Original article

Comparison of the effects of two kinds of commercially available pollen on colony development and queen production in the bumble Bombus terrestris L (, )

MF Ribeiro, MJ Duchateau, HHW Velthuis

Department of Comparative Physiology, Ethology and Socio-Ecology Group, Utrecht University Postbus 80.086, 3508 TB Utrecht, the Netherlands

(Received 16 November 1995; accepted 3 April 1996)

Summary — The effects of two kinds of stored pollen (fresh and dried, kept in the freezer) on individuals and colonies of Bombus terrestris were investigated. The pattern of colony development, and the number and biomass of workers and males were similar for colonies fed on fresh-frozen pollen (FFP) and dried-frozen pollen (DFP). However, queens reared on DFP were smaller, had lower biomass, higher mortality and produced smaller colonies than queens reared on FFP.

Bombus terrestris / bumble bee / pollen / nutritive value / queen production

INTRODUCTION cated the use of fresh pollen to initiate colonies in captivity (Hasselrot, 1952; Free and Butler, 1959; 1966; In extensive commercial rearing, bumble Plowright and Jay, Röseler, 1985). bee colonies are normally fed on fresh pollen collected from pollen traps at Pollen and nectar are essential food hives. However, because fresh pollen is not sources for . Being rich in proteins, affects their abil- easy to obtain all the year round, dried- pollen growth, longevity, to rear as as the frozen pollen (DFP) is a possible alterna- ity brood well development of their fat bodies and tive. DFP has advantages over fresh pollen ovaries, hypo- in the sense that it is cheaper and is less pharyngeal glands (Maurizio, 1950, 1951; likely to become contaminated by fungi and Standifer, 1967; Knox et al, 1971; other microorganisms. On the other hand, it Duchateau and Velthuis, 1989). is unattractive to queens probably because However, not all kinds of pollen have the of its unpalatability (Pomeroy and Plowright, same nutritive value. One of the main rea- 1980). In fact, many authors have advo- sons is that the chemical composition (ie, amount of amino acids, lipids, sugars, min- MATERIALS AND METHODS erals and vitamins) varies considerably from one to another and plant (Todd We have used two kinds of pollen in our study, Bretherick, 1942; Vivino and Palmer, 1944; one fresh and the other dried. The fresh pollen Standifer, 1966; for reviews see Haydak, was collected from pollen traps at honey bee 1970; Herbert, 1992). colonies kept at Utrecht University. The dried pollen was bought from a commercial company. In that has a low starch general, pollen Although both kinds of pollen differed in compo- content (and high oil or sugar levels) is con- sition (table I), and time of storage in the freezer sidered to be a valuable food source for bees. (1 year for the fresh pollen and 2 years for the In contrast, pollen containing abundant starch dried pollen), they could be considered of good of them were starchless and/or (and only a small amount of lipids) is of little quality (ie, many had a content, table Moreover, value and Baker, Trees and high protein I). (Baker 1979). we decided to compare them because both are that for plants produce high quality pollen available commercially and there is a great inter- honey bees include Crocus, Papaver, Salix, est in obtaining more information about the most Erica, fruit trees (plum, apple, pear, chest- adequate pollen diet to rear bumble bees. nut), and clovers. Poor quality pollen comes The pollen pellets were ground and supplied from coniferous trees (Pinus, Cedrus, etc) liberally to the colonies throughout the experi- (Vivino and Palmer, 1944; Maurizio, 1950; ment. The colonies were also given a sugar/water and Linskens, 1974). solution at a concentration of 50%, in separate Stanley containers. The nutritive value of pollen may also be B terrestris queens were collected in early influenced by factors such as drying, ageing spring 1990 in the Gimborn Arboretum and in the or storage conditions. Dried old pollen may Botanical Gardens of Utrecht University, The have very negative effects on bee growth Netherlands. Colonies were reared in the labo- (reduced dry weight of heads, thoraces and ratory following the method of Duchateau (1985) and Duchateau and Velthuis Their devel- abdomens), on the development of (1988). opment