Virtual Beekeeping Gallery

Galerie Apicole Virtuelle

Galeria Apicola Virtual

Virtuelle Imkerei Galerie

FACULTY OF AGRICULTURAL, FOOD AND ENVIRONMENTAL QUALITY SCIENCES (FAFEQS)
76100 REHOVOT, ISRAEL

Report for 1976-1997 - Page n°3

TABLE OF CONTENTS

1. Introduction
2. Staff
3. Teaching
4. Graduate research students
5. Undergraduate research trainees
6. Visiting research students and postdoctorals
7. Visiting scientists
8. Research support
9. Research facilities
10. Collaborators at the Hebrew University of Jerusalem and other Scientific Institutions
11. Domestic and international activities
12. Publications
13. Research reports
1. Differentiation of female honeybee larvae to queens
2. Swarming; construction of queen swarming cups and cells
3. Reproduction; mating; pheromones of queens and drones
4. Thermoregulation at high ambient temperatures
5. Defensive behavior of bee colonies; Alarm pheromones of workers and queens
6. Exocrine glands: structure and secretory products
7. Proteins of body surface and compartments, haemolymph and vitellogenins
8. Attraction of honeybees to flowers; pollination
9. Control of Varroa jacobsoni

13. Research Reports
Abstracts are identifiable from the alphanumeric symbols indicating Sections 4,5,12.

We developed a method for rearing worker and queen bees from larvae to adults in the laboratory in absence of nurse bees. We used it in the study of the differentiation of female larvae, and of their nutritional requirements.

Best results were obtained when larvae were reared on queen larval food at 35° C and 86% R.H. and then transferring, the 5^th stage larvae to queen cells at 30% R.H. From larvae reared in the laboratory emerged 78% adults; 50% queens, 16% workers and 12% intermediates. There were no significant differences between the number of adults emerging from larvae reared on fresh larval food from queen cells containing 1- to 5-day-old larvae and those reared only on larval food from queen cells containing 3-day-old larvae. However, from larvae reared on queen larval food stored at -20° C emerged either workers or intercastes, but not queens.

The growth rate of larvae up to 5^th day was faster on worker larval food than on queen larval food either fresh or stored at -20° C. The slowest growth rate was recorded in all larval instars reared on queen larval food stored at -20° C. The fastest growth rate was found in 5- and 6-day-old larvae reared in fresh queen larval food.

Larvae reared on dialyzed queen larval food only or supplemented with various substances died within 48 hr.

Larvae reared on worker larval food collected from worker cells containing 3-day-old larvae reached the prepupal stage, but neither pupated on soft paper nor in natural queen cells. Addition of fructose and glucose (16% of each) to worker larval food caused pupation, but no adults emerged. Supplementing worker larval food with honey (16%) and pollen (2%) produced 62% pupae of which 38% adults emerged. 4a(5)

We studied the acceptance and rearing of queen larvae in queenright colonies with non-restricted queens. Larvae grafted in queen cups were reared by the colonies, regardless of the queen's presence or absence in the brood-rearing part of the hive. The experiments were carried out during July and August when no swarming occurs. 13a(25)

We examined the effect of topical treatment with Juvenile Hormone-1 (JH-1) on the induction of "Queenliness" in undifferentiated female honeybee larvae reared in the laboratory on basic or modified diets of Worker Jelly (WJ) or of Stored Royal Jelly (SRJ).

Female larvae reared on a modified diet of WJ with 40 mg Fructose and 40 mg Glucose added to each gram WJ (WJ-40) emerged as workers only.

A dose-response effect of JH-1 on the differentiation of larvae reared on WJ-40 to adult queens was achieved: larvae treated with 0.01 m g/larva of JH-1 emerged as workers only, 0.1 m g- as workers, intercastes and queens; 1.0 m g-workers and queens; 10 m g-queens only; High levels of JH-1 (100 m g/larva) caused death.

Stored Royal Jelly (SRJ) for 4 years at -18° C, when used as a non-supplemented basic food enabled female larvae to differentiate and emerge as workers only. SRJ supplemented with 40 mg F and 40 mg G/gr food (SRJ-40) resulted in the emergence of adult workers and intercastes, but not of queens. About 40% of larvae treated with 1 m g JH-1 and reared on SRJ-40 emerged as queens. JH-1 treatment of larvae restored their capabilities to differentiate and emerge as queens, even though they were fed with SRJ.

An adverse effect of JH-1 was observed when only 17% of larvae which were treated with 1 m g JH-1 (each) and reared on a Fresh Royal Jelly (FRJ) diet emerged as queens. It is concluded that FRJ does not lend itself to be a reliable basic medium for the study of the effect of various compounds on female dimorphism in honeybees.

The fate of a presumptive adult female can be predicted by the quantity of food ingested until the last larval instar: before cocoon spinning the weight of larvae that differentiated and emerged as queens was higher (213-255 mg) than that of intercastes (182-233 mg) or of workers (179-214 mg). Sugars (F and G), act as phagostimulants in the induction of "Queenliness." 4b(1), 13b(1,13,17).

We studied the composition of sugars and their level in fresh royal jelly (RJ), worker jelly (WJ) and stored royal jelly (SRJ), as well as the effect of SRJ soluble sugars on the differentiation of female larvae. The sugar level of RJ removed from cells with 1- to 3-day-old larvae (RJ 1-3) was about 4 times higher than that of WJ of corresponding age (WJ 1-3) (12.4 vrs 3.1%) when calculated on the basis of fresh matter. The ratio fructose: glucose (F:G) in RJ 1-3 (1.43) and in WJ 1-3 (1.37) was almost equal. The crystals formed in SRJ and analyzed by GLC, as well as by optical, polarizing and scanning microscope, were composed of fructose (F), glucose (G) and sucrose (S). The level of F was 83.7% of the total sugar content of crystals. Female honeybee larvae reared on SRJ supplemented with 198 mg F + 38 mg G/g food, an amount similar to that of SRJ crystals, resulted in the emergence of adults as: queens (40%), intercastes (23%) and workers (37%). The importance of availability of soluble sugars versus solid crystals in the larval food on the induction of "queenliness" is discussed. 4b(1), 13b(20)

