The nesting behaviour of the mason (Latr) with special reference to its pollinating potential (, ) J Bosch

To cite this version:

J Bosch. The nesting behaviour of the Osmia cornuta (Latr) with special reference to its pollinating potential (Hymenoptera, Megachilidae). Apidologie, Springer Verlag, 1994, 25 (1), pp.84-93. ￿hal-00891142￿

HAL Id: hal-00891142 https://hal.archives-ouvertes.fr/hal-00891142 Submitted on 1 Jan 1994

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Original article

The nesting behaviour of the mason bee Osmia cornuta (Latr) with special reference to its pollinating potential (Hymenoptera, Megachilidae)

J Bosch

Dept Biologia , Fac Biologia, Univ Barcelona, Diagonal, 645, 08028 Barcelona, Spain

(Received 5 October 1992; accepted 30 July 1993)

Summary — The nesting behaviour of Osmia cornuta was studied in managed populations released in almond orchards. A correlation between the dry weight of a provision and the number of pollen grains in the provision was established. Male and female provisions contain a mean of 4 634 304 and 8 046 425 almond pollen grains, respectively. A correlation between the dry weight of a provi- sion and the number of foraging trips was also established. The mean number of foraging trips nec- essary to complete a provision was 16.76 for male provisions and 26.79 for female provisions. Dur- ing an almond blooming period, each O cornuta female potentially visits from 9 500 to 23 600 almond flowers. Given the pollinating efficiency of O cornuta on almond, 3 females per tree are con- sidered sufficient to maximize pollination in an almond orchard.

Osmia cornuta / nesting behaviour / pollination / Prunus amygdalus

INTRODUCTION linator can be increased in orchard envi- ronments, and that O cornuta is a very ef- ficient of almond flowers (Bosch Because of its pollen-collecting prefer- and Blas, 1994). ences (Rosaceae (Taséi, 1973b; Asensio, 1984; Marquez et al, 1994)) and its flying A necessary following step in the devel- period (February-April) in synchrony with opment of O cornuta as an orchard polli- bloom of fruit trees, Osmia cornuta has nator is the estimation of the number of been studied as a potential pollinator in necessary per unit area of crop. The almond and apple orchards (Asensio, population introduced in a given orchard 1984; Torchio and Asensio, 1985; Torchio should be large enough to ensure ade- et al, 1987; Krunic et al, 1989, 1991; quate levels of pollination, and low enough Bosch, 1994a, 1994c). These studies so that it can be supported by the avail- have shown that, through adequate man- able pollen-nectar sources. The super- agement practices, populations of this pol- saturation of an orchard with Osmia bees is likely to cause an increase in the disper- Life cycle of Osmia cornuta sal of prenesting females (Bosch, 1994b) and/or a decrease in size and nectar O cornuta is a solitary bee that nests in pre- concentration of the pollen-nectar provi- existing cavities, in which it builds linear se- sions constructed, with a in- subsequent ries of cells separated by mud partitions. crease in immature mortality (Torchio, Each cell is provisioned with a pollen- 1985). nectar loaf on top of which an egg is depos- The number of individuals needed per ited. At the end of each series of cells the unit area will from orchard to orchard vary female builds a thick terminal mud plug that to numerous such according parameters, seals the nest. One female may build as as size of attractiveness of weather, trees, many as 7 nests (unpublished data). cultivars, number and distribution of pol- Like many other aculeate cavity nesters lenizers, populations of other pollinating in- (Krombein, 1967), O cornuta females usu- sects, etc. While it is not possible to meas- ally lay their female eggs in the first (inner- ure all these parameters for each orchard, most) cells of the nest, and male in it is possible to make general