was monitored daily. hypopharyngeal glands and on brood rear- ing (Levin and Haydak, 1957; Haydak, 1961, Initially, all colonies received fresh-frozen when the to 1963; Hagendorn and Moeller, 1968). But pollen (FFP). However, queen began lay the eggs of the second brood, the colonies the of degree impairment depends mainly were divided into two groups, in such a way that on the drying methods (Groot, 1953; Maur- both groups had a similar number of egg cells. izio, 1958, 1960). For this reason, it is rec- The first group continued to receive FFP (n = 8) ommended that the temperature during the while the second started to receive DFP (n = 7). From this moment the of con- drying process should not exceed 45 °C, quantity pollen and the moisture should be removed grad- sumed was measured by weighing the amount provided and the amount left unconsumed after ually (Chambers, 1990). 24 h. Because there was an obvious difference in So far, however, we have hardly any moisture, specific amounts of the two kinds of information about the nutritional value of pollen were kept in an oven at 60 °C, for 24 h, and then measured The differences pollen to bumble bees, although Regali and again. between the final and the initial weights were wrote a on Rasmont (1995) paper recently used to correct for the water content. Since the the subject. Since bumble bees are becom- length of the experiment varied according to the ing increasingly important as greenhouse colony (see below), the pollen consumption per pollinators, investigations into this matter day was also calculated. No correction was made for the number or are urgently needed. biomass of individuals because the amount of food eaten by workers, males and In this work our aim was to compare the queens, either in the larval or adult stages, was effects of two kinds of pollen available com- not measured separately. on L at indi- mercially Bombus terrestris the Because we were interested only in the off- vidual and colony level. spring produced by the queen the colonies were killed no later than 25 days after the workers A minimum of ten workers was the criterion started to lay eggs. used to define a colony. Queens that produced fewer than ten workers were not considered to Several parameters were recorded for the be colonies, namely the number of egg cells, the colony-producers. number and biomass (dry weight) of workers, Data on the queens’ mortality were recorded males and queens produced, the day on which so that we could calculate the percentage of sur- the queen switched to laying haploid eggs (switch vivors. The size of the queens was determined point) and the day of the first workers’ oviposition by measurements of their biomass (dry weight) (competition point; both moments were calculated and length of their wing’s radial cell (Owen, 1988). in relation to the emergence of the first worker; The radial cells of the mother-queens were also for details see Duchateau and Velthuis, 1988). measured. The young queens produced were given The statistical tests used were Mann-Whit- opportunities to mate and hibernate. Queens that ney U, Chi-Square, ANOVA and ANOVA nested did not mate after several attempts were left out- (Sokal and Rohlf, 1981). In some cases, both side or killed. A few queens did not enter the mean and median are included in the figures to hibernation phase and showed the typical give a more complete picture of the data. behaviour that indicated they were ready to lay eggs. They were put directly into small cages to see if they would find colonies. The number of RESULTS queens was corrected in the light of these fac- tors when necessary (for instance, to calculate the of percentages survivors). The water content of the FFP was 17.5% Hibernation occurred in a cold room (5 °C) and that of the DFP was 5%. The results and lasted 3 or 6 months (for details see for the corrected consumption of pollen (total Duchateau, 1985). After that, the queens were and are in placed in flight cages for 10 days and then in consumption/day) presented fig- wooden boxes kept in a room at 28 °C and 60% ure 1. Colonies which received DFP con- of relative humidity. The rearing method used for sumed less pollen than the ones which these queens was the same as that used for their received FFP (almost 1.5 times less), but mothers. The the entire pollen provided