitin and then put fr 2 hr. with a nestling. Eight intact males impregnated 6/8 females and eight damaged males impregnated 5/9 females. Hwever, access fr day t a new-brn nestling was apparently inadequate t 'prime' previusly unfed male fleas. Thus n gravid females were impregnated during tw 8 hr. experiments in which such males were used. Table. The prprtins f immature female fleas impregnated during experimental expsures with male fleas which had had varius preliminary maintenance rigimes Maintenance regime fr 6 fleas prir t experiment On pregnant rabbit up t parturitin, days On male rabbit, 29 days On nn-pregnant female rabbit, 29 days fleas frm nn-pregnant rabbit 9 fleas frm male rabbit *- *, <- \ N nestling in Nestling in N nestling in Nestling in arena (2 hr.) arena (7 hr.) arena (2 hr.) arena (7 hr.) /8 / / Attempts t ' prime' intact male fleas by keeping them fr 2 hr. clse t, but nt in prbing cntact with, a new-brn rabbit failed. The fleas, which had previusly spent 26-28 days n a male rabbit, were placed in an arena n ne side f a duble-layered gauze partitin thrugh which they culd nt reach a nestling n the ther side. At the end f the 2 hr. the nestling was changed and gravid female fleas were put with the males. Nne ut f females was impregnated in three experiments. The muthparts f tw grups f males were then damaged and these fleas were put with females int the nestling cmpartment but again there was n successful mating. It was cncluded that male fleas culd nt be 'primed' by such dur cntact with a nestling. In all the experiments s far reprted gravid female fleas were used. This wuld apprximate t the situatin in nature, since mst f the female fleas taken t the nest by the parturient de will have mature varies. It was, hwever, f interest t determine whether well-fed female fleas with undevelped varies mated under ur experimental cnditins. Tw grups f experiments were carried ut, ne with female fleas that had been maintained n a nn-pregnant female rabbit fr days and the ther with females frm a male rabbit after a similar perid. The experimental scheme and the results are summarized in Table. As anticipated there was n mating in the absence f a nestling but nly ne flea was impregnated, ut f a pssible sixty-nine, when nestlings were in the arenas. This flea received many (ver ) spermatza i/9 /8 /9 /8 / / / / /5 / /i

Sme requirements fr mating in the rabbit flea 5 and had been with males that had been ' primed' n a pregnant hst. It may be cncluded that fleas with undevelped varies are rarely inseminated. The results fr mating success f the fleas were examined accrding t the sex f nestling used but n likely effect f sex was indicated, althugh the data were sparse. (2) The number f spermatza transferred An estimate was made f the number f spermatza in each spermatheca and the results are summarized in Table. These results shw a general similarity in pattern t thse fund fr verall mating success. The cncept that the male requires t be 'primed' by prbing cntact with a pregnant rabbit r a nestling is supprted, and this 'priming' appears t be quantitative. Table. The estimated number f spermatza in the spermathecae f gravid female fleas after they had been cnfined tith males that had been given varius preliminary maintenance rigimes (In ne grup the males were damaged t prevent their feeding n the new-brn rabbit als present in the arena). Maintenance regime fr <J fleas prir t experiment On pregnant rabbit up t parturitin Ditt and then 2 hr. n nestling On male rabbit Ditt and then 2 hr. n nestling On nn-pregnant female rabbit Ditt and then 2 hr. n nestling Ttal On pregnant rabbit up t parturitin Ditt and then 2 hr. n nestling On male rabbit Ditt and then 2 hr. n nestling On nn-pregnant female rabbit Ditt and then 2 hr. n nestling Ttal, S.» 26 2 27 6 9 5 9 29 82 N. f spermathecae cntaining -2 S. 2 5 2 - S. - Spermatza 2 8 2 2 «S. -S. $ fleas intact 5 8 9 8 > S. 5 9 9 28 8 9 $ fleas with muthparts damaged 9 6 6 H 7 7 6 8 8 ^ Ttal 6 5 6 27 6 28 285 9 5 9 7 255 %S9 impregnated In the experiments with intact males the prprtin f females receiving large quantities f sperm was highest when the males had cme frm a pregnant hst and Jwest when they were frm male r nn-pregnant hsts and their nly cntact with 98 92 57 9 55 97 79 5 75 22 29