estimates eggs the last-built cells (outermost cells). Fe- that apply to most places and years. Sev- male are than eral approaches have been used to esti- provisions typically larger male and female bees are mate concentrations of Osmia provisions, typi- cally larger than males. needed per unit area of orchard. Some of these approaches (direct approaches) are Eggs hatch a few days after they have based on the measure of fruit-set, in plants been laid (February-March), and larvae visited by different numbers of pollinators. feed on the pollen-nectar provision until This can be accomplished by caging this is completely consumed (May-June). plants (Torchio, 1979), or by using differ- Then, fifth instar defecated larvae spin ent concentrations of pollinators in differ- their cocoons, but pupation does not occur ent places (Yamada et al, 1971) or years until August-September. At this time, bees (Torchio, 1985). Other methods (indirect) reach the adult stage and remain in dia- are based on the estimation of the number pause inside their cocoons for the winter. of flowers to be pollinated, the number of A period of low temperature is necessary visits necessary to pollinate a flower, and to break diapause (Taséi, 1973a), and the number of visits made per individual adults reactivate in February-March when pollinator (Yamada et al, 1971; Maeta and temperatures rise. Kitamura, 1974, 1981). Male emergence starts a few days be- In this study, the time spent in each fore female emergence. After mating, fe- nesting task of O cornuta, and the number males initiate nesting activities; that is, of trips necessary to provision cells, and they select a hole in which to nest, collect construct cell partitions and nest plugs pollen and nectar to provision it, lay eggs, were measured. The number of pollen and collect moist soil to build cell partitions grains per provision was also estimated. and terminal nest plugs. These data combined with results ob- tained on cell production, flower-visiting rates, and pollinating efficiency of O cornu- MATERIALS AND METHODS ta (Bosch and Bias, 1994) have provided a framework to measure accurately popula- Field observations were carried out in 1991 at 3 tion densities of O cornuta required to nesting shelters; 1 located in an experimental al- maximize pollination of almond. mond orchard at the Agricultural Research Sta- tion of Mas Bové (IRTA), and the other 2 in com- sexed by cutting a small slit below the nipple mercial almond orchards in Parc de Samà and and observing the facial pilosity of the bee. The Vila-Seca (Tarragona, NE Spain). Laboratory pilosity of males is more abundant and white, work was undertaken at the Department of Ani- whereas that of females is scant and black. The mal Biology (Univ of Barcelona) facilities. sexed cocoons were then placed in winter refrig- erators.

Provisions and mud structures Foraging and nesting activity Nests recently built in paraffinated paper straws 8 mm inside diameter and 15 cm long by fe- To determine the temperature threshold for O males of a managed population released in an cornuta foraging activity, nesting shelters were almond orchard in Parc de Samà (Baix Camp, observed in the early morning, and when all or NE Spain) in 1990 were dissected. Eggs were most of the nesting females had left their holes removed, and pollen provisions were carefully and started foraging activities; air temperature placed in pre-weighed individual aluminium was recorded at a neighbouring weather station. paper envelopes and assigned male or female In the late afternoon, when most females had re- condition according to their size and position turned and remained inside their nests, air tem- within the straw. Mud partitions and terminal perature was again recorded. were also removed and plugs carefully wrapped Some holes in which nesting activities were in of aluminium Enve- pre-weighed pieces paper. under way were continuously