during the difference was not significant. The pro- was fresh-frozen and com- colony development of cells in the third posed of 88% Cruciferae (Brassica), 9% Acer- ductivity (number egg brood and number of individuals aceae (Acer), 2% Compositae (Taraxacum), 1 % per egg Rosaceae (fruit tree). cell) was similar for the two groups of colonies (table II). No significant difference fed on FFP and DFP, respectively. For the was found in the number of workers, males competition point the average values were and queens produced (fig 2). The switch 32.2 ± 9.7 days (FFP) and 31.3 ± 8.9 days and competition points were similar, too. (DFP). The biomass of workers, males and The averages for the switch point were 17.0 queens was lower for colonies fed on DFP ± 6.5 days and 16.7 ± 8.8 days for colonies colonies than for colonies which received FFP, but only for queens was the difference highly significant (P ≤ 0.001, table III). Queens produced by colonies fed on DFP were much smaller than those pro- duced by colonies fed on FFP. The length of the radial cell was highly significantly dif- ferent (P < 0.001; fig 3). The radial cells of the mother-queens were of similar length and averaged 4.25 ± 0.06 mm (n = 4; colonies fed on FFP) and 4.28 ± 0.10 mm (n = 6; colonies fed on DFP). Young queens reared on FFP and DFP had a moderate mating success; for both groups the success was above 60% (no sig- nificant difference; table IV). The percentage of survivors before hibernation was much lower in the colonies which received DFP than in the colonies which received FFP (highly significant dif- ference, P = 0.0001; table IV). After hiber- nation the survival was also lower, how- ever, the difference was not significant (table IV). When we consider the total mor- tality of the young queens at colony level, we find that although there was a general high mortality, the colonies which received DFP presented larger percentages; four colonies even had a mortality of 100%! Among the colonies that received FFP, how- ever, only two colonies had a high mortality: around 68% (fig 4). Queens that did not go into hibernation behaved in different ways. The two queens reared on FFP produced small colonies. The five queens reared on DFP did not pro- duce any colony at all. As regards the founding of colonies by queens that went into hibernation, the queens from colonies fed on FFP were again more successful than those from colonies fed on DFP, but the difference was Furthermore, both groups of queens dif- not the significant (table IV). However, fered tremendously with regard to the num- colonies produced by queens reared on DFP ber of days that elapsed before they started were much smaller than the colonies pro- to the reared on duced by queens reared on FFP (averages lay eggs; queens 49 ± 35.55 and 112 ± 65.06 workers, respec- FFP began to lay eggs much sooner than the difference was < tively); significant (P the queens reared on DFP (P = 0.0001; 0.05, Mann-Whitney test). fig 5). DISCUSSION smaller than the queens reared on FFP. This result was confirmed when the same The quantity of DFP consumed was lower kind of pollen (Brassica) was compared in than the quantity of FFP pollen consumed. A possible reason for the difference is that the drying process modified some of the substances that act as phagostimulants (a free fatty acid and/or a neutral lipid; Lepage and Boch, 1968; Robinson and Nation, 1968; Schmidt, 1985). Another possible explanation is that the palatability of DFP was reduced (Robinson and Nation, 1968) because there was an accumulation of phospholipids due to the loss of membrane integrity of the pollen grains. DFP did not have a negative effect on colony development and on workers and males (number and biomass). Most of the values for colonies fed on DFP tended to be smaller than those for colonies fed on FFP, but the differences were not signifi- cant. The young queens, however, were dras- tically affected by DFP. They were much fresh-frozen and dried-frozen conditions: Although we did not calculate heritabil- significantly smaller queens were produced ity (Falconer, 1981), our results indicate that by the DFP in comparison to the FFP the size of the young queens was a result of (Duchateau, unpublished data). environmental (pollen) rather than genetic forces, because mother-queens were of similar size.