5 A. R. MEAD-BRIGGS AND J. A. VAUGHAN a nestling was in the mating arena. Preliminary expsure t a nestling fr 2 hr. f fleas frm male and nn-pregnant hsts clearly increased the verall quantity f sperm transferred. The practice f damaging the muthparts f sme males t prevent their successfully prbing the nestling during the experiment reduced nt nly the percentage f females impregnated but als the relative number f spermatza transferred; nevertheless, the pattern was similar. The nly females t receive large numbers f spermatza were thse put with males given preliminary access t a pregnant rabbit and a nestling. Table 5. T shw that the number f spermatza transferred t the spermathecae f gravid female fleas by male fleas previusly maintained n male r nn-pregnant female rabbits increased with time fr up t at least 2 hr. Duratin N. f spermathecae cntaining Surce and cnditin f fleas sure (hr.) f exp- S.* -2 S. - S. -S. > S. Male rabbit, Muthparts intact 6 days 5 2 5 Muthparts damaged 2 Nn-pregnant Muthparts intact 6 female rabbit, days 5 2 2 s Muthparts damaged 2 S. = Spermatza The results f all experiments are incrprated in Table irrespective f the length f time the sexes were tgether; this varied frm 6 t 2 hr. (Table 2). Cnsequently all the data are nt strictly cmparable but the nly imprtant differences cncern the categries f intact male fleas frm male and nn-pregnant female rabbits and whse nly access t a nestling was in the arena. In ne series f experiments grups f similar fleas were left tgether fr 6, 5 and 2 hr. The numbers f spermatza fund in the spermathecae were greater the lnger the experiment (Table 5). These results supprt the cncept f a quantitative ' priming' f malefleaswhilst they are in prbing cntact with a new-brn nestling. The influence f the duratin f the experiment was relatively unimprtant in the case f fleas 'primed' n a pregnant rabbit since nearly all the females received a large number f spermatza, appraching the maximum spermathecal capacity, within 6 hr. It was als unimprtant with damaged males since the peratin impaired their survival and few were fully active beynd 6-8 hr. When the experiments were planned it was cnsidered impracticable t use nly ne male flea per bx. Almst certainly very many mre replicate experiments wuld have been required and sufficient new-brn rabbits were nt readily available. A sixreplicate experiment was carried ut t btain sme idea f what a single, fully ' primed' male culd achieve under cnditins similar t thse already described. One male flea maintained n a pregnant de fr 28-29 days up t parturitin, ten gravid

Sme requirements fr mating in the rabbit flea 5 5 females and a new-brn nestling were put tgether in an arena fr 7 hr. The results were as fllws: Sex f nestling in arena <J s $? 9 $ S.' 2 IO 8 Number f spermathecae cntaining t-is. S. = Spermatza. - S. -S. >is. 2 Table 6. T shw that changes ccur in the mbility f spermatza within the genital ducts f male fleas when maintained n pregnant rabbits but nt when kept n male r nn-pregnant female rabbits Prprtin f preparatins with spermatza ut f testes and at varius psitins in system N pressure applied Slight pressure applied t preparatin Maintenance regime fr 6* fleas Unfed (straight frm culture) On male rabbit fr 9 days days 2 days 28 days On nn-pregnant female rabbit fr On pregnant rabbit (Day f hst's parturitin) 9 days days 2 days 28 days fr days 9 days days 7 days 9 days 2 days 2 days 25 days 27 days 29 days In epid. i/ 2/ 2/ In vasa def. # i/ 2/ */ In epid. J / 2/ i/ In Ex vasa def. gnpre / indicates spermatza were present, r frced int, bth vasa deferentia f ne individual ut f three, i/ means int nly ne vas deferens f ne individual and \I int ne vas deferens in each f tw ut f three individuals, etc. i/ / / i/ i/ 2/ 2i/ / 2/ () The influence n the male reprductive system f different maintenance rigimes A typical series f bservatins is summarized in Table 6. Unfed male fleas were released n male, nn-pregnant and pregnant rabbits and samples f three fleas were remved at varius intervals frm each hst and dissected. N difference was bserved between the fleas frm the male rabbit and the nn-pregnant female. Befre pressure