observed, and the lopes with mud partitions were dried in a stove at time spent by nesting females inside and out- 60°C, and envelopes with pollen-nectar provi- side the nest was measured. This was done for sions were dehydrated in a lyophilizer. All sam- females collecting pollen and nectar, and for fe- ples were then weighed on a precision balance. males collecting mud. To establish a correlation between pollen weight and number of pollen grains, small pieces of dehydrated provisions were cut and Mud loads weighed on a precision balance. These pieces were then introduced into glass vials with 10 ml distilled water and some detergent to facilitate To determine the quantity of mud that a bee separation of pollen grains. The liquid was transports to the nest on each mud-collecting stirred until the pollen provision portion had com- trip, holes in which partitions or plugs were be- pletely disintegrated. The vial was placed in a ing built were observed until the nesting female vortex mixer, to allow for even suspension of came back with a mud load. When the bee was pollen grains throughout the liquid, and 10 5-μl about to enter the hole, it was gently pressed samples of the suspension were taken with a with a finger against the nesting material, which micropipette and individually deposited on a caused it to release the mud load and fly away. glass slide. Pollen grains in each droplet were Mud loads obtained in this way were collected, counted under the microscope, and the mean dried at 60°C and weighed. number of pollen grains counted was multiplied by 2 000 to obtain an estimation of the number of in each of pollen grains piece provision. Pollen loads

Cocoons To estimate the quantity of pollen transported by a bee on each foraging trip, holes in which a bee was about to provision a new cell were se- Other nests in 8 mm x 15 cm and 7 mm x 15 cm lected. These holes were continuously observed straws from the same population were left undis- and the number of times a bee entered them turbed until October, at which time nests were with pollen loads on the were recorded. dissected, cocoons were separated, and all fe- At the end of the day the straws containing cal particles were removed. Each cocoon was these half-provisioned cells were removed from the nesting material and taken to the laboratory. the nest and moulded with the mandibles The straws were then dissected and the partial until a concave partition covering the hole provisions were carefully removed and placed in diameter of the nesting hole is formed. pre-weighed aluminium envelopes. These sam- ples were dehydrated and weighed as de- Following the sequence described scribed above. In this way, an estimation of the above, a series of cells are provisioned and dry weight of a pollen load was obtained. sealed. At the end of each series, females may optionally build 1 (rarely 2) vestibular cells (empty cells), and the nest entrance is RESULTS AND DISCUSSION then sealed with a thick mud plug. The times spent on each nesting activity are included in table I. Thus, mud collect- behaviour Nesting ing trips are shorter than foraging trips, but mud deposition takes about as long as Nesting O cornuta females initiated forag- nectar-pollen deposition. These data are ing activity in the morning when the air similar to those reported for Japanese Os- temperature approached 12°C (mean: mia species, which spend at least 6-8 min 12.19; SD: 0.55; n = 11). The temperature on each pollen-nectar foraging trip in apple at which bees suspended foraging activity orchards, and 1-5 minutes on each of the in the afternoon or evening ranged from 13 other nesting activities (Maeta, 1978). to 16°C, and it increased from the begin- Similarly, propinqua spend about 1 min ning to the end of the nesting season. per mud collecting trip, and 14 min in the field O cornuta females collect nectar and per foraging trip (Levin, 1966). The time spent this species