The fact that colonies reared on DFP produced smaller queens may be related to a lower amount of proteins in this pollen. DFP was composed almost only of Bras- sica which is considered to be a very good quality pollen, probably because of the high protein content (around 22%, Regali and Rasmont, 1995). Indeed, as already men- tioned, fresh-frozen Brassica pollen pro- duced larger queens (Duchateau, unpub- lished data). This could be an indication that the drying process affected the pro- teins of the pollen. In fact, loss of protein in dried (old) pollen was detected by Svo- boda (1940) and Haydak (1963). Of the ten amino acids found to be essential for honey bees (Groot, 1953), half of them (methio- nine, lysine, arginine, tryptophan and cys- tine) are particularly sensitive to heating (Liener, 1958). These amino acids are often the pollen was observed. This could indi- found in pollen collected by honey bees cate a reaction between amino acids and (Auclair and Jamieson, 1948; Bieberdorf et reducing sugars (Leiner, 1958).) Therefore, al, 1961), and could be affected by the dry- as a result of leakage and loss and/or dam- ing process. Stored pollen was also found age of substances, the bees may have less to contain fewer amino acids (Dietz and nutrients available for digestion and absorp- Haydak, 1965). Besides destroying some tion. It seems that the queen larvae are amino acids, heating may also lead to especially sensitive to reduced amounts of changes in the molecular structure of the nutrients. The latter then becomes more protein. Our results also indicate that DFP pro- resistant to enzymatic digestion (Liener, voked higher mortality (especially before 1958). Since the final pollen digestion hibernation) in the young queens produced. occurs through the action of enzymes Larger queens do indeed have a higher sur- and Linskens, this could (Stanley 1974), vival rate (Holm, 1972; Owen, 1988). have serious consequences for the absorp- In our the survivors also had tion and use of such substances by the experiment bees. less success in producing colonies. Holm (1966) suggested that a queen’s ability to Other nutrients such as vitamins might start and develop a colony depends on her also be damaged or destroyed by drying, previous nutritional condition. and storage (predominantly the vitamins the lower survival rate and less that are water-soluble, eg, carotene and Probably ascorbic acid; Liener, 1958). Ascorbic and success in producing colonies were also panthotenic acids (essential nutrients) in associated with the lower biomass of these mass to several fact were found to be very unstable under queens. Body is related storage conditions (Nielsen, 1956; Haydak factors including fat content (Alford, 1969a; 1963; Hagedorn and Burger, 1968). Holm, 1972). During hibernation queens use most of the reserves (fat and glyco- the content of the DFP may Finally, lipid have accumulated in also have been affected during the drying gen) they previously the fat body. Alford (1969a,b) found that process. Polyunsaturated fatty acids, such the amount of fat in young B terrestris as linoleic and linolenic acids (present in queens about to go into hibernation was many plant species; Stanley and Linskens, on average 34% of their dry weight. Con- 1974) are essential nutrients for most that pollen is important for the and may not be synthesized by them (Dadd, sidering of the fat 1973). The membrane of pollen grains loses development body (Maurizio, 1950; Duchateau, it is its integrity during the drying-ageing pro- unpublished data), that DFP caused an underdevel- cess. As a consequence, lipids are de-ester- possible ified and lysophospholipids and free fatty opment of the fat body of young queens in acids accumulate. An excessive leakage of our experiment. the internal contents occurs with rehydration In conclusion, fairly good colonies may (Crowe et al, 1989; Bilsen and Hoekstra, be reared with DFP (if the temperature dur- 1993; Bilsen et al, 1994a,b). (It has been