56 A. R. MEAD-BRIGGS AND J. A. VAUGHAN was applied the spermatza were entirely restricted t the testes prper, but with pressure sme spermatza entered the epididymes but nt the vasa deferentia. These bservatins were in marked cntrast t thse n males frm the pregnant rabbit. In mst f these spermatza were already present in the epididymes and in several fleas taken during the latter half f the gestatin spermatza were present in ne r bth vasa deferentia. Gentle pressure frequently caused spermatza t pass int the vasa deferentia but at first they rarely penetrated beynd the level f the junctin with the accessry glands and the ejaculatry ducts. With fleas taken when the hst was nly a few days pre-partum, hwever, spermatza culd usually be made t pass right thrugh the system and ut f the gnpre int the surrunding saline. Thus spermatza in fleas kept n a pregnant rabbit exhibit much greater mbility within the genital ducts than d spermatza in fleas kept n male and nn-pregnant female hsts. As already mentined there is little change in the verall size f the testes that can be crrelated with the type ffleamaintenance. This bservatin is prbably cnsistent with the fact that spermatgenesis is largely cmpleted by the time the adult flea emerges frm its ccn. The ranges f size fund fr the testis prper, i.e. the part f the capsule cntaining bundles f spermatza were as fllws: 5 specimens frm pregnant hst 285- x -6/ 2 specimens frm nn-pregnant hst 28- x 5-55/ 2 specimens frm male hst 29-2 x 5-55/i There were spaces between the bundles f spermatza in the testes f many unfed fleas and f fleas which had been n a hst fr less than 2- days. These spaces appeared t be filled by afluidcntaining small granules. The heads f the spermatza f each bundle were typically embedded in a granular, gelatinus cap, but when the sheath f a testis was ruptured at least sme freely mtile spermatza were t be seen, irrespective f the feeding regime f the flea. The size f the accessry glands increased prgressively in fleas kept n a pregnant hst, but did nt d s in fleas n nn-pregnant and male hsts. Data included in Table 7 shw that the length and width f the lateral accessry glands increased by abut 5%. These glands, which cnsist f a simple clumnar epithelium, appeared t be in an active secretry phase in fleas taken frm a hst shrtly pre-partum. Table 7 als includes data n the heights f the epithelial cells lining the midgut. The results shw that these epithelial cells were significantly larger infleastaken frm a pregnant hst during the final week f gestatin than frm nn-pregnant nes. A similar, but mre marked, develpment f these cells was fund in the case f female fleas kept n pregnant hsts (Mead-Briggs, 96 b; Rthschild, 965 a). () Mating behaviur In rabbit nests rabbit fleas may be fund in cpula bth n the skin f the nestlings and amngst the nest material. Similarly, during the bservatins n the fleas in the cardbard bxes it was fund that cpulatin culd be initiated between fleas bth n and ff the bdy f the new-brn nestling enclsed with them. Hwever, n mating attempts were seen when there was n nestling in the bx. Twelve sets f fleas were watched, and details are given in Table 8 f the numbers

Sme requirements fr mating in the rabbit flea 57 Table 7. The mean length {height) fmidgut epithelial cells {m.ep.c), and the mean length and width f lateral accessry glands {a.gl.) f male rabbit fleas dissected after maintenance fr varius perids n male, and nn-pregnant and pregnant female rabbits (Each mean is based n the dissectin f a sample f three fleas.) () Fleas frm $ rabbit N. f days n rabbit 9 2 28 5 '5 6 6 6i 6 6 6 2 2 9 2 28 6 8 6 7 6 59 6 6 2 6 2 2 6 2 2 2 7 2 Mean length m.ep.c. (/*) Mean length a.gl. (/*) Mean width a.gl. (ji) (2) Fleas frm nn- N. days n rabbit Mean length m.ep.c. (ji) pregnant $ rabbit Mean length a.gl (/*) Mean width a.gl. (JI) ( a) Fleas frm N. days n rabbit pregnant? rabbit N. days pre-partum (a) Mean length m.ep.c. *) Mean length a.gl. (ji) Mean width a.gl. (/*) (6) Fleas frm N. days n rabbit pregnant $ rabbit N. days pre-partum (b) Mean length m.ep.c. (JI) Mean length a.gl. (ji) Mean width a.gl. (ji) 7 9 8 25 2 69 7 2 9 6 9 2 7 8 6 25 S 2 5 8 8 5 Table 8. Details f the numbers f fleas which cpulated, r attempted t d s, when male and female fleas, which had been maintained n pregnant rabbits up t parturitin, were put tgether with a new-brn rabbit in a glass-tpped bx n a warmed bench 7 2 8 8 7 7 2 9 7 79 7 N. f fleas used N. f N. f successful cpulatins Time t, * > attempted, *, st cpulatin 6* 9 cpulatins In bx On nestling Ttal (min.) 5 a 5 8 2 2 9 8 2 5 8 