de- pollen in tandem when they visit almond by positing mud and pollen-nectar loads is flowers (Bosch and Blas, 1994). Nectar is about 1.5 min (Levin, 1966). carried in the crop, and pollen on the ven- O cornuta females also tral scopa of the abdomen. When the bee spend variable returns to its nest, it enters head first and amounts of time searching for and inspect- regurgitates the nectar load. It then crawls ing holes in which to nest, cleaning holes, intruders from their etc. out of the hole, where it turns and re- expelling nest, enters the nest backwards. The bee then Mean weights of mud partitions, termi- crawls back to the bottom of the nest and nal plugs and mud loads carried by the brushes its abdomen with its hind legs so that the pollen load is deposited on the pro- vision initiated. After several of these trips, the provision is completed and the female lays the egg on the surface of the provision. Upon egg deposition, the bee exits the hole and flies away, to return some time later with a glob of mud held under the head and behind the mandibles. This nest- ing material is collected with the mandibles at crevices and holes in the ground, where soil is moist but not super-saturated. Many females have been observed to collect moist soil at the same source, adjacent to one another. Mud loads are deposited in bees are indicated in table II. Intercell parti- chio and Tepedino, 1980). The expected tions are the lightest, followed by initial par- sex ratio of O cornuta calculated from its titions, vestibular partitions, and nest provision weights is 1.74 males:1 female, plugs. By dividing the weights of these dif- which is slightly higher than sex ratios ob- ferent mud structures by the mean weight tained in wild populations (1.1-1.5 males:1 of a mud load, the number of mud collect- female) (unpublished data). ing trips necessary to build each structure Female cocoons recovered from the is obtained The number of loads (table II). 8 mm straws weighed 0.235 ± 0.032 g (n = used to build an intercell is lower partition 50), and male cocoons weighed signifi- than in O (4.4) lignaria propinqua (8-12) cantly less 0.142 ± 0.024 g (n = 50) (t — (Torchio, 1989). Student = 16.55, p < 0.0001). Female co- coons from the 7 mm straws weighed 0.202 ± 0.021 g (n = 50), and male co- Theoretical sex ratio coons weighed 0.129 ± 0.022 g (n = 50) (t — Student = 17.03, p < 0.0001). Differenc- The distribution of dry weights of male and es in weight between female cocoons from female provisions overlapped, but the either 8 or 7 mm straws and between male mean weight of the latter (0.451 ± 0.069 g, cocoons from both types of straws were n = 61) was significantly higher than that of statistically significant (t — Student = 6.10, < — = < the former (0.260 ± 0.034 g, n = 58) (t- p 0.0001; and t Student 2.74, p Student = 19.05, p < 0.0001). 0.01, respectively). The theoretical sex ra- tio calculated from cocoon was According to Fisher’s (1930) theory, pa- weights rental investment should be equally distrib- similar to the theoretical sex ratio calculat- ed from 1.66 males uted among progeny of both sexes, and provision weights: per female form the 8 mm straws, and 1.57 therefore, it is expected that fewer mem- males female for the 7 mm straws. bers of the sex with highest production per costs will be produced. Weights of pollen- nectar provisions made by bees are easy Number to measure and can be considered an in- of pollen grains per provision dex of the energy invested by parent fe- males in their progeny. For these reasons, All samples used to establish a correlation Fisher’s theory has been applied to a num- between pollen weight and number of pol- ber of species of solitary bees to obtain len grains contained more than 90% al- their expected (optimal) sex ratio (eg, Tor- mond pollen by number. Correlation be- tween these 2 parameters was significant (N = 37, r = 0.936, p < 0.00001 ) (fig 1) with