Ontogenesis of the mite Varroa jacobsoni Oud. in worker brood of the honeybee Apis mellifera L. under natural conditions

Transcription

1 E:~-perimental & Applied AcarohJgy. 18 (1994) Ontogenesis of the mite Varroa jacobsoni Oud. in worker brood of the honeybee Apis mellifera L. under natural conditions S. J. Martin National Bee Unit, MAH", CSL, l,,uddmgron. Stra~rd-upon-Awm, Warwickshire. CV37 9S J. UK. (Accepted 22 February 1994) ABSTRACT Martin, S.J., Ontogenesis of the mite Vcuroa jacohso~zi Oud. in,a'orker brood of the honeybee Apis mellijera L. under natural conditions. Exp. Appl. Acarol. 18: A study was carried out during May in southern England, on eight chemically untreated Apis mellifera L. colonies heavily inl~sted with Varroa jacobsoni ( 15-40% of worke, sealed brood). The position and time of capping of 3,228 worker sealed br~n)d were recorded. At two hour intervals, starting from when each cell was capped, groups of worker cells were uncapped and their contents recorded. It was found that each ~: jacobsoni female could deposit five or sometimes six larvae in a worker cell, of which four (1 male and 3 females) may reach maturity before the bee emerged from its cell. However. mortality of the oft~pring resulted in only 1.45 female offspring reaching maturity. for each normally reproducing mother mite, before the bee emerged. The mean development time of the firsl three female offspring was 134 hours (_+ =3 h. n=3), shorter than that of lhe male ( 154 hours). The first larva was deposited approximately 60 hours alter the cell was capped, and developed into a male, Subsequent larvae were deposited at intervals varying from hours, and all developed into females. INTRODUC~ON Varlva.jacobsoni Oud. is a parasite mesostigmatid mite which lives exclusively on honeybees (Apis spp.). The mites reproduce only in the sealed brood of the honey bee and feed on the developing bee. Fertilised females survive broodless periods (e.g. during winter) on the adult bees. The development of the mite is restricted mainly to drone (male) brood in Apis cerana E, the mite's natural host. This is believed to be due to various behavioural and physiological traits (Rosenkranz et al., 1993). However, in A. meilifera L., the mite reproduces successfully in both worker and drone sealed brood. Worker brood is always more numerous than drone brood in a colony, and unlike drone brood, is present throughout the entire breeding season. These facts, coupled with the lack of any of the behavioural and Science and Technology Letters. All rights reserved.

2 88 S.J. MARTIN physiological traits present in A. cerana, allow the mites to build up to levels which eventually cause colony collapsc if left untreated (Ritter, 1981). In the 1960s, A. meliifera colonics were brought into close contact with infested A. cerana colonies in southern Chinain 1960 and east Russia in 1964 (reviewed by Smirnov, 1978). Subsequently the mite has spread rapidly throughout the rest of Asia, reaching Europe, North Africa and South America by the 1970s (Griffiths and Bowman, 1981). Despite various exclusion attempts, the mite has also been reported in the USA during 1987 and UK during During its development, V. jacobsoni, like most other mesostigmatid mites, passes through four stases: larva, protonymph, deutonymph, and adult (Evans, 1992). The egg develops internally with the viable ofi~pring being deposited as a hexapod larva. Since V. jacobsoni can only breed within sealed brood, the number of mature offspring that can develop is affected by the length of time the cell remains sealed. Despite considerable research into the biology and population dynamics of V.jacobsoni, the ontogenesis (development) of the mite offspring within the cell has only been studied in detail by Infantidis (1983), who calculated that up to two female mites could mature by the time the bee emerges. However, Rehm and Ritter (1989a) showed that the first mite offspring was male and not female as assumed by lfantidis (1983). This was probably due to the different rates of development of the male and female protonymph stases. Rehm and Rittcr (1989a) recalculated Infantidis' data, and showed that two (and in some cases, where the sealed brood stage is extended, three) new adult female mites could mature before the bee emerged. Three new adult females can be produced in a worker cell from one mother mite (12%, Schulz, 1984; 4%, Accorti and Nannelli, 1990; 23 %, n=106, personal observation), suggesting that the true development time may be shorter than that proposed by Rehm and Ritter (1989a). Limited observations by Rehm and Ritter (1989b) and Ifantidis's (1990) recalculations of his 1983 data, suggested a shorter developmental time for the female mite. Infantidis (1983) combined results from both the drone and worker sealed brood which may tend to misrepresent the development times in worker sealed brood alone, since the time that the mites remain in brood food after the cell is capped is different in the worker (6 hours) and drone (20 hours) sealed brood (Infantidis, 1988) and the speed of mite development is different in the drone and worker cells (Piletskaya, 1988). Also Ifantidis (1983) re-used his groups of cells, which may have caused periodic periods of brood cooling while the frames were out of the hive, which will have prolonged mite development (Piletskaya, 1988). Emphasis has been placed on breeding Varroa 'resistant' bees by selecting bees that have a shorter sealed brood period (Moritz, 1985; Le Conte and Cornuet, 1989; Btichler and Drescher, 1990; Mortiz and Jordan, 1992). This would reduce the number of mites able to complete development. The aim of this study is to repeat lnfantidis' original work in greater detail using the knowledge that the first mite offspring is male.