The fluctuations of Juvenile Hormone and ecdysone during the postembryonic development of queen and worker honeybees were established by radio-immuno assays. In both cases the Juvenile Hormone levels were high during the prepupal stage, whereas increased levels of ecdysones were present prior to the imaginal molt. Only in queen larvae a high increase in Juvenile Hormone levels (mainly JH-1) was recorded between the end of the 3^rd and the beginning of the 5^th larval instar, when caste induction and ovarian development occur. The high level of ecdysones found only in queen prepupae and pupae may explain the enhanced development of queen pupae, as compared to that of worker pupae.13b(3)

In a series of experiments conducted during non-swarming periods with queenright colonies we have tried to elucidate the following questions: Is cell construction inhibited by volatile substances emitted by the queen or do some other secretions, presumably secreted by the tarsal glands of queen legs, contribute to this inhibition? Does restriction of queen movements affect this inhibition? Are queen cups constructed along comb edges or upon any existing or freshly exposed surface of the comb to which the queen had no previous access? Does overcrowding of nurse bees on the comb containing the restricted queen interfere with the even distribution of the inhibitory substances upon it?

To answer these questions we have exposed fresh surfaces in combs by either razing the edges, cutting out gulfs, or fenestrating the center of a comb in queenright and queenless compartments of the same colony.

For this purpose six experiments, using 31 queenright colonies, were carried out at our apiary between May and November, and the results are summarized below:

Queen cup and cell construction prior to swarming are controlled by inhibitory substances secreted by freely moving and non-restricted young (up to 1 year old) queens. Restriction of the queen's movements or her confinement within a part of a comb induces construction of queen cups and cells upon any comb in the queenless part of the colony.

Exposure of fresh surfaces upon combs, either by destroying the existing cups or by razing comb edges, cutting out gulfs, or by fenestrating, induced construction of queen cups in these sites or in their vicinity.

There was a partial inhibition of queen cup and cell construction in the comb area with freely moving queens, if no overcrowding of bee workers was induced.

There was a preference for constructing queen swarming cups on the edges of combs containing non sealed brood.

Volatile substances secreted by queens did not inhibit the cup and cell construction on combs where the queen could not be physically present, or on the comb where she was restricted and which was overcrowded with bee workers.

In queenless colonies the majority of emergency queen cells are constructed at the center of a brood area in a comb upon a basis of a worker cell containing an egg or a larva. In queenright colonies, there was a preference for building queen cells along comb edges, even in the presence of a mutilated queen. Queen cups were also constructed in low population density areas of queenless compartments of a colony.

Exposure of fresh surfaces by razing the edges of combs containing either brood or honey induced construction of queen cups in strong colonies, but not in weak ones.

No construction of queen cups and cells occurred in the control colonies headed by queens that were free to move upon all the combs in a hive during the experimental periods.

We tried synthetic compounds to inhibit queen swarming cell construction during swarming period, both in: queenless colonies and small groups of workers in cages, as well as in queenright colonies during the swarming period.

However, phenylacetic, phenylpropionic, p-hydroxybenzoic, azelaic, and sebacic acids that were reported as components of "queen substance" neither had any inhibitory effect on queen cells construction, nor were detected by TLC in an extract of whole, laying queens.

The trans-9-oxo-dec-2-enoic acid inhibited the construction of queen emergency cells in queenless colonies. Its inhibitory effect was more pronounced when the acid was applied on filter paper strips rather than mixed with sugar candy. But, the trans-9-oxo-dec-2-enoic acid did not inhibit queen swarming cell construction in queenright colonies during the swarming period.

Although the extract of one year-old queens inhibited queen emergency cell construction in queenless groups of worker bees in cages, it did not inhibit the construction in queenright colonies during the swarming period.

The volatile substances secreted from the queen's body surface neither had an inhibitory effect on queen emergency cell construction, nor a synergistic effect together with the trans-9-oxo-dec-2-enoic acid and with queen extracts in queenless colonies.

We have demonstrated that the construction of queen cups that proceeds to queen swarming cells is a general phenomenon occurring in all colonies during the spring. Only a part of the queen cups is further constructed to queen swarming cells.

The factors affecting queen swarming cell construction during the swarming period are as follows: high density of bee workers, inadequate ventilation and overheating of the brood nest. . 4a(6), 13a(39)

We studied the role of worker population density of queenright honeybee colonies, and that of the queen bee pheromonal secretions on the induction and inhibition of queen swarming cup construction during swarming and non-swarming seasons. Construction of queen cups was induced experimentally in overcrowded queenright colonies during winter, which is a non-swarming season. This construction was induced by a high population density of bee workers: above a threshold of 2.3 bee workers/ml there was a relationship between the number of cups constructed and the worker population density of a colony. During the swarming season a relationship was established between the free volume of a hive (population density) and the number of queen cups constructed: 1.5 cups in a colony that occupied 80,960 ml, compared with 77 cups in a colony hived within a volume of 20,240 ml. Observations of the queen's movements upon combs in colonies of high and normal worker population densities showed that in an overcrowded colony the queen bee was almost absent from the bottom edges of the comb, where queen swarming cups and cells are constructed.

The tarsal glands of queens are located in the fifth tarsomere and the glandular oily secretion is deposited by the foot-pads upon the comb surface. The rate of secretion by the queen's tarsal glands was about 13 times higher that by those of the workers. A bioassay for testing the inhibitory effects of the queen's glandular extracts on the construction of queen cups was developed. It was based on increasing worker bee population densities, and can be used effectively throughout the year in a subtropical climate.