ing the drying process is not too high). But to observed that bumble bees moisten the ensure that young queens are of high qual- pollen before ingesting it, or when preparing ity (adequate size, biomass, survival and it to feed the larvae. For obvious reasons capability to produce good colonies) it is the amount of moisture was always larger in essential that the pollen used to feed the the containers of DFP than in the FFP. colonies during the queens larval develop- Sometimes even a change in the colour of ment is fresh (or FFP). ACKNOWLEDGMENTS der Überwinterung verbrachten die überle- benden Königinnen in einem Flugkäfig. Danach wurden um neue Völker zu star- We would like to thank W Punt for identifying the sie, pollen, B Kraus for critically reading the ten, bei 28 °C und 60% relativer Luftfeuch- manuscript and writing the German summary, A tigkeit in Holzkisten gehalten. Zur Bestim- van Doom for giving valuable suggestions, I van mung der Größe der Königinnen wurde ihre der Tweel for helping with the statistics, M Schim- Biomasse (Trockengewicht) sowie die mel for preparing the figures and S McNab for Länge der radialen Fügelzelle bestimmt. advice. we also acknowl- giving linguistic Finally, Getrockneter Pollen wurde konsu- edge the financial support provided by Koppert- weniger Biological Systems and CNPq (Conselho miert als frischer Pollen (nicht signifikant, Nacional de Desenvolvimento Cientifico e Tec- Abb 1). Eine mögliche Ursache hierfür nológico, Brazil) and the two anonymous refer- könnte die durch den Trocknungsprozeß ees who helped to improve this manuscript. bedingte Veränderung einiger Substanzen sein, die vor dem Trocknungsprozeß den Pollen attraktiver für Hummeln machen. Zusammenfassung — Vergleich der Aus- Eine andere Ursache könnte sein, daß wirkung von zwei handelsüblichen Pol- getrockneter Pollen weniger schmackhaft lensorten auf Kolonieentwicklung und für Hummeln ist. Die Entwicklung der Kolo- Königinnenproduktion bei der Hummel nie sowie die Anzahl und Biomasse der Bombus terrestris (L) (Hymenoptera, Api- Arbeiterinnen und Drohnen wurde durch dae). Verschiedene Pollen haben unter- den getrockneten Pollen nicht beeinflußt schiedlichen Nährwert für Bienen. Hierfür (Tabelle II und III; Abb 2). Königinnen, die gibt es zwei Gründe: 1. Die chemische mit getrocknetem Pollen aufgezogen wur- Zusammensetzung des Pollens ist je nach den, waren jedoch kleiner und hatten eine Pflanzenart unterschiedlich. 2. Faktoren wie geringere Biomasse und eine höhere Sterb- Trocknung, Alterung und Lagerungsbedin- lichkeit vor der Überwinterung als mit fri- gungen beeinflussen den Nährwert. Es exi- schem Pollen aufgezogene Königinnen. Sie stieren in der Literatur keine Angaben über begannen später mit der Eiablage und pro- den Effekt von getrocknetem Pollen auf die duzierten eine geringere Anzahl Individuen Entwicklung des Hummelvolkes und die (Tabelle III und IV; Abb 3, 4 und 5). Mit Produktion von Hummelköniginnen. Zwei getrocknetem Brassica Pollen gefütterte im Handel erhältliche Pollensorten wurden Völker produzierten ebenfalls signifikant in diesem Experiment verwendet. Die Pol- kleinere Königinnen als mit frischem Bras- lensorten stammten von unterschiedlichen sica Pollen gefütterte Völker. Dies kann Pflanzen (Tabelle I) und wurden vor Ver- möglicherweise durch eine geringere Menge suchsbeginn unterschiedlich gelagert. Der an Proteinen in getrocknetem Pollen bedingt Pollen wurde entweder frisch oder getrock- sein. Der getrocknete Pollen bestand net verwendet. Getrockneter oder frischer hauptsächlich aus proteinreichem Brassi- Pollen wurde sofort verwendet oder tiefge- capollen. Der Trocknungsprozeß könnte froren. Beide Pollensorten waren auf Grund jedoch die Proteine beeinflussen. Viele wich- ihres hohen Proteingehaltes und ihres gerin- tige Aminosäuren reagieren besonders emp- gen Stärkegehaltes als qualitativ hochwer- findlich auf Erwärmung. Erwärmung könnte tig anzusehen (Tabelle I). Wir Kontrollier- die Molekularstruktur eines Proteins