3 ONIOGENP~glS OF THE MI'fE VARROA JAUOBSONI O 'D. IN WORKER BROOD OF IIlE HONEYBEL- 89 MATERIALS AND METHODS The study was carried out during May 1993 using eight chemically untreated A. meilifera colonies in which 15-40% of the sealed worker brood were infested with V.jacobsoni. Over a period of three consecutive warm (max. temp. 14.1, 14.3, 14.5 C), sunny days (13.0, 13.6, 12.6 h.), suitable frames (containing many mature larvae) from the central region of each colony were removed and the position of all sealed brood recorded on transpm-ent sheets. Frames were then replaced in the hives. Thereafter, the marked fiames were examined at two hour intervals during the day and the position of brood which had been capped in that interval was recorded. The cells that were capped during each two hour period were classed as a group. A total of 157 groups were recorded, allowing the entire development of the mite and bee sealed brood to be monitored. The average duration of the sealed brood stage is 12 days (288 hours) in A. mellifera in the UK (Jay, 1962). The relevant combs were removed from the colonies at pre-determined times and the cell contents of each group, as indicated by the transparent sheets, were recorded. If this could not be carried out immediately, the frames were placed in a freezer to kill all brood (bees and mites). Each frame was used once. The development stage of the bee in each cell was recorded: mites, if present, were removed from the bee's body and/or cell wall. Mites were placed on a microscope slide using a fine brush and examined using a stereo zoorn microscope. The condition of the mother mite was recorded according to the state of body swelling (Table 1 ). Mite larvae and protonymphs were not sexed, as this cannot be done reliably using external characteristics (Steiner, 1988). Only eggs which had developed into hexapod larvae at the time of deposition were considered viable. Some eggs were deposited before they develop into the larval stases and were not viable (Akimov and Pileskaya, 1983). Protonymphs and deutonymphs were distinguished by counting the number of sternal setae (3 pairs = protonymph: 5-6 pairs = deutonymph) using a 100 microscope. The male and female dcutonymphs were easily distinguished using body shape and arrangement of sternal setae in the young deutonymph stages. The mites were not classified into immobile/mobile (deuto/proto chrysalis) stages. The presence of exuviae was used to confirm the number of adult offspring present in each cell. TABI,E 1 Cla,,,silication of mother mites. (.'lass A B C D Condition of body Shields of idosoma concave Shields of idosoma level vath edge of dorsal shield ldosoma shields convex, intersutal membrane visible betv,'een shields ldosoma shields highly convex, intersutal membrane visible between dor~,al and metapodal shields