The application of tarsal and mandibular glands secretions to comb bottom edges in overcrowded colonies (bioassay) caused the inhibition of queen swarming cup construction. Neither of these two secretions affected the construction of these cups when applied separately. We presume that due to colony overcrowding the queen bee is unable to deposit the non-volatile secretions from her tarsal glands along the comb edges and that the deficiency of the footprint pheromone triggers the construction of swarming cups along the non-inhibited areas. 4a(2), 13b(9)

Extracts of laying, mated queens or synthetic 9-oxo-trans-2-decenoic acid (9-ODA) partially inhibited queen replacement following the removal of the queen. Synthetic 9-hydroxy-trans-2-decenoic acid (9-HDA) was ineffective and not synergistic with 9-)ODA. During the swarming season, a continuous treatment with 9-ODA did not reduce the number of queen cups and queen swarming cells built by queenright colonies. 13(2)

The control of the construction of queen cups and swarming cells is a complex event which depends on both ecological (temperature, availability of nectar) and biological (density of workers inside the brood nest) factors. Other components include the chemical composition of both the mandibular (MG) and tarsal (TG) gland secretions of queen bees. The chemical composition of these secretions depends on queen's age and her physiological state. Due to the abundance and the complexity of pheromonal components secreted by the queen MG and TG, at this stage of our research we cannot offer yet a synthetic mixture of pheromonal components which might be applied to inhibit swarm cell construction in a colony headed by an old queen or to a congested colony. In Israel, by applying the results from our previous studies, we have minimized the production of new queens, and thus the incidence of swarming by (1) annually requeening in September-October so when swarming starts in March colonies are headed by 6- to 7-month old queens; (2) increasing the volume of hives by adding supers in excess; (3) raising 7 to 8 brood frames from the brood nest into the supers above queen excluders; (4) preventing overheating by providing additional openings to facilitate ventilation and (5) providing a constant water supply in the bee yard.

Perhaps future research on the chemical composition of mandibular and tarsal gland secretions will provide a better insight into the problem of swarming and will contribute to a useful control of the phenomenon. 13b(32)

During two swarming seasons we studied the effect of increased hive volume and ventilation, as well as that of colony overheating on the construction of queen swarming cups and cells, issuing of swarms and honey yields. Queen swarming cups were constructed by all colonies, but their number varied from one year to another. Only 1 cup, out of 10 or 20, was transformed to a queen swarming cell. Fewer queen cups were transformed to queen swarming cells by colonies maintained inside large-volume hives (low population density) than by congested colonies.

Increased hive volume and ventilation exerted an inhibitory effect on the construction of queen swarming cups and cells and on issuing of swarms. Overheating of congested bee colonies significantly affected the early construction of queen swarming cups and cells and the incidence of swarming. Although insignificant at P<0.05, the honey yields obtained from colonies in high-volume, ventilated hives were higher than those from control colonies. 4a(6),13b(7)

In a study carried out in Israel, we found that brood areas, worker populations and honey yields were higher, and less queen swarming cells have been constructed in colonies headed by 7- to 10-month-old than in those headed by up to 20-month-old queens during the nectar flow season. There was a significant effect of seasonal conditions on the number of queen cups constructed but not on that of queen swarming cells. Neither brood area nor worker population affected construction of the two kinds of cells in non-congested colonies. The queen’s age and the season affect honey yields through the population level of workers. 4b(12), 13b(39)

The secretion rate from the tarsal glands was 100%, 11% and 33% by young (6 months), old (18-24 months) and 3-day-old queens, respectively.

Gas chromatography using packed (5% OV-101, 2% OV-101, SE-30) and capillary (FFAP and SE-30) columns revealed characteristic components for young and old queens, respectively. The tarsal secretions were separated into up to 160 components using capillary columns. 4b(14)

The rate of secretion from the tarsal glands was as follows: 1.005 mg/hr for 6 month-old queens; 0.64 mg/hr for 2 year-old queens; 0.0718 mg/hr for drones.

The footprint substances revealed no behavioral effects in the following experiments: Worker foot-print substances, orientation to hive entrance, attraction to food source, survival of isolated workers, acceptance of grafted larvae in queen cups, drone mating flight. Drone foot-print substance: drone mating flight. Queen foot-print substance: survival of isolated workers, drone mating flight.

Although no effects of footprint substances on the behavior of workers and drones were found, we do not preclude the possibility that they may act as synergistic and/or slow release substances for other pheromones, in addition to being an adhesive substance for bees crawling on smooth surfaces.

Gas Chromatography-Mass Spectrometry: Characterization of the foot-print substance showed the presence of alkanes, alkenes, alcohols, organic acids, ethers, esters and aldehydes. 4b(9)

For detailed abstract see section: The structure and secretory products of exocrine glands

Multiplication of honeybee colonies and the matings of queens and drones occur in Israel mostly during the spring. Accordingly, commercial queen rearing is carried out during this season and seldom in the fall. We examined the possibility of extending the season of queen rearing in Israel and the mating success of queens throughout the year. We were carried out observations during all seasons to establish:

1. The relationship between queen and worker bees during the mating flight days;
2. The total number of queen mating flight days, the frequency of flight and the duration of each flight;
3. At what time of the day queens' and drones' flights occur;
4. The effect of climatic factors on mating flights and the onset of egg-laying;
5. The effect of CO₂ anesthesia on queens' mating flights and on the onset of egg-laying;
6. The success of mating and the effect of the mating season on the build-up of colony population and honey yields.

Excited worker bees run upon combs, direct the virgin queen towards the hive entrance and fly outside the hive for about 60 min. before the queen leaves for her mating flight. The workers expose their Nassanof glands about 50 min. after they have started directing the queen. Shortly before she flies out from the hive, the queen exposes her Renner gland.

A delay in queen and drone flights was caused by cloudiness above 7 octaves and by wind velocity above 3.9 m/sec. The duration of the queens' flights was reduced at temperatures between 15-20° C and at wind velocity between 2.6-3.9 m/sec.

The peaks of flight activity hours of queens and drones overlapped throughout the year, however the range varied according to the season (13.00± 0.09 in January and 15.30± 0.15 in June). The hours of flight activity were affected by the season rather than by the monthly fluctuations of climatic factors. It seems that the mating flight time is controlled by the diurnal rhythm (20.19± 0.12 hours from sunset to flight) and/or the azimuth of the sun (196.31° -266.01° ) and/or the angle of the zenith (36.04° -57.89° ).