the regression equation: When the average weights of male and fe- male provisions were added to the equa- No pollen grains = 17 858 298 * weight — tion, we found that 13.48 to 21.40 foraging 3188 trips (mean: 16.76, SD: 1.76) are neces- sary to complete a male provision, and Thus, male provisions in the 8 mm x 15 18.78 to 35.36 (mean: 26.79, SD: 3.62) are cm straws contain 4 634 304 ± 598 283 needed to complete a female provision. pollen grains (range: 3 516 682- Maeta (1978) reported that a provision of 6 213 285), and female provisions contain O cornifrons requires 11-35 pollen loads, 8 046 425 ± 1 232 940 pollen grains and a provision of O taurus 12-37 loads. (range: 5 322 156-10 961 807). Maeta Similarly, 14-35 foraging trips are neces- (1978) estimated about 7 000 000 apple sary to provision a cell in O lignaria (Tor- pollen grains per provision in Osmia corni- chio, 1989). frons and about 9 000 000 in Osmia taur- As indicated in table I, foraging trips av- us. erage 11.48 min. Some of the time is spent flying from the nest to the first tree visited, grooming, and flying back to the nest. If we Number of flowers visited per provision presume that these activities consume a third of the time the bee spends outside its A significant correlation was found be- nest, bees visit flowers for 7.65 min per for- tween the number of foraging trips and the aging trip. Visitation rates of 77 O cornuta weight of pollen-nectar provision (n = 34, females averaged 7.9 visits per min r = 0.853, p < 0.00001) (fig 2) with the re- (Bosch and Blas, 1994), which translates gression equation: to 60 flowers visited per foraging trip. Therefore, 1 000 flower visits are made to egg-laying potential in O cornifrons, O lig- produce a male provision and 1 600 visits naria and O cornuta seems to be around to produce a female provision. Maeta and 30 eggs (Maeta and Kitamura, 1974; Mae- Kitamura (1974) assumed that all time ta, 1978; Phillips and Klostermeyer, 1978; spent by O cornifrons on a foraging trip Torchio et al, 1987; Torchio, 1989). These was devoted to visiting flowers, and they, data suggest that an O cornuta female can therefore, concluded that this species visit- visit from 9 500 to 23 600 flowers while ed as many as 90 flowers per foraging trip. nesting in almond orchards. As a consequence, they estimated that an O cornuta females contact the stigma of average O cornifrons provision required almond flowers in 98.71% of their visits, 1 620 flower visits. and single visits of this pollinator to other- wise unpollinated flowers of the cultivar Ferragnes yielded fruit sets of 21.84 and of O cornuta Pollinating potential 38.10% in 1989 and 1991, respectively on almond (Bosch and Blas, 1994). Hand-pollination experiments on Ferragnes in 3 different or- In earlier studies with marked and un- chards in 1900 and 1992 yielded fruit-sets marked bees released in almond orchards, of 8.92, 25.49 and 38.29% (unpublished the number of male and female cells pro- data); 40% is considered the maximum duced per O cornuta nesting female was bearing capacity of almond (Kester and determined (Bosch, 1992). In some of Griggs, 1959). These data suggest that al- these releases, the time available for nest- most all visits by O cornuta females cross- ing activities was considerably reduced for pollinate almond flowers. a of reasons. Under these circum- variety An average almond tree in a commer- stances each O cornuta female construct- cial orchard in NE Spain bears about ed 2.5 female cells and 5.5 male cells on 12 000 flowers (Romero, personal commu- 1994a, 1994c, and un- average (Bosch, nication). According to our results, 3 O cor- Under more favourable published data). nuta females would make a minimum of conditions and (extended blooming periods 28 500 visits, which means that each flow- cell was as good weather) production high er would receive, on average, a minimum as 6 female cells and 14 male cells per of 