4 90 S.J. MARTIN RESULTS The position and time of capping of 3,228 worker brood cells were recorded. Of these, only 89 cells (2.8%) had been uncapped and the contents removed by adult worker bees. There was no relationship between the level of comb infestation (15-20% or %) and number of sealed brood removed (4.0% and 3.3%) respectively. Of the remaining 3,139 cells, 908 (29%) were infested with a total of 1,334 mother mites. The reproductivity of these mites is given in "Pable 2. Only 2% of the mother mites died whilst in the cells and 18% of them failed to produce viable offspring (1 or more mated females). However, these figures do not take into account changes that occur over time and assume that a normal mite will remain normal and no mortality will occur, if the recorded mite was allowed to develop. This causes an overestimation of the norm',d group. Figure 1 shows the percentage change in the various groups given in Table 2 during the development of the bcc pupa. The data now show that 35% of the mites fail to produce viable oltspring by the time the bee emerges. The percentage of non-reproducing, laying of non-viable eggs (Fig. 1 ) and dead mites remains fairly constant throughout the development of the bee scaled brood. No mother mite was found to produce more than one male offspring. A similar observation was made by Akimov and Piletskaja (1985). In the first group of cells examined (sealed for 0-2 hours), five mites were found lying motionless in the brood food, facing the bee larva, and a further two mites were on larvae. While in the brood food the mites raised their peritrcmes, presumably to aid respiration. It was noticed that in the next group (sealed for 2-4 hours) all mites were found on bee larvae, or occasionally on the cell wall, despite the presence of small amounts of brood food up to 4-6 hours after scaling. This suggests that the mites may be able to escape from the brood food rather than wait TABLE 2 Repn~uctivity of mother mites. Reproduction of mites Time from Number of % of when measured mites mites (hours) Normal > Abnormal: with viable offspring > with no viable offspring > only single male off.,;pring > non-viable egg/s & no offspring > non-viable eg~s & male offspring >141) 17 1 non-viable egg/.,, & offspring > Non-reproducing > Mite dead: trapped in cell wall >30 18 I in cell >0 6 1

5 ONTOGt:~NESIS OF THE MITE ~'7~RROA Z,~,COBSONI Or-D, IN WORKER BROOD OF THE IIONEYBI:.E 91 80' \. uj 60",< i- "' 40 -t : l w "-..,,.. -" "'. ~,,,,"/" ' "~N,V IA B L E A "v" L, ///%//' "~NORMAL /NO VIABLE O..,-7'~'-.,-. ~ ~ : ",..... ",.-"."... [-'NQN-REPRODUCING TIM E(holrl) Fig. 1. Changes in the frequency of reproduction type {Table 2) and viability of offspring produced by mother limes in relation to the development of the worker bee sealed brood. Mites were grouped into consecutive 20 hour perk,s. to be released by the feeding activity of the bee larva, as suggested by Infantidis (1988). Infantidis (1988), using much larger samples, found.that 86% and 100% of mites were free from the brood food two hours and six hours, respectively, alter sealing. During the next 30 hours the idosoma of the mother mite swelled as the first egg/larva matured. This caused the interscutn membranes to stretch, forcing the anal shield (alter approximately 10 hours) and later the metapodal shields (after 30 hours) to protrude away from the dorsal shield. The mother mite remained swollen until the fifth larva was laid at around 172 hours (Fig. 2). Nine percent (n=159) of the mother mites present alter 212 hours had produced six offspring. The first larva was laid on the cell wall near the top (cap), thereby preventing the larva being damaged during the bee's pre-pupal moult. Subsequent larvae were usually deposited further down the cell or sometimes on the developing pupa. The time and interval between each larva being deposited is given in Table 3. The duration of the development of each mite offspring in worker sealed brood is given in Fig. 2. As the bee developed, the later mite larvae often fail to develop and subsequently die. To quantify this, only the mother mites which were tbllowing a normal course of reproduction (as indicated by Fig. 2) were used. During each developmental stasis (larva, protonymph, etc.) of each offspring the percentage of that offspring (lst, 2nd, etc.) that were dead or showed abnormal development was calculated. The results showed (Fig. 3) that there was no mortality in the larvae and protonymphs

6 -C.,-... ith ell 7 OZ J =~- ~.Q: 5th oil i w a~ ~ J G. u~ /tl I " oo [...1 ; w I/d 211 el - lfltll.j oj CONDITION OF : MOTHER MITE:I S -- -/--/- "i DEVELOPMENT OF~_cl os WORKER BEE -... L 30kt "'--:_.. 1=63 ~=159 L 22Jtl L 24~ --, P 34h~I~ ;~3-9,..112 O 7611f1 L 22brs ' P 32~ li =48 n=32 p L.st ~---=-'.- IS 11=2,.2 D P.2_8_~. '- ~- - T:,:... ~L 221iS ' n=4g m=37 - ~-~i- -~ D 861ol1 ~ : ;=8s D 721[s.---- s l.... p w _ po PP pr.... yt \ D 80~rs,=124 D P A i:28 1 A 9 M~105 i n= ! L / i I ' 0 20 TIME(IItll 20 " ~oo "-~2"0... 1io ' i~o~-~o~ o ' 2,;o TAME [Murl] - - 1'o ~" ---1 '~ Fig. 2, Timing and duration of development of Varroajacobsoni offspring in relation to the development of the worker bee sealed brood and condition (sec Table I) of the mother mite. Mite: I~ larva, P= protonymph, D= deutonymph, A= adult, (est)= duration estimated not measured; developmental time is given above the line with the sample size (n) given below, the dotted line indicates the range. Bee: el= cell capped, cs= cocoon spinning, sl= stretched larva, pw= pupa with white cyes, po= eyes pale, pp= eyes pink, pr= eyes purple, yt= thorax yellow, gp= wing pads grey, gt= thorax grey, m= pupal moult, r= resting adult, e= bee emerges,