Following CO₂ treatment of emerging queens the onset of mating flights was delayed (2 days), the flights occurred later (44 min.) and the onset of egg-laying was enhanced (2 days).

The queens successfully mated during most of the months (100% success), except for September's 70% success due to lack of drones.

Colonies headed by 2- to 10- month-old queens yielded similar honey crops. This finding contradicts the prevailing view that successful mating can be achieved only during the spring season and in September and that colonies headed by 7-month-old queens produce less honey than those headed by 7- to 24-month-old queens.

This study demonstrates that successful queen rearing and mating can be achieved in the coastal plain of Israel almost throughout the year.4b(6), 13b(18)

Virgin queen bees aged from three to eight days were experimentally introduced into mating nuclei. The acceptance of aged virgins did not approach the success of those emerging from queen cell in the nuclei used as controls. 13a(32)

Mating success of virgin queen honeybees of different ages

The mating success rate of virgin queen honeybees aged from three to eight days was determined. In general, the success rate was very similar between queens aged from 3 to 6 days and the control. The possibility that mating success declines with age is discussed, even though virgin queens stored for more than 60 days mated successfully. 13a(33)

Juvenile hormone speeds up the development of the ovaries and the rate of egg production. When young queens were topically treated with 0.2 m l of juvenile hormone, oocytes and trophocytes were present in their ovaries by the fifth day, whereas control queens' ovarioles were totally empty.

Laying queens, treated similarly during the winter, considerably increased the rate of oviposition as compared to that of the controls. External factors, such as photoperiod and the act of mating were found to influence ovigenesis. These factors may influence the corpora allata to increase juvenile hormone secretion, thereby accelerating the process of ovigenesis.

Queens treated with juvenile hormone attracted workers more strongly. The increased attractiveness of juvenile hormone-treated queens seems to be the result of their production of pheromones attracting bee workers. 4a(7)

We developed a method for identifying sugars based osazone precipitation of reducing sugars and on differential resistance of sucrose and trehalose to acid hydrolysis. Examination of haemolymph of queens and drones, the spermathecae of virgin and mated queens and the seminal vesicles showed the presence of glucose, fructose, trehalose and one unidentified sugar. A test for trehalase in the above samples showed that its activity is similar in haemolymph of males and females. High trehalase activity was found in the spermathecae. 13a(20)

The degree of attractiveness to worker bees of queens of the following ages was: 0-1 day-old: none; 2- and 4 day-old: medium; 5 and 6 day-old, as well as 16 and 18 month-old: high.

The observed attractiveness of queens' body parts to workers was: abdomen>head>thorax. However, heads of living queens were 4.5 and 8.4 times more attractive than abdomens and thoraces, respectively, when calculated as the number of attending worker bees per surface of a body part.

Queens with tergites and sting chambers covered with paraffin were less attractive (16 and 22%, respectively) to worker bees than the intact controls.

The attractiveness of ethanol extracts of mandibular glands, tergal glands and abdomen tips was as follows: MG>TG=AT. Of mandibular gland extracts prepared from 1-, 2-, 3-, 5-day and 16-18 month-old queens, the highest attraction was observed in 5-day-old MG extracts only.

Of 24 chemical components bioassayed for attraction of workers both inside cages and in bee colonies, 9-oxo-trans-2-decenoic acid was the most attractive.

The results of the study show that the exocrine secretions from both mandibular and tergal glands of the queen bee contribute to the attraction of worker bees and that its intensity is age-dependent. 4a(13),13b(24)

The survival of virgin and mated queens in queenless and queenright reservoir colonies was studied during different seasons, for 5 years.

In queenless reservoir colonies, 80% of mated queens survived for 5 months. In queenright reservoir colonies with two laying queens, 78% of mated queens survived for 2 months. In queenless or queenright compartments within queenright reservoir colonies, 69 to 91% of virgin queens survived for 5 months. About 78% of mated queens stored in a super with a laying queen, above a queenright brood nest, survived for 2 months.

No seasonal differences in queen survival were observed, but an ample supply of young bees, protection from heat and cold, and syrup feeding after the main citrus flow ended, were necessary. 13b(10)

We studied the role of drone mandibular gland secretion in attracting flying drones, and the effect of drone age on secretory activity.

Extracts of mandibular glands and of cephalic tissues were applied to cotton lures attached below balloons tethered at 8-12 meters above the ground. Most flying drones were attracted to mandibular gland extracts, but a few flew to other cephalic tissues or to solvent controls.

Histological and electron microscope studies showed that the structure of the tiny (0.12 mm long) mandibular gland varied with age. In 0-3 day-old drones the secretory activity is evident from the abundant R.E.R. At 7 days the glands are fully developed and contained neurosecretory-like organelles indicating the secretion of an exocrine product. After 9 days the glands are no longer active and the product is stored in the gland lumen for further emission during the drones' mating flights. 4b(4), 13b(15)

Mature spermatozoa of drone honeybees obtained from seminal vesicles, were examined with an electron microscope by negative staining and by serial sectioning. The structure is described as illustrated in the micrographs. 13b(5)

Spermatozoa obtained from the vesiculae seminales were found to be motile. In these samples as well as in slightly diluted samples from the spermathecae, the spermatozoa were found to move in a swirling mass. In diluted samples from all sources, individual spermatozoa of circular and snake -like forms were observed. The pattern of motility and the shape of the spermatozoa were dependent upon the state of densitity and tonicity of the diluent and not the organ from which they were taken.

The pH of the spermathecal fluid and of the spermathecal glands was found to be 9 and 7, respectively. This negates the supposition that the immobilization of spermatozoa inside the spermatheca is due to the accumulation of the respiratory CO_2, and that the spermathecal glands activate the spermatozoa by neutralization of the acid medium in the spermatheca.