und ten die Völker täglich, um Daten über die damit seine Verdaubarkeit durch Enzyme Entwicklung der Individuen und der Völker verändern. Während des Trocknungspro- zu gewinnen. Den geschlüpften Königinnen zesses und der Lagerung könnten auch wurde Gelegenheit zur Paarung und Über- Vitamine oder Fettsäuren (wichtige Nähr- winterung geboten. Die ersten 10 Tage nach stoffe für einen Großteil der Insekten) geschädigt werden. Durch den Verlust expérience. Leur composition différait und/oder die Schädigung wichtiger Sub- (tableau I), ainsi que leur mode de conser- stanzen steht den Hummeln bei der Fütte- vation : à l’état frais + un an au congélateur rung mit getrocknetem Pollen möglicher- pour le premier, séché + deux ans au weise eine geringere Menge an Nährstoffen congélateur pour le second. Néanmoins, en zur Verfügung. Königinnenlarven scheinen raison de leur qualité probablement bonne besonders empfindlich auf die geringere (pas d’amidon et/ou forte teneur en pro- Qualität des getrockneten Pollens zu rea- téines) et de la nécessité d’obtenir des infor- gieren. Die geringere Biomasse der Köni- mations sur les régimes de pollen conve- ginnen steht möglicherweise im Zusam- nant aux bourdons, nous les avons utilisés menhang mit der Größe ihres Fettkörpers. pour nourrir des colonies et comparer leurs Während der Überwinterung benötigen effets éventuels sur les insectes. Des don- Königinnen ihre im Fettkörper zuvor ange- nées sur les individus et le développement reicherten Reserven (Fett und Glykogen). des colonies ont été relevées journellement. Pollen hat eine wichtige Funktion bei der Les jeunes reines produites ont été libres Entwicklung dieses Fettkörpers. Die Fütte- de s’accoupler et d’hiberner. Après l’hiber- rung mit getrocknetem Pollen könnte zur nation les survivantes ont été placées dans Unterentwicklung des Fettkörpers führen. des cages de vol pendant 10 jours, puis Aus den Ergebnissen unserer Untersuchung dans des boîtes en bois en conditions folgern wir, daß Durchschnittsvölker mit contrôlées de température (28 °C) et d’humi- getrocknetem und eingefrorenem Pollen dité relative (60 %) pour la fondation des aufgezogen werden können. Zur Zucht qua- colonies. La taille des reines a été détermi- litativ hochwertiger junger Königinnen sollte née en mesurant leur biomasse (poids sec) jedoch frischer (oder eingefrorener frischer) et la longueur de la cellule radiale alaire. Le Pollen verwendet werden. pollen séché a été moins consommé que le pollen frais, mais la différence n’est pas Bombus terrestris / Pollen / Nährwert / significative (fig 1). La raison pourrait en Königinnenzucht être que le processus de séchage a modifié certaines des substances phagostimulantes et/ou une réduction de l’appétibilité du pol- Résumé — Étude comparative de l’action len. Le développement de la colonie de deux types de pollen du commerce (tableau II), le nombre d’ouvrières et de sur le développement de la colonie et mâles (tableau III, fig 2) et leur biomasse sur la production de reines chez le bour- (tableau III) n’ont pas été affectés par la don Bombus terrestris (L) (Hymenoptera, consommation de pollen séché. Pourtant Apidae). La valeur nutritive des pollens pour les reines élevées avec du pollen séché les insectes varie d’un pollen à l’autre, étaient plus petites (fig 3), avaient une bio- d’abord parce que leur composition chi- masse inférieure (tableau III) et une morta- mique est fonction de l’espèce botanique, lité avant l’hibernation plus élevée (fig 4, ensuite du fait de l’action de divers facteurs tableau IV), commençaient à pondre plus tels que le séchage, le vieillissement et/ou tardivement (fig 5) et produisaient de plus les conditions de conservation. Il n’existe petites colonies (tableau IV). Le pollen séché pas dans la littérature d’information sur les de Brassica a également donné des reines conséquences de l’utilisation de pollen significativement plus petites que celles éle- séché sur le développement de la colonie et vées avec du pollen de