7 ONTOGENESIS OF "I'HE MITE VARROA JACOBSO,~,70UD. 1N WORKER BROOD OF TIlE HONEYBEE 93 TABLE 3 Time (min./average/max.) when each larva is deposited, the interval between larvae is given in brackets. Varroa lar~;a number Author Hours 56/60/64 90/9~94 -/118/120 -/144/152 -/17~- This study aflcr (32) (26) (26) (28) capping Infantidis, (33) (27) (29) (40) tu 60" 0 < A I- Z "' 40' U nl A D 20' 0 A D A D p 2nd 3rd 4th 5th 9? 9 9 ORDER OF OFFSPRING D Fig. 3. Percentage of mortality incurred by each offspring during its development. Calculated by comparing the number of normal offspring with offspring which died or failed to develop during the time period of each stasis. P= protonymph, D= deutonymph, A= adult, represent the time period when the mortality occurred, not the mortality of that developmental stasis.

8 94 S.J. MARTIN stases, except for the 5th offspring protonymph stasis. Nearly all mortality occurred at the deutonymph stasis. This occurred mainly during the adult time period due to the stopping/slowing down of the mite development, so they never reached maturity. This reason for this is not clear. However, it is possible that the cuticle of the developing pupa became too thick for the mite to penetrate and feed. The dramatic increase in the mortality of the 3rd and 4th off:springs (Fig. 3) accounts for why although three female mites could be produced, the actual number produced is much smaller in most cells, Figure 3 shows that a average of 1.45 (94%, 2nd offspring + 38%, 3rd offspring + 13%, 4th offspring survive to become adults) female mites will be produced. No visible physical damage, e.g. crippled bees, was seen at any stage of the bees' development, despite some pupae being host to 2-3 mother mites and their offspring. Similarly, De Jong et al. (1982) found only 6% of infested bees with obvious physical deformations, in the form of wing damage. After the pupal moult, adult mites were often found already positioned between the gastric tergites of the bee. Some mother mites did not reproduce and remained in the same position, usually near the tip of the gaster, resulting in a distinctive build up of white faeces suggesting that the mite was consuming large amounts of bee haemolymph. The body of these mites remained in the un-swollen state. Other mites laid non-viable eggs (Table 2), sometimes accompanied by offspring. If present, normal offspring usually preceded the laying of non-viable egg/s and not vice-versa. DISCUSSION The overall development of the worker sealed brood follows the sequence given by Jay (1962) and Rcmbold and Kremer (1980). However, the timing of individual phases differs slightly since changes in pigmentation levels arc gradual, so the change from one phase to the next is often arbitrary and should bc taken as a general indicator and not an accurate measurc of pupal age. The timing of deposition of mite larvae and the intervals between dcpositions is comparable to other studies (lnfantidis, 1983; Akimov and Pilctskaja, 1985; Rehm and Ritter, 1989a). Very little variation in the development of the offspring was found. The juvenile mortality accounts tor why only small numbers ( 12%:, Schultz, 1984; 4%, Accorti and Nannclli, 1990; 23%, personal observation) of cells produce three mature females. The development of the egg and larva has been studied in detail (Steiner, 1993), and the developmental times given here are in accordance with those lbund in that study. The rapid expansion of the egg during vitellogencsis (Akimov and Yastrcbtsov, 1984), when the egg more than doubles in size, occurs between 20 and 35 hours (Steiner, 1993), 26 hours (Akimov et al., 1990) after the cell is sealed. This correlates well with the observed timing of swelling of the mother mite's opisthosoma (Fig. 2). Subsequent larvae are deposited at hour intervals, during which