The honeybee spermatozoa were resistant to a wide range of temperatures (2-45° C) at the low and high temperatures of this range they were inactivated, but could be reactivated on transfer to room temperature. 13a(15)

We studied the survival of workers of spring generation at 45, 50, 55 and 60° C during 15, 30, 45 and 60 minutes exposures. The survival rate was established at the end of their exposure, and 24 hours later. Lethal effects of heat were evident immediately after a 30 minute exposure to 50° C. These effects could be noticed in the survivors 24 hours following exposure.

Effect of nutrition on heat resistance at 32° C and at 50° C was studied on one group of bees supplied with a 30% honey solution; on another- with water only, and on a third group which served as a control (no food). Their survival at 32° C after 12 hours exposure was 100%, 81% and 48%, respectively. However, an exposure duration of 45 minutes at 50° C resulted in a survival of 22% of control bees as compared to a total survival of those fed on honey solution. 4a(1), 13a(12)

General weight losses were established by weighing the workers before and after their exposure to heat. Temperatures and exposure durations were as follows: 36, 42, 48, 50, 52 and 55° C during 15, 30, 45 and 60 minutes. A regression expressing weight losses as a function of duration of exposure was adjusted for the temperature range from 36 to 50° C and within the intervals 15 to 60 minutes. Weight losses increased with temperature, but the slope of the curves decreased with exposure duration. No " transition-point" of cuticular lipids was found within the temperature range of the exposure. Mortality at temperatures above 50° C was marked by a temporary cessation of weight losses. Workers maintained during several days on a liquid and dry diets were exposed to 46° C during 15, 30, 45 and 60 minutes. Weight losses were superior in the group fed liquids.

We studied the group-effect on groups consisting of 1, 5, 15 and 50 individuals exposed to 46° C during 60 minutes. The weight loss of single isolated bees were 19.7% on the average as opposed to a mean of 9.4% per individual within a group of 50 bees. This effect is probably due to liquid soliciting as well as to creation of more favorable microclimate conditions.

Two groups of bees, one with ligatured the other with free wings, were exposed to 46° C during 60 minutes. Weight loss was more pronounced in freely ventilating bees.

Liquid exchange resulted in a loss of weight in the liquid-fed group, whereas bees maintained on a dry diet actually gained weight. However, when no contact was permitted, both groups lost weight. We examined water consumption within a temperature range of 28-50° C and exposure of 60 minutes. The water intake remained low within the temperature range of 28-45° C: 3.5-4.5 mg/hour/bee). However, it rapidly increased starting at 45° C (up to 13.3 mg/hour/bee). Under similar conditions of temperature and duration of exposure, water consumption was lower than weight losses. 4a(1), 13a(13)

We studied the effect of high temperatures on thermoregulation in a honeybee colony and on the behavior of the bees. Thermoregulation was limited to brood-nest. Its temperature did not exceed 37.6° C while the ambient one was 48° C. The activity of forager bees continued in spite of the extremely high temperatures (up to 48° C). During the middle of the day this activity decreased, but it was resumed as a result of transport of liquids when the ambient temperature was still 47° C.

The activities of ventilating bees at the hive entrance and the velocity of the air current were influenced by the fluctuation of the temperature of the brood-nest. The activity of water carriers was not affected directly by the ambient temperature but by that of the brood-nest. On the contrary, the activity of sucrose solution carriers depended upon external temperatures. A similar trend of daily activity was found both in ventilating bees and in water carriers, as opposed to the sucrose solution carriers. 4a(1), 13a(9)

We studied the effect of paint coating of beehives on the overheating of bee colonies during the summer. The lowest temperature values were recorded inside empty hives that were either whitewashed or placed in a partial shade. We recorded temperature using thermocouples at seven different points in each of the bee colonies hived in hives coated with aluminum paint, whitewashed or protected by shade. Temperature was almost constant in the brood nest, while in other parts of the hive it considerably fluctuated according to the coating or protection of a hive. Highest temperature values and activity of ventilating and water carrying workers were observed in bee colonies hived in aluminum-painted hives. Colonies in shaded or whitewashed hives produced more honey during hot summer months than those in aluminum painted and hives fully exposed to sun. 4a(1), 13a(11)

We devised an apparatus consisting of odor delivery, response and recording compartments, and a bioassay of airborne alarm pheromones, to measure the extrusion of the stinger, the stinging response (SR) of an isolated honeybee worker in controlled conditions. The recipient worker responds to a stimulus with abdominal contractions; their intensity is measured by an a manometer as the difference in air pressure inside a tube. The SR is specific for the alarm pheromone produced by the sting apparatus. The rise of the SR level corresponds to the number of donors or stingers, and hence to the concentration of pheromone used for stimulation. 4b(20), 13b(40)

The strongest alarm behavior response was recorded in bees aged 15-21 days. The highest amounts of alarm pheromones are also present in the same age group. No alarm or stinging reaction to stings of bees younger than 12 days could be recorded. 4b(20)

To evaluate the contribution of the setaceous membrane of the worker to the defensive behavior of a bee colony, we studied its fine structure and the effect of alarm pheromones present on the membrane surface on the reaction of guards. Groups of guards at the hive entrances displayed defense behavior when stimulated with volatiles from the setaceous membranes. Transmission electron microscope sections of the setaceous membrane did not reveal any structure characteristic of an exocrine gland, in contrast to the sting sheath or the Koschewnikow glands. Both surfaces of the membrane are abundantly covered with setae, which can considerably increase the general surface of the membrane and thus are suited for a quick discharge of alarm pheromones. The setaceous membrane serves only as a platform to release the alarm pheromones originating from the glandular part of the sting sheaths and the Koschewnikow glands. 4b(20),13b(38,41,43,46)

We studied the fine structure and the effect of the secretions of the sting sheaths on the defensive behavior of a bee colony. (1) The structure of the proximal part of the sting sheath integument has all the characteristics of a primitive exocrine gland. It has porous cuticle with enlarged pore canals and epicuticular pores, and hypertrophied epithelial cells secreting an electron dense material. The non-volatile part of the secretory product embedded the setae of the sheaths. (2) Individual in the laboratory and groups at hive entrances displayed defense behavior only when stimulated with either sting sheaths or with setaceous membrane. We observed enhanced defensive reaction of guards when a venom gland or Koschewnikow glands are added to either sting sheaths or to setaceous membrane.(3) We demonstrate for the first time that the sting sheaths have the structure of an exocrine gland and that their secretions induce defensive behavior in guards. 4b(20),13b(36,38,43,46)

The Koschewnikow glands (KG) of honeybee queens are composed of type III glandular units; secretions are emitted from ducts that open onto the entire surface of the intersegmental membrane.