Brassica frais sur la production de reines chez les bour- (Duchâteau, comm pers). Cela peut être dons. Deux types de pollen disponibles dans mis en relation avec la quantité de protéines le commerce ont été utilisés dans cette peut-être plus réduite dans le pollen séché. Bien que le pollen séché ait été constitué Auclair JL, Jamieson CA (1948) A qualitative analysis of de de riche amino acids in pollen collected by bees. Science principalement pollen Brassica, 108, 357-358 en protéines, il est possible que le séchage Baker HG, Baker I (1979) Starch in angiosperm pollen ait affecté les protéines. De nombreux grains and its evolutionary significance. Am J Bot acides aminés essentiels sont sensibles à la 66, 591-600 chaleur. Des températures élevées peuvent Bieberdorf FW, Gross AL, Weichlein R (1961) Free modifier la structure moléculaire des pro- amino acid content of pollen. Ann Allergy 19, 867-876 téines et augmenter la résistance aux Bilsen DGJL van, Hoekstra FA (1993) Decreased mem- Les vitamines brane integrity in aging Typha latifolia L pollen. Plant enzymes digestifs. également Physiol 101, 675-682 peuvent être abimées ou détruites par le Bilsen DGJL van, Hoekstra FA, Crowe LM, Crowe JH séchage et la conservation. Certains lipides (1994a) Altered phase behavior in membranes of (nutriments essentiels pour la plupart des aging dry pollen may cause imbibitional leakage. insectes) peuvent avoir subi une dégrada- Plant Physiol 104, 1193-1199 tion au cours du séchage. En conséquence, Bilsen DGJL van, Roekel TV, Hoekstra FA (1994b) suite à la et/ou la détérioration de sub- Declining viability and lipid degradation during pollen perte storage. Sex Plant Reprod 7, 303-310 stances, les nutriments pour la disponibles Chambers S (1990) Feeding pollen to honey bees for digestion et l’absorption peuvent être dis- colony development. Bee World 71, 35-37 ponibles en moins grande quantité. Il semble Crowe JH, Hoekstra FA, Crowe LM (1989) Membrane que les larves de reines soient particulière- phase transitions are responsible for imbibitional dam- ment sensibles à la réduction des nutri- age in dry pollen. Proc Natl Acad Sci USA 86, 520-523 ments. La biomasse plus faible des reines Dadd RH (1973) nutrition: current developments and metabolic implications. Annu Rev Entomol 18, est liée à la teneur en Au cours de graisse. 381-420 l’hibernation les reines utilisent la majeure Dietz A, Haydak MH (1965) Causes of nutrient defi- partie de leurs réserves (graisse et glyco- ciency in stored pollen for the development of newly gène) accumulées auparavant dans le corps emerged honey bees. Proc 20 Int Beekeeping Congr, gras. Le pollen ayant un rôle important dans Bucharest, Apimondia, 222-225 le développement du corps gras, il se peut Duchateau MJ (1985) Analyse einiger Methoden der Hummelzüchtung. Apidologie 16, 225-227 que le pollen séché donné en nourrisse- Duchateau MJ, Velthuis HHW and ment entraîne un du (1988) Development sous-développement reproductive strategies in Bombus terrestris colonies. corps gras. En conclusion, il est possible Behaviour 107, 186-207 d’élever de belles colonies de bourdons avec Duchateau MJ, Velthuis HHW (1989) Ovarian develop- du pollen séché, mais pour s’assurer de la ment and egg laying in workers of Bombus terrestris. bonne qualité des jeunes reines, il est essen- Entomol Exp Appl 51, 199-213 tiel que le pollen utilisé pour nourrir les colo- Falconer DS (1981) Introduction to Quantitative Genet- ics. 2nd London nies pendant le développement des larves ed, Longman, Free JB, Butler CG (1959) . Collins, London de reines soit frais ou congelé à l’état frais. Groot AP de (1953) Protein and amino acids require- ments of the honeybee (Apis mellifica L). Physiol Bombus terrestris / pollen / valeur nutri- Comp Oecol 3, 197-285 tive reines / élevage Hagedorn HH, Moeller FE (1968) Effect of the age of pollen used in pollen supplements on their nutritive value for the honey bee. I. Effect on thoracic weight, REFERENCES development of hypopharyngeal