7 2 2 8 2 6 7 5 9 5 5 7 O I 5 5 7 2 f fleas which were used, and the numbers which cpulated r attempted t d s. The number f matings r attempted matings bserved was lw, presumably because the cnditins btained n the bench were nt s satisfactry as in the incubatr used fr the ther experiments. Perhaps darkness, the usual situatin in nature, is imprtant. Althugh it was thught that all thefleasf the same sex wuld be in a similar physilgical state, as they were similarly prepared, there was a wide range f reactin t individuals f the ppsite sex. Thus, in sme cases males and females encuntered each ther as they mved abut, r they came t rest, r t feed, side by side and yet shwed n signs f mating activity. Smetimes a male apprached a female, ran arund her and attempted t cpulate but the female wuld nt accept. As he at- 98 2 2 8 7 86 2 2 6 8 9 7 25 22 8 9 27 22 9 2 29 26 99 2

58 A. R. MEAD-BRIGGS AND J. A. VAUGHAN tempted t slip beneath her abdmen with his abdmen curved upwards and the terminalia raised she might either mve away r, in the case f sme feeding females, repel him by kicking with the hind leg. Using individually marked fleas additinal situatins invlving unsuccessful mating attempts were nted. An individual male might attempt t cpulate with several females, smetimes in quick successin, but be rejected each time. The male did nt necessarily attempt t mate with every female encuntered. Smetimes ne r mre males made repeated attempts at cpulatin with a particular female but were rejected each time. Sme females might mate with ne male and later reject anther, r vice versa. When a male was accepted and cpulatin successfully achieved the female might mve abut, dragging the male with her, r remain attached and feeding n the nestling. The male was nt seen t feed whilst m cpula. The duratin f cpulatin varied widely frm sec. t 9 r min. The actual mating 'behaviur' appeared t have a very simple pattern, with n preliminary 'curtship'. The male made a definite apprach t certain females, and was either allwed t cpulate r rejected. The 'characteristic zig-zag apprach' f the male referred t by Rthschild (965a) seemed t be a purely mechanical effect as the flea scrambled thrugh lse fur r scuttled ver the bdy f the nestling, since bth sexes mved in an irregular path in this way. DISCUSSION The results described in the preceding sectins indicate that several cnditins must be satisfied befre successful cpulatin can ccur between male and female rabbit fleas. Thus, the male fleas, and prbably the females, need t be in an apprpriate ' physilgical state', and there must be a live, yung nestling in the vicinity f the fleas. Fleas which will mate sn after reaching an ccupied nest never cpulate whilst still n the adult hst, even if it is pregnant r immediately pst-partum. Althugh fleas were frequently fund in cpula ff the bdy f the nestling in the bxes (and als in natural rabbit nests) and were smetimes seen t cmmence mating whilst away frm the nestling they were never fund t have mated successfully, using impregnatin as the criterin, in the cmplete absence f a nestling. Grups f male fleas which failed t impregnate any females in an therwise empty bx mated effectively with similar females as sn as a nestling was included. The present direct bservatins and the literature suggest that in many species f fleas the female determines whether cpulatin will be permitted. If she is nt ready t accept cpulatin curting males may be repulsed by vigrus kicking with the hind legs. In several species the males are believed t be psitively attracted t females when the latter are ready fr cpulatin, and it has been suggested that the females may prduce a phermne which is detected at a distance by receptrs n the maxillary palps f the male (Geigy & Suter, i96). Humphries (967) fund that male Ceratpkylhis gallinae (Schrank) d nt lcate females frm a distance, but that mating behaviur is initiated after the male has accidentally cllided with a female and a cntact-chemical stimulus has been received by his maxillary palps. It wuld be interesting if this difference between shrt-distance and cntact attractiveness f the female culd be crrelated with the need r therwise fr bld meals by the female prir t mating. There is evidence that the females f several