9 ONTOGENFLSIS OF ]'lie MITE VARROA JACOBS03,'I OUD. IN WORKER BROOI') OF TIlE tloneybee 95 time the mite remains swollen. This supports Akimov's et al. (1990) observation that as each larva is deposited, the next egg is already undegoing vitellogenesis. Tables 4 and 5 compares the results of this work with previous studies. They show that although the developmental time of the male mite has remained at approximately 6 days ( 144 hours) (Table 4), the reported female mite development time varies considerably, becoming shorter with each study (Table 5). This is probably because intervals between observations have become shorter, e.g. from 1 day to 2 hours, thereby improving the accuracy. However, misinterpretation of the order in which sexes are deposited i lfantidis, 1983: Yoshida, et al., 1987: Wet and Huang, 1993) could add up to 23 hours to estimates of the development time of the first female. Piletskaya (1988) showed that a 4 C (32 C to 36 C) change in temperature greatly affected the development time of the egg in both drone (46.2 h. to 22.6 h.) and worker (54.1 h. to 24.5 h.) brood. So studies performed outside the colony, using incubators and mites artificially introduced into cells, may also distort the results. It is clear from their results that Choi and Woo (19731) rotsidentified the various mites stases which led to greatly increased times. This may also have occurred in other studies. The low level of brood removed (4% from infested combs) means that only, around 1% of the infested cells was removed. This suggests that the worker bees from these colonies were unable to detect and remove infested cells, unlike A. ceraj~a workers (Rath and Drescher, 1990). However. A. mellifera colonies in Germany, that have been exposed to the mite for many years, have been found able TABLE4 Development times of male Varroajacob.s'ot+i on Apis mell(.fera brotx.4. All times are given in hours. Figures in brackets are for those which the developmental stases,are not clear. Larva F'rotonymph Deuton', mph Total Author 24-4, Salchenko " (egg 48J 144 Choi & Wo~, "* (egg 24 II larva 24) Langhc et ol., Muravskaya (.I lfantidis, Laurcnt & Santas 'Yoshida et al "** 1(~0 Rehm & Rifler. 198% Rehm & Riller. 1989b [ Akimov et al, Wet & Huang, This qudy Scc GorNw ** The female larvae stage is probably male. meaning that the developmental time of the male was 144 hours which is not clear from the original paper. *~* Host Apis cerana.

10 96 S.J. MARTIN TABLE 5 Development times of female Varroajacobsoni on Apis mellifera bro~. All times are given in hours. Figures in brackets are for those which the developmental sta,ses are not clear. The figures given from this study represent the average values for all ~ 1st, 2nd, etc.) the female offspring. Larvae Protonymph Deutonymph Total Author 168 lan Tin-He, 1965" (egg 24) (larva 24) Saltchenko, 1966, 1972" Peking, 1972 (egg 48) (larva 96) Choi & Woo, 1973"* (larva 24-48) Lo & Chao, lsozaki, 1976"** (egg 24) (larva 24) Lange et al., Sakai et al, Muravskaya, Ifantidis, Yoshida et al., 1987"*** 149 Rehm & Rittcr, 1989a Rehm & Rittcr, 1989b 144 I fantidis, Akimov et al Wei & Huang, _+ I _ _+ 3 This study * See Gorbov, ** It is clear from the original paper that the protonymph and deutonymph stases were not correctly identified and the larva stases refers only to the male offspring, which means that the developmental time for the female is nearer 144 hours than 240 hours. *** Sakai etal., **** Host Apis cerana to remove % of cells containing I-2 mother mites after 10 days (Boecking and Drescher, 1992). Moosebeckhoffcr et al. (1988) found that 30% of cells containing bees at the stretched larvae or prepupae stages, already had some "eggs" (probably larvae) or "larvae" (probably protonymphs) of V. jacobsoni and suggested that these were deposited early by mites on their second reproductive cycle. In thc present study only three out of 119 mites had deposited their first larva during the first 60 hours alter sealing and these were all laid 56 hours after sealing. Moosebeckhoffer et al (1988) showed that during the period betbre the prepupai moult 574 mites produced 147 (26%) larvae and 11 (2%) protonymphs. However, in a colony in which the mites deposit their first larva alter 60 hours and the bees' prepupai moult occurs after 86 hours, onc would expect 574 mites to produce 172 larva and no protonymphs, since thc first protonymphs in this study appeared 84 hours after sealing. It is probable that grouping of the data for the stretched larval and prepupal stages has lead the authors to their conclusions and early deposition of larva has yet to be shown.