glands, and brood rearing. J Apic Res 7, 89-95 Hagedorn HH, Burger M (1968) Effect of the age of Alford DV (1969a) A study of the hibernation of bumble pollen used in pollen supplements on their nutritive bees (Hymenoptera: Bombidae) in Southern Eng- value for the honey bee. II. Effect of vitamin content land. J Anim Ecol 38, 149-170 of pollens. J Apic Res 7, 97-101 Alford DV (1969b) Studies on the fat-body of adult bum- Hasselrot T (1952) A new method for starting bumble ble bees. J Apic Res 8, 37-48 bee colonies. Agron J 44, 218-219 Haydak MH (1961) Influence of storage on the nutritive Nielsen N (1956) The vitamin content of pollen after value of pollens for newly emerged honey bees. Am storing. Acta Chem Scand 2, 332-333 Bee J 10, 354-355 Owen, RE (1988) Body size variation and optimal body Haydak MH (1963) Influence of storage on the nutritive size of bumble bee queens (Hymenoptera: Apidae). value of pollen for brood rearing by honey bees. J Can Entomol 120, 19-27 Apic Res 2, 105-107 Plowright RC, Jay SC (1966) Rearing bumble bee Haydak MH (1970) Honey bee nutrition. Annu Rev Ento- colonies in captivity. J Apic Res 5, 155-165 mol 15, 143-156 Pomeroy N, Plowright RC (1980) Maintenance of bum- Herbert Jr EW (1992) Honey bee nutrition. In: The Hive ble bee colonies in observation hives (Hymenoptera: and the Honey Bee (JM Graham, ed), Dadant & Apidae). Can Entomol 112, 321-326 Sons, Hamilton Regali A, Rasmont P (1995) Nouvelles méthodes de test Holm SN Problems of the domestication of bum- (1966) pour l’évaluation du regime alimentaire chez ble bees. Bee World 47, 179-186 des colonies orphelines de Bombus Holm SN (1972) Weight and life length of hibernating terrestris L (Hymenoptera, Apidae). Apidologie 26, bumble bee queens (Hymenoptera: Bombidae) under 273-281 controlled conditions. Entomol Scand 3, 313-320 Robinson FA, Nation JL (1968) Substances that attract Knox DA, Shimanuki H, Herbert EW (1971) Diet and caged honey bee colonies to consume pollen sup- the longevity of adult honey bees. J Econ Entomol plements and substitutes. J Apic Res 7, 83-88 64, 1415-1416 Röseler PF (1985) A technique for year-round rearing of Lepage M, Boch R (1968) Pollen lipids attractive to Bombus terrestris (Apidae, ) colonies in cap- honey bees. Lipids 3, 530-534 tivity. Apidologie 16, 165-170 Levin MD, Haydak MH (1957) Comparative value of dif- Schmidt JO (1985) Phagostimulants in pollen. J Apic ferent pollens in the nutrition of Osmia lignaria. Bee Res 24, 107-114 World 38, 221-226 Sokal RR, Rohlf FJ (1981) Biometry. 2nd ed. WH Free- Liener IF (1958) Effect of heating on plant proteins. In: man and Co, San Francisco Processed Plant Protein Foodstuffs (AM Altschul, Standifer LN (1966) Some constituents of ed) Academic Press, New York lipids pollens collected by honeybees. J Apic Res 5, 93-98 Maurizio A (1950) The influence of pollen feeding and Standifer LN (1967) A comparison of protein quality of brood rearing on the length of life and physiological pollens for growth-stimulation of the condition of the honey bee - preliminary report. Bee hypopharyn- World 31, 9-12 geal glands and longevity of honey bees, Apis mel- lifera L. Insectes Soc 14, 415-425 Maurizio A (1951) Studies on the influence of pollen Stanley RG, Linskens HF Pollen: Bio- feeding and brood rearing on the length of life and the (1974) Biology, physiological condition of bees. The Norfolk Bee- chemistry, Management. Springer-Verlag, Berlin keepers’ Association 101, 6-8 Svoboda J (1940) The value of pollen as a food for bees. Maurizio, A (1958) Einfluss der Trocknungsmethode auf Bee World 21, 105-107 die biologische Wirksamkeit des Pollens für Bienen. Todd FE, Bretherick O (1942) The composition of pol- Z Bienenforsch 4, 59-62 lens. J Econ Entomol 35, 312-316 Maurizio, A (1960) Bienenbotanik. In: Biene und Bienen- Vivino AE, Palmer LS (1944) The chemical composition zucht (A Büdel, E Herold, eds) Ehrenwirth Verlag, and nutritional value of pollens collected by bees. Munich Arch Biochem 4, 129-136