Sme requirements fr mating in the rabbit flea 59 species ffleasd require t feed befre they mate. Geigy (95) bserved that cpulatin ccurs in Txmga penetrans (Linn.) nly after the female has bred int the skin and varius rgans have undergne hypertrphy. Female Megabthris c. clantm (Hubbard) need access t a suitable hst fr at least ne day befre they will mate (Ple & Underhill, 95). Geigy & Suter (i96) shwed that befre cpulatin female Echidnphaga galunacea (Westwd) must feed fr 2- days, during which time their varies reach a certain, but undefined, stage f maturity. Suter (96) implied that the females f all species f fleas must underg a perid f maturatin befre they are ready fr cpulatin and becme attractive t males. Hwever, the females f many ceratphyllids d nt require a bld meal prir t cpulatin althugh feeding will dubtless be necessary fr egg prductin. Examples include the bird fleas Ceratphyllus farreni Rths. (Waterstn, 9), C. niger Fx, C. idius Jrdan & Rths, and C. riparius Rths. (Hlland, 955) and C. galhnae (Schrank) (Rthschild, 965a; Humphries, 967) and the grey squirrel flea Orchpeas h. hwardi (Baker) (R. J. C. Page, unpublished). Unfed female rabbit fleas d nt mate. Even fleas which have been allwed t feed fr days n male r nn-pregnant rabbits, but f curse still have immature varies, mate nly rarely when put in an arena with fecund males and a nestling. By cntrast, in a similar situatin nearly every female with mature varies wuld be impregnated. A systematic study t determine whether the prpensity t mate can be crrelated with the precise degree f varian develpment has nt been undertaken. (Adams & Mulla (968) have demnstrated that mating in the eye gnat Hippelates cllusr ccurs when varian develpment is in genetic stages 6 9, with a peak in stage 7. This can be crrelated with prductin f a phermne sex attractant frm stage 6 t stage 9, after which the females becme repellent during stage, the stage f mature eggs.) Indeed such a study in the rabbit flea wuld be difficult because varian develpment is very rapid, passing frm immaturity t near maturity in 2 hr., when the female flea can feed n a nestling rabbit, and it is essential t prvide a nestling if mating is t be pssible. Presumablyfleaswuld have t be taken frm pregnant rabbits when their varies were at different stages f develpment and then prevented frm feeding further, n the nestlings, by damage t their muthparts r sme similar technique. Mating ccurred in all three f the present experiments in which the muthparts f gravid females taken directly frm an adult hst were deliberately damaged t prevent their feeding n the nestling in the arena. This appears t cnflict with the belief f Rthschild (965 a, b) that the female flea, even if mature, needs t feed n a yung rabbit befre she will accept cpulatin. Under natural circumstances feeding wuld ccur n the nestling in any case t prvide nurishment fr cntinued cyte maturatin. Rthschild (965 a) and Rthschild & Frd (966) have shwn experimentally that different hrmnes f the hst are invlved in varian develpment and the acceptance f cpulatin in the fleas. Apparently the ' cpulatry factr' can prduce its releaser effect withut being imbibed. Other wrkers have fund that the males f mst species f fleas d nt require t feed befre they can cpulate successfully. This is the case fr all thse ceratphyllid species in which the females d nt need preliminary meals, and fr T. penetrans, E. galunacea and Xenpsylla chepis (Rths.) which d (Geigy & Suter, i96). Hwever, males f M. c. clantm fed n rat bld are insufficiently aggressive t cpulate

5 A. R. MEAD-BRIGGS AND J. A. VAUGHAN satisfactrily, by cntrast with nes fed n mice (r vles) (Ple & Underhill, 95)* Our experiments indicated that the male rabbit flea must be in the crrect ' physilgical state' befre it can mate successfully. This state is usually reached when the flea has been in prbing cntact with a pregnant rabbit nt mre than 2- days prepartum, r a new-brn rabbit. The cntact with a nestling needs t be fr nly a few hurs, particularly if the fleas have already fed fr sme days n an adult rabbit. The nestling cntact has t be ne that allws prbing, and presumably feeding. Males which have been nly in dur cntact r did nt prbe because f damage t their muthparts fail t mate. It is pssible that the accessry glands f the reprductive system f the male rabbit flea have t be brught int gd secretry activity befre the spermatza can pass readily thrugh the genital