2.37 visits. Given the high pollinating ef- female These nesting (Bosch, 1994c). fig- ficiency of O cornuta females, this seems ures are similar to cell data re- production to be more than enough to secure ade- for other Osmia O ported pollinators: lig- quate pollination. naria propinqua foraging in an apple O cornuta is a close relative to O corni- orchard produced 7.97 cells per female re- frons and O which have leased when pollen-nectar resources were lignaria propinqua, too low to adequately support the entire been developed as orchard pollinators in and the US, population, and 17.59 cells per female re- Japan respectively (Maeta and The leased when these resources were abun- Kitamura, 1974; Torchio, 1993). dant (Torchio, 1985); when pollen-nectar number of recommended females per tree is 2.5-3 and resources in apple orchards were moder- (Maeta Kitamura, 1974, 1981) ately abundant, O cornifrons provisioned 4 for O cornifrons on apple (Yamada et al female and 6 male cells, and when they [1971 ] give values of 3-11), and 3 for O lig- almond were abundant, these numbers increased naria propinqua on (Torchio, 1991). to 5 and 7, respectively (Maeta and Kita- Finally, the pollinating potential of male mura, 1974; Maeta, 1978). The maximum O cornuta should not be neglected. Males do not live as long as females and they séchées, pesées et mises en suspension surely visit fewer flowers, but they are dans l’eau. Des échantillons de ces sus- more numerous than females, and al- pensions de pollen ont été prélevés et les though their pollinating efficacy has not grains de pollen ont été dénombrés. On a been directly measured, they always land pu ainsi établir une corrélation entre le on the reproductive organs of almond flow- nombre de grains de pollen et le poids sec ers, and are, therefore, likely to be good d’une charge (fig 1). Les provisions dans pollinators. les cellules de mâles et les cellules de fe- melles contenaient en moyenne 4 364 304 et 8 046 425 grains de pollen d’amandier, ACKNOWLEDGMENTS respectivement. Pour certains nids on a compté le nombre de voyages d’approvi- sionnement. En for- I am grateful to J Calzadilla and MA Escolano pesant les provisions for their assistance in field work, and to M mées, on a pu établir une corrélation entre López for her assistance with laboratory work.I le poids sec d’une provision et le nombre thank J Retana (CREAF) for reviewing an early de voyages de butinage (fig 2). Le nombre draft of the manuscript. This study was support- moyen de voyages de butinage nécessaire ed by an FPI grant of the Spanish Ministerio de pour faire une était de 16,76 pour Educación y Ciencia, and by funds of the INIA provision les des cellules de mâles et de (project no 8191). provisons 26,79 pour celles des cellules de femelles. La combinaison de ces résultats avec les Résumé — Comportement de nidifica- données portant sur le nombre de fleurs vi- tion de l’abeille maçonne Osmia cornu- sitées par unité de temps et sur la produc- ta (Latr) en rapport avec son efficacité tion des cellules donne une estimation du pollinisatrice (Hymenoptera, Megachili- nombre de fleurs d’amandier visitées par dae). On a étudié le comportement de ni- une femelle d’O cornuta qui nidifie dans un dification d’Osmia cornuta sur des popula- verger d’amandiers : il varie entre 9 500 tions introduites dans des vergers (dans des conditions météorologiques dé- d’amandiers. Le temps passé par les fe- favorables et des périodes de butinage melles à l’intérieur et à l’extérieur du nid a courtes) et 23 600 (lorsque le temps est fa- été mesuré. Les parois en boue des cellu- vorable et la floraison prolongée). Étant les et les charges de boue transportées donné l’efficacité pollinisatrice d’O cornuta, par les abeilles ont été rassemblées et pe- on considère que 3 femelles par arbre est sées. Les voyages des femelles pour ré- un nombre suffisant pour assurer une polli- colter la boue duraient en