Secretory activity is characterized by the appearance of dense granules containing mainly glycoproteins, stained by periodic acid-thiocarbohydrazide-silver proteinate, extracted by pronase. In one-year-old mated queens, the gland degenerates. Secretory products originating from mitochondria are likely to be only carriers of pheromonal compounds.

Topical treatment of worker bees ("pseudoqueens") with EtOH extracts of queen Koschewnikow glands induced typical queen balling behavior in workers of a bee colony.

None of them is present in worker alarm pheromone that is secreted from worker KG. 4b(7b),13b(28)

The pheromone-secreting exocrine glands of the honeybee are associated with mouth parts (mandibular glands), sting apparatus (Koschewnikow glands), legs (tarsal glands) and abdominal tergites (Renner glands of queens; Nassanof glands of workers). These glands belong to types I or III previously defined by Noirot and Quennedey (1974). Their ultrastructural characteristics, the sex- or caste-dependent differences and temporal ontogeny have been described.

The pheromonal secretions, acting alone or synergistically, according to various conditions and situations, regulate some major functions which secure the continuation of the society of A. mellifera, such as nuptial flights, monogynous status, " retinue" stability, swarming, optimal exploitation of food sources, territorial defense, etc. 13b (30,31,45)

Tarsal glands of queens, workers and drones are located in the distal part of the tarsus. No morphological or structural features could be related either to sex or to caste.

The gland consists of a glandular epithelium lined by a thin endocuticular layer surrounding the cuticle, and a cellular pocket originating from a post-imaginal delamination of the cuticle that is secreted by the glandular epithelium.

Glandular cells are well characterized by their structural and functional polarity. Apical cellular membranes are provided with numerous cristae and microvilli lining crypts which are large in relation to the subcuticular space. The apical cortical part of the glandular cells shows numerous and various vesicles (pinocytotic vesicles, multivesicular bodies, residual dense bodies). R.E.R. is well developed in the pernicular and basal parts of the glandular cells, but the Golgi apparatus is a discrete organelle deprived of secretory granules. We found no secretory features (exocytosis). To penetrate the glandular pocket, the non-protein secretory product must be able to go across the subcuticular space, the cuticular intima, and the space comprised between the intima and the cuticular wall of the glandular pocket. 4b(2), 13b(9,16)

Each mandibular gland of the queen bee, consists of an axial cavity lined with a thin cuticular intima, secreted by a flat epithelium, and numerous glandular units (type 3), each unit with one duct cell and a large polyploid glandular cell. Mitochondria and endoplasmic reticulum are preponderant organelles. During the ontogeny of the queen bee, the ultrastructure of the glandular cells evolves as her level of pheromonal activity. Variations mainly concern the mitochondrial system. Hence, at the beginning of the imaginal life, the homogeneous population of small mitochondria increases. Towards the 3^rd-5^th day of the queen's imaginal life, the pheromonal activity increases and the mitochondrial differentiation results in the appearance of giant forms. During the highest activity phase (6 to 18-month-old queens), giant mitochondria, associated with endoplasmic reticulum, invade cytoplasmic areas. In queens aged from 18 to 24 months, the reduction of pheromonal and secretory activities is associated with the reduction of mitochondrial population, as well as with the accumulation of lipid droplets and various lytic structures. 13(25)

4b(17), 13b(27)

4b(4), 13b(15)

13b (36)

13b (41)

For detailed abstract see: Alarm pheromones

13b (18)

The numerous gland units are closely associated with oenocytes and fat body cells. Glandular and fat body cells show the sytoplasmatic structures involved in the biosynthesis of proteins: endoplasmatic reticulum, Golgi apparatus and dense vesicles discharging their electron-dense content at the periphery or into the reservoir. The protein fraction of Nassanof secretion may act as a slow release compound of the pheromone. 13b (37)

The wax gland complex of the honeybee worker consists of three cell types, epithelial cells, oenocytes and adipocytes, which act synergistically to secrete wax, a complex mixture of hydrocarbons, fatty acids and proteins (lipophorins). We studied the structure of the wax mirrors and the different types of cells with scanning and transmission electron microscopy. The hydrocarbons coming from the oenocytes and the proteins from the haemolymph transit across the epithelium via the large SER cisternae and through the mirror plate along a well-developed extracellular and pore canal filamentous system connected to wax canal filaments of the epicuticle. 13b( 42)

In this research, soluble proteins are demonstrated on the surface of the honeybee cuticle. Polyacrylamide gel electrophoresis revealed an 80 kDa doublet on workers, queens, and drones showed a single band at a slightly higher molecular weight. Several additional queen- or drone-specific proteins were observed. The presence of soluble cuticular surface proteins may be a general phenomenon because they were also present on insects of two other orders, Orthoptera and Hemiptera. Exocrine gland secretions from the Nasanof glands of workers and the tarsal glands of both female castes and drones also contained soluble proteins. Proteins of tarsal gland secretions may contribute to the milieu of surface proteins because the electrophoretic profiles of cuticular proteins and tarsal secretions were similar. In contrast, the electrophoretic profile of proteins from Nasanof gland secretions was distinct from the profile of the tarsal gland secretion. Proteins on the cuticular surface or in exocrine gland secretions could also affect social interactions. Surface cuticular proteins and those associated with exocrine glands may function in pheromone degradation, as slow release substances, or as caste- or sex-specific modulators of pheromone activity. 13b(34)

Proteins in adult worker bees are secreted from their exocrine glands in royal jelly and in venom or in eggs laid by laying workers in queenless colonies.

We established the identity of proteins from various compartments by double-diffusion, immuno-electrophoresis, gradient acrylamide slab gel electrophoresis, and iso-electric focusing.