ducts. It is generally cnsidered that reprductive prcesses in the male insect are little affected by nutritin althugh mating behaviur may be elicited by many kinds f stimuli, including varius types f fd. It has nt been pssible t deduce whether the 'factr' needed by the male rabbit flea t bring it int the crrect 'physilgical state' fr mating, and btained frm the pregnant hst and/r the new-brn rabbit, is a nutritinal requirement r a sensry trigger. The same difficulty was fund when cnsidering the ' factr' needed by the female flea fr initiatin and maintenance f vitellgenesis (Mead-Briggs, 96a). These tw factrs may well be ne and the same, since their distributin appears t be identical, i.e. present during the final stages f a rabbit's pregnancy, and at high level in the new-brn, but waning t zer during the first 7 days, r s, f life. SUMMARY. Rabbit fleas Spilpsyllus cuniculi (Dale) d nt mate while n adult rabbits; cpulatin is usually initiated when thefleasare n the bdy f a new-brn rabbit and at least the clse prximity f such a nestling is essential. The female prbably becmes receptive nly after clse cntact, but nt necessarily prbing cntact, with a yung nestling. 2. Fleas with mature varies mate much mre readily than immature fleas.. Male fleas must be in a suitable 'physilgical state' befre they can mate and inseminate females. This state is reached by prbing cntact with rabbits in the final stages f pregnancy r, fr a relatively shrter perid, with new-brn rabbits. Spermatza f such fleas shw much greater mbility within the genital ducts than thse frm fleas kept n male r nn-pregnant rabbits. It is a pleasure t recrd the help given by Mr R. J. C. Page in the earlier stages f this wrk. Many f ur clleagues thrughut England and Wales assisted by sending us flea-infested wild rabbit nests. REFERENCES ADAMS, T. S. & MUIXA, M. S. (968). Ovarian develpment, phermne prductin, and mating in the eye gnat, Hipptlates clhur. J. Insect Pkytil., 627-5. GEIQY, R. (95). Sandflh-Prbleme. Naturwutenschaften, -2. GEIGY, R. & SUTER, P. (i96). Zur cpulatin der Flflhe. Revue suisse Zl. 6rj, 26-. HOLLAND, G. P. (955). Primary and secndary sexual characteristics f sme Ceratphyllinae, with ntes n the mechanism f cpulatin (Siphnaptera). Trans. R. ait. Sc. Lnd. 7, 2-8.

Sme requirements fr mating in the rabbit flea 5 HUMPHRIES, D. A. (967). The mating behaviur f the hen flea CeratpkyUus gautnae (Schrank) (Siphnaptera: Insecta). Arntn. Behav. 5, 82-9. MEAD-BRIGGS, A. R. (962). The structure f the reprductive rgans f the Eurpean rabbit flea, Spilptyllus cumculi (Dale) (Siphnaptera). Prc. R. ent. Sc. Lnd. (A), 7, 79-88. MEAD-BRIGGS, A. R. (96a). The reprductive bilgy f the rabbit flea SpUpsyllus cumculi (Dale) and the dependence f this species upn the breeding f its hst. J. exp. Bil., 7-2. MEAD-BRIGGS, A. R. (966). A crrelatin between develpment f the varies and f the midgut epithelium in the rabbit flea Sptlpsyllut cumculi. Nature, Lnd. ai, -. MEAD-BRIGGS, A. R. & RUDGE, A. J. B. (i96). Breeding f the rabbit flea, Spilptyllus cunuuli (Dale): requirement f a 'factr' frm a pregnant rabbit fr varian maturatin. Nature, Lnd. 87, 6-7. MINCHIN, E. A. (95). Sme details in the anatmy f the rat flea, Ceratphyllus fatciatut Bsc. J. Quekett micrsc. Club (),, -6. POOLE, V. V. & UNDERHILL, R. A. (95). Bilgy and life histry f Megabthris clantm clantni (Siphnaptera: Dlichpsyllidae). Walla Walla Cll. Publt Dtp. bil. Set. (9), -9. ROTHSCHILD, M. (965). The rabbit flea and hrmnes. Endeavur, 62-8. ROTHSCHILD, M. (9656). Fleas. Scient. Am., -5. ROTHSCHILD, M. & FORD, B. (96a). Breeding f me rabbit flea (Spilptyllus cumculi (Dale)) cntrlled by the reprductive hrmnes f the hst. Nature, Lnd. ai,. ROTHSCHILD, M. & FORD, B. (966). Maturatin and egg-laying f the rabbit flea (Spilptyllus cumculi Dale) induced by the external applicatin f hydrcrtisne. Nature, Lnd. 2, 2-. ROTHSCHILD, M. & FORD, B. (966). Hrmnes f the vertebrate hst cntrlling varian regressin and cpulatin f the rabbit flea. Nature, Lnd. an, 26 6. SUTER, P. R. (96). Bilgie vn Echidnphaga gallinacea (Westw.) und Vergleich mit andern Verhaltenstypen bei Fldhen. Ada trap, ai, 9-28. WATERSTON, J. (9). Sme habits and hsts f bird CeratpkyUi taken in Sctland in 99; with descriptins f a new species (C. rthscmldt), and recrds f varius Siphnaptera. Prc. R. Pkys. Sc. Edinb. 8, 77-9. Exp. BL 5, 2