moyenne 2,72 nisation maximale d’un verger d’aman- min et le déchargement 1,87 min (tableau diers. I). Le nombre de voyages nécessaires pour construire une paroi en boue était de Osmia cornuta / nidification / pollinisa- 6,88 pour les premières parois, de 4,41 tion / Prunus amygdalus pour les parois intermédiaires, de 11,88 pour les parois du vestibule et de 21,09 pour les bouchons terminaux (tableau II). Zusammenfassung — Das Nistverhal- Les voyages de récolte de nectar et de ten der Mauerbiene Osmia cornuta Latr pollen duraient en moyenne 11,48 min et (Hymenoptera, Megachilidae) mit be- le dépôt du nectar et du pollen 2,28 min. sonderer Berücksichtigung ihrer Bedeu- La ponte durait 4,11 min (tableau I). Les tung für die Bestäubung. Das Nistverhal- provisions de nectar et de pollen ont été ten von Osmia cornuta wurde in künstlich gehaltenen Populationen im Bereich von blüten besucht. Berücksichtigt man die Lei- Mandelkulturen studiert. Es wurden die stungsfähigkeit von O cornuta bei der Be- Zeiten gemessen, welche die Weibchen stäubung der Mandel, so kommt man zu inner- und außerhalb des Nestes verbrach- dem Schluß, daß drei Weibchen pro Baum ten. Die Zellwände aus Erde und die von zur optimalen Bestäubung in einer Mandel- den Bienen transportierten Erdladungen pflanzung ausreichen. wurden gesammelt und gewogen. Die Weibchen verbrachten im Durchschnitt Osmia cornuta / Nistverhalten / Bestäub- 2,72 min auf einem Ausflug zum Sammeln ung / Prunus amygdalus von Nistmaterial, und 1,87 min zur Anbrin- gung des Materials (Tabelle I). Die Zahl der Ausflüge zum Sammeln von Nistmate- REFERENCES rial betrug für Zellwände am Beginn des Nestes 6,88, für Zwischenwände 4,41, für Asensio E Osmia cornuta Latr Vorraum-Wände 11.88 und für den Ab- (1984) (Osmia) pollinisateur potentiel des arbres fruitiers en schluß-Pfropf 21,09 (Tabelle II). Ausflüge Espagne (Hymenoptera, Megachilidae). Ve zum Sammeln von Pollen und Nektar dau- Symp Int sur la Pollinisation, Versailles, erten im Durchschmitt 11,48 min, das Ab- INRA, 461-465 lagern von Pollen und Nektar 2,28 min, Bosch J (1992) Osmia cornuta (Latr) (Hymenop- das Ablegen von Eiern 4,11 min (Tabelle I). tera, Megachilidae) como polinizador poten- Der Pollen-Nektarvorrat der Zellen cial de almendros. Ph D Thesis, University of Barcelona. 177 p wurde getrocknet, gewogen und in Wasser Bosch J Osmia cornuta suspendiert. Dann wurden Proben von (1994a) (Latr) (Hyme- diesen Pollensuspensionen entnommen, noptera, Megachilidae) as a potential pollina- tor in almond orchards: releasing methods um die Pollenkörner zu zählen. Auf diese and nest-hole length. J Appl Entomol 115 (in Weise konnte eine Korrelation zwischen press) der Pollenzahl und dem Trockengewicht Bosch J (1994b) Comparison of nesting materi- werden Vorräte in gewonnen (Abb 1). Die als for the orchard pollinator Osmia cornuta männlichen und weiblichen Zellen enthal- (Latr) (Hymenoptera, Megachilidae). Entomol ten im Mittel 4.634,304, bzw 8.046,426 Gen 19 (in press) Mandelpollenkörner. Bosch J (1994c) Improvement of field manage- In einigen Nestern wurde die Zahl der ment of Osmia cornuta (Latr) for almond polli- Sammelausflüge gezählt. Durch Feststel- nation (Hymenoptera, Megachilidae). Apido- lung des Trockengewichts der Vorräte logie 25(1), 62-74 Blas M behaviour and wurde eine Korrelation zu der Zahl der Bosch J, (1994) Foraging Ausflüge hergestellt (Abb 2). Die mittlere pollinating efficiency of Osmia cornuta (Latr) and mellifera L on almond Zahl der für die Vorrä- Apis (Hymenopte- Sammelflüge betrug ra, Megachilidae and Apidae). Appl Entomol te/Zelle in männlichen Zellen 16,76 und für Zool 29 (in press) weibliche Zellen Eine Kombination 26,79. Fisher RA (1930) The Genetical Theory of Natu- dieser Zahlen mit den Daten über die Zahl ral Selection. Clarendon Press. Oxford der Blütenbesuche Zeiteinheit und die pro Kester DE, Griggs WH (1959) Fruit setting in the von Zellen und die Produktion