Royal jelly is composed of proteins secreted from hypopharyngeal, mandibular, and postcerebral, but not from thoracic glands. No immunological identity was found between the proteins of royal jelly, venom and haemolymph. The glands producing the two secretions seem to be compartments separate from the haemocoele with regard to macromolecular traffic. In contrast to the exocrine glands, many haemolymph proteins were shared by the fat body and the ovaries. 4b(3), 13b(11, 47)

During the pupal ecdysis, which lasted 4 to 6 hr at 28°C; We recorded about 3000 body contractions.

Molting fluid proteins were separated into twenty-one bands on polyacrylamide gels. Eight were specific to molting fluid only; four were common with larval and prepupal haemolymph and four with pupal haemolymph. There were five additional bands with the same R_x in molting fluid, larval, prepupal and pupal haemolymph. Using antiserum against haemolymph following immuno-electrophoresis, eight lines were precipitated in molting fluid and fourteen in prepupal haemolymph.

Absorption of small molecules took place through the general surface of the pupal cuticle. Macromolecules (bromphenol blue tagged haemolymph; peroxidase, etc.) penetrated into new cuticle only at particular sites. 13a(30)

Molting in insects is always accompanied by the production of a molting fluid which fills the exuvial space between the old and the new cuticle and digests the inner layers of the old cuticle. In Hyalophora cecropia, molting fluid is secreted at the outset of adult development and persists until two days before eclosion, whereupon it is absorbed. The present report examines the protein composition of the molting fluid of Cecropia, the origin of the molting proteins, the relation of these proteins to blood proteins and the exchange of macromolecules between the molting fluid and the blood. It also examines the sites of absorption of molting fluid.

Disk electrophoresis on acrylamide gels reveals that the molting fluid of Cecropia contains about fifteen protein bands that can be resolved at pH 8.6. Some of these protein bands are detected in the molting fluid at all stages, whereas others appear only at specific times. About ten of the bands are peculiar to molting fluid and are not detected in the blood. About five bands are detectable in both blood and molting fluid, but none of these common bands appears to be a major component of the molting fluid, and only one is a major blood protein. In contrast, the epidermis contains most of the protein bands found in molting fluid but lacks all but one of the major protein bands present in the blood.

Immunological analysis reveals that blood and molting fluid share five antigens. At least four of these common antigens also occur in the epidermis which appears to secrete these antigens in both the molting fluid and the blood.

Native and foreign proteins do not penetrate from the exuvial space into the blood or vice versa. Apparently the epidermis and cuticle act as a barrier to the exchange of most macromolecules between the blood and molting fluid. The exuvial space is clearly a separate fluid compartment.

In addition the exuvial space itself is compartmentalized and the fluids in the compartments do not admix several days before eclosion.

Absorption of molting fluid during the final two days of adult development occurs most readily through particular regions of the integument. In the abdomen the principal sites of absorption are pits which represent the points through which the tonofibrils make attachments to the old cuticle. Two days before ecdysis, the attachments between the tonofibrils and the pupal cuticle rupture, exposing the points of attachment on the new cuticle. It is through these exposed surfaces that much of the molting fluid is absorbed. Molting fluid is also absorbed in the head and thorax through various flexible membranes at the bases of appendages. 13a(23)

Characterization of haemolymph proteins of honeybee worker larvae by several electrophoretic methods and by ultracentrifugation revealed that they resemble by these criteria the human serum globulins, but not albumins. However, we found no immunological cross-reactions were found between the haemolymph and human, rabbit or avian sera.

The haemolymph was separated into twenty-one bands in 15% acrylamide gels, providing the highest resolution of all the methods and media investigated.

Following immuno-electrophoresis, 14 precipitation lines were detected, most of which were cathodic. Dialysis of haemolymph against dodecyl sodium sulfate, and incorporation of Ampholine into agar, changed the migration of several components toward the anode. These two treatments improved the distribution and facilitated identification of precipitation lines. 13a(27)

We investigated the haemolymph proteins of the developing honeybee workers by disc electrophoresis, double-diffusion and immuno-electrophoresis using antisera against haemolymph of larvae and adults.

Larval haemolymph proteins have been found in 3-day-old eggs. Only slight changes of haemolymph proteins occurred during the larval stage. The first appearance of "adult" proteins was recorded in white pupae. In tanned pupae most of the haemolymph proteins were of the "adult" type.

Three main patterns of haemolymph proteins: larval, adult and common to all stages have been determined during the development of a bee worker from egg to adult. 13a(28)

We have established the presence of drones' haemolymph proteins in the reproductive organs by disc electrophoresis, double-diffusion and immuno-electrophoresis using antiserum against haemolymph.

The samples were separated by disc electrophoresis in 15% acrylamide gels into about 30 components in each. The number of major bands was: haemolymph-14; seminal vesicles-20; mucus glands-20; penis-18. In addition to three haemolymph bands present in all the organs, band 11 was detected in penis and band 13 in seminal vesicles.

Three identical antigens common to haemolymph and organs were detected by double-diffusion. The remaining seven slow diffusing antigens were haemolymph-specific.

Immuno-electrophoresis revealed 14 lines in haemolymph and from two to four lines in the organs.

Penis and mucus glands absorbed bovine serum albumin, but not rabbit g -globulin from the haemolymph. 13a(29)

We analyzed the spermathecal fluid and haemolymph proteins by both disc electrophoresis and immuno-electrophoresis and double-diffusion. Female-specific proteins were detected in the haemolymph. Although almost all proteins of both fluids had similar migration rates, as shown by disc electrophoresis, other methods demonstrated that only some of the protein fractions were immunologically identical or similar. 13a(21)

blood and ovarian proteins of honeybee queens

We analyzed protein components of haemolymph, ovaries, eggs, fat body and midgut wall were analyzed by disc electrophoresis, double-diffusion and immuno-electrophoresis.

We found caste- and sex-specific haemolymph proteins in queens (4 components), workers (2) and in drones (1).