11 C)NIOGENF~SIS OF rile MITE i;arroa JA('OB.~.ONI OUD. IN WORKt~R BR(R)D OF TIlE HONI;YBI!E 97 Biichler and Drescher (1990) obtained a correlation coefficient of r = 0.48 between the duration of the sealed stage and levels of mite infestation. Howevm: the number of mite offspring that could complete development in worker sealed brood of colonies with the shortest {278 hours) and longest (289 hours) sealed periods would, in fact, be the same ( 1 male and 3 female) (see Fig, 2) or at most only' affects the 4th offspring. Figure 3 shows that the 4th offspring only contributes a small proposition (13%) to the total number of female mites produced, whereas the development of the 2nd and 3rd offspring ( 138% of total female mite production) will not be affected. This may account for the low coefficient value the,,' obtained. The difference in the observed mite levels could be due to variations within the colony (up to 19 hours) in the development of the worker scaled brood, mite re-invasion, small variations in the initial mite infestation and known law~e variations m the reproductivity of the mites. This is supported by the fact that colonies studied by Biichler and Drescher (1990), with the same sealed period (286 hours), had both the lowest (500) and highest (3.200) numbers of mites. Similarly, Lc Conic and Cornuet (1989) found significant differences m the duration of the worker sealed brood phases in three European races of,4. melli/era. These were attributed to differences in n'ursmg behaviour of the worker bees and in the distribution of brood within the colony. However, the differences between the shortest ( days) and longest ( days) sealed period would according to the data presented in this study, again have no effect on the number of mites produced bv cells from each colony. It can be seen from Fig. 2 that the sealed period would have to be shorter than 9.25 days (222 hours) to prevent the first female offspring maturing before the bee emerges. Moritz and H~cl (1984) found that in A. melti/era capensis it was impossible for nemly hall" (42%) of the mother mites to finish their reproductive cycle (i.e. to produce at least one mature female offspring before the bee emerged). Howevm; in A. mellit'era carnica colonies all mites were able to produce at least one mature female offspring. This was probably' due to the very short (9.6 +_ 0.07 days) period of worker sealed brood found m A. m. capepzsis, whereas the period of A. m. carnica worker sealed brood was day, s (Moritz and H~inel, 1984). This short sealed period, along with more efficient grooming was thought to be the reason for the decline in the number of miles in A. m. c~q~ensis colonies over the season (Moritz and Mautz, 1990). The results from the A. m. capensis work (Mortiz and H~inel, fit well with the duration of development of the mite reported in this study {Fig. 2). However, recent work (Moritz and Jordan, 1992) lound that the sealed period ofa. m. capensis was hours ( 11 days). This fgure is more in agreement with data on other African races of A. mellifera (Fletcher, 1978) and may be due to a sampling error in the imported stock e.g. hybridization with A. m. sculellata. (Moritz, personal communication). The period for worker sealed brood in A. cetzma is 11 day's (blishra and Dogra, 1983) which is similar to tropical A. mellifera races, and does not appear to account for the resistance of A. cerana to V. jacobsoni. It is known that V. jacobsoni can