ermöglichte almond: the effect of cross-pollinating various Schätzung, daß ein Weibchen von O cor- percentages of flowers. Proc Am Soc For nuta in einer Mandelanlage zwischen Hort Sci 206-213 9,500 (bei ungünstiger Witterung mit be- Kromblein KV (1967) Trap-nesting Wasps and schränkter Flugzeit) bis 23,600 (bei günsti- Bees. Life Histories, Nests and Associates. gen Wetter und langer Blühzeit) Mandel- Smithsonian Press. Washington, DC 570 pp Krunic MD, Jankovic D, Stanisic T (1989) Os- Torchio PF (1979) Use of Osmia lignaria Say as mia cornuta Latr (Hymenoptera, Megachili- a pollinator of caged almond in California. dae) — Potential orchard pollinator. Arh Biol Proc IVth Int Sympt on Pollination Md Agric Nauka 41, 33-37 Exp Sta Spec Misc Publ 1, 285-293 Krunic MD, Brajkovic MM, Mihajlovic LS (1991) Torchio PF (1985) Field experiments with the pol- Management and utilization of Osmia comu- linator species Osmia lignaria propinqua Cres- ta Latr for orchard pollination in Yugoslavia. son, in apple orchards: V, (1979-1980), meth- Sixth Int Symp on Pollination. Acta Hortic ods of introducing bees, nesting success, 288, 190-193 seed counts, fruit yields (Hymenoptera, Mega- J Kans Entomol 448-464 Levin MD (1966) Biological notes on Osmia lig- chilidae). Soc 58, naria and (Hymenoptera: Torchio PF (1989) In-nest biologies and devel- , Megachilidae). J Kans Entomol opment of immature stages of three Osmia Soc 3, 524-535 species (Hymenoptera, Megachilidae). Ann Entomol Soc Am 599-615 Maeta Y (1978) Comparative studies on the bi- 82, ology of the bees of the genus Osmia in Ja- Torchio PF (1991) Bees as crop pollinators and pan, with special reference to their manage- the role of solitary species in changing envi- ment for pollination of crops (Hymenoptera, ronments. Sixth Int Symp on Pollination. Acta Megachilidae). Bull Tohoku Nat Agric Exp Hortic 288, 49-61 Stn # 57, 221 pp Torchio PF (1993) A case history in the develop- Maeta Y, Kitamura T (1974) How to manage the ment of Osmia lignaria propinqua as a man- Mame-Ko bee (, Radosz- aged pollinator of orchard crops. In: Experi- kowski) for pollination of fruit crops. Ask Co mental Studies in Pollination and Pollinator Ltd 16 pp Foraging Efficiency (SL Buchmann, ed) (in Maeta Y, Kitamura T (1981) Pollinating efficien- press) cy by Osmia cornifrons (Radoszkowski) in re- Torchio PF, Tepedino VJ (1980) Sex ratio, body lation to required number of nesting bees for size and seasonality in a solitary bee, Osmia economic fruit production. Honey-bee Sci- lignaria propinqua Cresson (Hymenoptera, ence 2, 65-72 (in Japanese) Megachilidae). Evolution 34, 993-1003 Márquez J, Bosch J, Vicens N (1994) Pollens Torchio PF, Asensio E (1985) The introduction collected by wild and managed populations of the European bee, Osmia cornuta Latr, of the potential orchard pollinator Osmia cor- into the US as a potential pollinator of or- nuta (Latr) (Hymenoptera, Megachilidae). J chard crops, and a comparison of its man- Appl Entomol 115 (in press) ageability with Osmia lignaria propinqua Phillips JK, Klostermeyer EC (1978) Nesting be- Cresson (Hymenoptera, Megachilidae). J havior of Osmia lignaria propinqua Cresson Kans Entomol Soc 58, 42-52 (Hymenoptera, Megachilidae). J Kans Ento- Torchio PF, Asensio E, Thorp RW (1987) Intro- mol Soc 51, 91-108 duction of the European bee, Osmia cornuta, Taséi JN (1973a) Observations sur le déve- into California almond orchards (Hymenopte- loppement d’Osmia cornuta Latr et Osmia ra, Megachilidae). Environ Entomol 16, 664- rufa L (Hymenoptera, Megachilidae). Apido- 667 logie 4, 295-315 Yamada M, Oyama N, Sekita N, Shirasaki S, Taséi JN (1973b) Le comportement de nidifica- Tsugawa C (1971) The ecology of the mega- tion chez Osmia (Osmia) cornuta Latr et Os- chilid bee Osmia cornifrons and its utilization mia (Osmia) rufa L (Hymenoptera, Megachili- for apple pollination. Bull Aomori Apple Exp dae). Apidologie 4, 195-225 Stn 15, 1-80