12 98 s.j. MARTIN reproduce in A. cerana worker sealed cells (De Jong, 1988) and that enough time is available for up to two mature female mites to develop. However, the number of mites which do reproduce in worker sealed brood is extremely small. This is due to the bees' efficient detection and removal of artificially infested cells (Rath and Drescher, 1990) or the mites not becoming fertile (Rosenkranz, et al., 1993). However, it is possible that a reduction in sealed brood period may significantly change the mortality rates of the offspring and this needs to be investigated in more detail. ACKNOWLEDGEMENTS I am grateful to Dr. J. B. Ford of University of Wales, Bangor for his comments on the manuscript and "also to Joan and Len Davie at Nymett Cottage, Devon Ibr their help with maintenance of the study colonies. Also many thanks to Dr. N. Milani for providing some of the more obscure references. This work was funded by the Horticulture and Potatoes Division of the Ministry of Agriculture, Fisheries and Food. REFERENCES Accorti, M. and Nannelli, R., [Oviposition sequence and developmental time of the progeny of Varroajacobsoni Oud. on drone brood ofapis rnellifera ligustica Spin]. Apicoltura (Roma) 6: (In Italian). Akimov, I. A. and Piletskaja, I. V., [Vitality of the eggs of Varroa jacobsoni mite[. Pchelovodstw~, 8: 20. (In Russian). Akimov, I. A. and Piletskaja, I. V., [Influence of temperature on oviposition and development of Varroa jacobsoni eggsl. Vestn. Zool., 3: (In Russian). Akimov, I. A. and Yastrebtsov A. V., [Reproductive system of Varroajacobsoni. I. Female reproductive system and oogenesis]. Vestn. zool., 6: 61~8. (In Russian). Akimov, I. A., Piletskaja, I. V., Yastrebtsov, A. V., [Reproductive cycle of Varroajacobsoni and its host connections]. Vestn. zool., 2: (In Russian). Boecking, O. and Dre~her, W., The removal response ofapis mellifera L. colonies to brood in wax and plastic cells after artificial and natural infestation with Varroajacobsoni Oud. and to freeze-killed brood. Exp. Appl. Acarot., 16: Bfichler, R. and Drescher, W., Variance and heritability of the capped developmental stage in European Apis mellifera L. and its correlation with increased Varroajacobsoni Oud. infestation. J. Apic. Res., 29: Choi, S. Y. and Woo, K. S., [Studies on the bionomics of the bee mite Varroajacobsoni Ouderoans and its chemical control (1)]. Research Reports of the Office of Rural Development, Suwon, Korea 15 (Livestock): (In Korean). Choi, S. Y., Brief report on the status of Korean Beekeeping. Prec. Exp. Consultation Beekeep. with Apis mellifera in Tropical and sub-tropical Asia, Bangkok/Chiang Mai, Thailand, Evans, G. O., Principles of Acarology. C.A.B. International, 563 pp. Fletcher, D. J. C., The African bee, Apis mellifera adansonii, in Africa. Ann. Rev. Entomol, 23: Grobov, O. F., Varroasis in bees. In: Varroasis, A honeybee disease: Apimondia, Verlag, Bucharest, : Griffiths, D. A. and Bowman, C. E., World distribution of the mite Varroajacobsoni, a parasite of honeybees. Bee World, 62: Ifantidis, M. D., Ontogenesis of the mite Varroajacobsoni in worker and drone brood cells. J. Apic. Res., 23:

13 ONTOGENESIS OF THE MITE ~,~4RROA Z4COBSONI dud. IN WORKER BROOD OF TIlE HONEYBEE 99 lfantidis, M. D., Some aspects of the process of Varroajacobsoni mite entrance into honey bee (Apis mellifera) brood cells. Apidologie, 19: lfantidis. M. D., Re-examination of reproduction of the mite Varroa jacobsoni Oudemans. Proceedings of the International Symposium on recent research on Bee Pathology. Gent, Belgium, 5-7 Sept., pp Jay, S. C Colour changes in honeybee pupae. Bee World; 43: Jong, D. De, Jong, P. H. De. and Goncalves, L. S., Weight loss and other damage m developing worker honeybees from infestation with Varroa jacobso,& J. Apic. Res., 21: Jong, D. De, Varroajacobsoni does reproduce in worker cells ofapis ceraml in South Korea. Apidologie, 19: Lange, A. B., K. V. Natskii and Tatsii, V. M [The biology of the mite varroa}. Veterinariya. (7): (In Russian). Laurent J. C. and Santas L., [Study of the larval development of Varroa jacobsoiti]. Apidologie, 18:53-60 (In French with English summary). Le Conte, Y. and Cornuet. J. M., Variability of the post-capping stage duration of the worker brood in three different races of Apis melli&ra. Prec. E. C. E/xperts Group Meeting, Udine.: Lo. Kang-chen and Chad, Rou-su., [The preliminary investigations on bee mites in Taiwan]. J. Agric, Res. China, 24 ( 1/21: (In Chinese). Mt.ravskaya, A. I., [Assessment of the reproductive capacity' of female Varroa jacob.wmi]. Veterinariya. 2: t!n Russian). Mishra, R. C. and Dogra, G. S., Post-embryonic development ofapis cera~ra imlica F worker bee. Int. Conf. Apic. "rrop. Climate, 2: Moosbeckhofer, R.. M. Fabsicz, and Kohlich. A Investigations on the curtelation between rate of reproduction of Varroa jacobsoni and infestation rate of honeybee colonies. Apidologie. 19: Moritz, R. F. A., Hcritabilily of the Ix~st capping stage in Apis mellil'era and its relation to varroatosis rcsistance. J. Hered.. 76: Moritz, R. F. A. and H~tncl. H., Restricted development of the parasitic mite Varroajacobso~i dud. in the Cape honcybee Apis mellif'era capem'is Esch. Z. Angew. Entomol., 97: Moritz. R. F. A. antl Jordan, M., 1992, Selection of resistance against Varroa/acobsorti across caste and sex in the honcybce (Apis mellifera L., Hymenoptera: Apidac). Exp. Appl. Acarol., 16: Mortiz, R, F. A. and Mautz. D., Development of l,'arroa.jac~bsoni in colonies ofapis mell~fer~t capensis and ~4pi.~ mvllifera carnico. Apidologie, 21: 53-58, P, eking Agrictalt~ral Institute ~;~f S,cie!!ce, Keeping the honeybee. Agricultural Publications, P, Pi!etskaya,!.v.,!9,88. [Some peculiar feat.ures of the deveiopm,ent of tile Vam~a ja~~bso~ti mite itl wo!,k~r bee ~'~!!d drone broodl. Ves!n. Zoo!., 2~ (In Russia, n). R{!til, W. and.d, re,.~cher~ W., Response 9fApis cero~'~ Fabr tow,ards brood infested with Vctrro~ jaf'o&roni dud and intest,al;ion rate of olol!ies in Thailand. Apidologi. 2!; 31 I-3~/. Rehm. S. M..and Rittc, r, W., 1989a. Sequence of the sexes in the offspring of Vam~,iacobsom and t!~e resu!ting ~o!~seq,u.en~os for tile calcujatmn of tile d.eyelopmema! period. Api~!ogie, , R,eilm, S, M. and Rit, ter, W., 1989b. Sueqession and period of development of the male and female offspring of Varroajoc~bsani in the brood of ta.).ing workers. In: Cav.alloro R. (E,d.), "Present st0- tus of varro;ltosis in Europe and progress in the Varroa mite control", CEC, Luxembourg; 97-99, Rembold, H. and Kr!n,er, J. P:,! 980, Characterisation of post.embl3'oni developmenta! stages of the fema!, castes of the honey bee Apis mellinra L. Apidologic,!!: 29-38~ Ritter, W., Varroa di.seage of the lmneybee Apis metlifera, Bee World, I Rose!!kranz, P., N. C, Tewarson, A. Rachmsky, A. Stramhi, C. Strambi, al~d Engels W,, Juvenile hormone til; r and reproduction of Varroajacobsoni in capped brood stages ofapis cel'c.tno io(lic(t in comparison to Ap(s mellifera ligt!sticet. Apidologie, 24:

14 100 S.J. MARTIN Sakai, T., Takeuchi, K. and Hara, A., [Studies on the life history of a honeybee mite Varroa jacobsoni Oudemans in laboratory rearing]. Bull. Agric. Tamagana Univ. 19: (In Japanese). Schulz, A. E., [Reproduction and population dynamics of the parasitic mite Varroa jacobsoni Oud. and its dependence on the brood cycle Of its hostapis mellifera L.]. Apidologie, 15: (In German). Smirnov, A. M., Research results obtained in USSR concerning aetiology, pathogenesis, epizootiology, diagnosis and control of Varroa disease in bees. Apiacta, 13: Steiner, J., Sex discrimination based on external structures in nymphal and adult Varroajacobsoni mites (Acarina: Varriodae). Entomot. Gener., 14: Steiner, J., Vom Ei zur milbe: Varroajacobsoni. Deutsches Bienen Journal, 6: Wei, H., Huang, S., Study on biological characteristics of Varroa jacobsoni. Apimondia, Beijing, China: (p 101 in abstracts of reports). Yoshida, T., Sasaki, M., Yamazaki, S., [Parasitism and reproduction of Varroa mite op~ the Japanese Honeybee. Apis ceranajaponica]. In: Sasaky M., [The reason why Apis cerana japonica is resistant to Varroa mite]. Honeybee Science, 10: (In Japanese).