The vitellogenic proteins associated with egg-production were found in 2-year-old laying queens (4 haemolymph and 3 ovarian protein components), but not in 1-week-old non-laying virgin queens. These proteins were also present in the fat body and midgut wall of 2-year-old queens and in eggs.

The uptake of haemolymph proteins by ovaries and oocytes was selective. Common proteins were found as follows: 14 to haemolymph and ovaries; 10 to haemolymph and eggs and 16 to ovaries and eggs. However, foreign protein (bovine serum albumin) injected into the haemocoel was taken up by the ovaries.

During the 3 days of embryonic development a gradual shift occurs from queen haemolymph proteins to larval-specific haemolymph components.

The presence of proteins common to haemolymph, ovaries and eggs in the fat body and midgut wall of the queen bee, suggests that they may originate from these tissues. Also, three protein components seem to be produced in the ovaries and transferred into the oocytes. 4a(4)

Vitellogenin was localized in laying queen and worker ovaries and eggs using antisera against laying and emerging queens' ovaries in double-diffusion, immuno-electrophoresis, and radioimmuno-electrophoretic analyses. Vitellogenin was preparatorily separated by polyacrylamide gradient (5-11%) gel electrophoresis, and the eluate was used to prepare antiserum against vitellogenin. Vitellogenins were detected in the blood of both laying and non-laying females of the two castes, but not in drones.

Vitellogenin was separated by isoelectric focussing into 9 components with PI values between 5.9 and 7.2. Vitellogenin was separated into 26 components by SDS polyacrylamide gradient (5-15%) gel electrophoresis. Female-specific proteins were present in the blood of the two castes, but only a part of them were vitellogenins. Vitellogenins were found in the fat body and the ovaries of laying queens, but not in those of emerging ones. No vitellogenin was found in royal jelly or in the mid-gut of laying queens. 4a(7), 13b(6)

We studied the origin of protein constituents present in larval food of the honeybee by disc-electrophoresis, immuno-diffusion and immuno-electrophoresis using antiserum to larval food. We analyzed: (1) hypopharyngeal, mandibular, postcerebral and thoracic glands of 1 to 19-day-old nurse bees and of field workers, and: (2) food of 1- to 6-day-old queen, worker and drone larvae.

No antigenic components common to the haemolymph of adult bee workers and the larval food were found. This indicates that no protein components of the glandular secretion are taken up from the haemolymph. The hypopharyngeal, mandibular and postcerebral glands participate in the secretion of protein components into the larval food. The thoracic glands did not cross-react with antiserum to larval food.

The secretion of protein components by the hypopharyngeal glands begins in 1-day-old nurse bees and is maximal in 2-week-old bees. All the 14 antigenic components present in larval food are secreted by this gland. The secretion of the mandibular glands begins in 8-day-old nurse bees and is maximal in 2-week-old nurse bees. Two to three antigenic components of the mandibular gland are identical with those secreted by the hypopharyngeal glands and with larval food components. The secretion of protein components from the postcerebral glands begins in 5-day-old nurse bees, but is not steady during the aging.

The secretory activity of protein components by these glands is maximal in 2-week-old nurse bees and corresponds to the age at which they feed larvae.

Immuno-diffusion analyses of food of queen, worker and drone larvae did not reveal any significant differences between the antigenic components. However, immuno-electrophoresis showed that antigens 16 and 17 were present in the food of queen larval of all ages, but were absent from both worker and drone food 4- to 6-day-old larvae. Band 4' was found by disc-electrophoresis in the food of 4- to 6-day-old worker and drone larvae, but not in the food of queen larvae. 4a(3)

Conditioned honeybee workers can discriminate between enantiomeric pairs which are both congruous and incongruous odorants for human beings. These results are discussed in terms of appropriately chiral chemoreceptors, which would enable honeybee workers to distinguish floral odor sources that differ only in enantiomeric identity. 13a(31)

In this study we tried to: (1) correlate, according to the age and polyethism of worker bees, the ontogeny of their mandibular glands and the age-dependent changes of 2-heptanone levels and to: (2) evaluate the role of 2-heptanone in some socio-behavioral interactions of the honeybee.

Each mandibular gland of the worker honeybee consists of an axial cavity lined with a thin cuticular intima elaborated by a flat epithelium, and of numerous glandular units (type 3), each with a duct cell an a large polyploid glandular cell. Mitochondria and endoplasmic reticulum are preponderant organelles. The Golgi apparatus is reduced; its activity is lysosomal. With increasing age, the size of the gland and the amount of 2-heptanone per headspace sample progressively increased. The increasing quantity of 2-heptanone from 0.1 m l at emergence to 7 m l in foraging bees, is well correlated with hypertrophy of the mitochondria, which are engaged in the production of heterogeneous bodies.

We found similar age-dependent changes of 2-heptanone levels in mandibular glands of workers from docile and aggressive colonies: the amount of 2-heptanone in foraging bees of the two groups was alike. Neither crushed glands nor 2-heptanone showed any direct effect as an alarm pheromone on guardian bees at hive entrances. 2-heptanone had either an attractive or repulsive effect on guards, according to the season; it showed a repulsive effect when added to sucrose solution which was visited by foragers. 2-heptanone had a temporary, repulsive effect on the visitation of flowers by foraging bees, hence it seems to act as a "forage marking" pheromone. 2-heptanone does not inhibit the germination of pollen grains. 4b(17), 13b(27)

2-heptanone, a forage-marking pheromone secreted from mandibular glands, was detected in emerging workers and its level progressively increased with age, peaking in foragers, similar to what has been observed for JH-III. This trend is well correlated with the progressive maturation of the mandibular gland, but not with the level of ecdysteroides. 13b(26)

We studied the anatomy of the nectary, the relative amount of nectar secreted, the composition of the nectar and its sugar concentration, in 9 species of the family, Labiatae (Coridothymus capitatus, Melissa officinalis, Phlomis viscosa, Rosmarinus officinalis, Salvia fruticosa, S. hierosolymitana, S. judaica, Satureja thymbra, Stachys aegyptiaca). We investigated the visiting rate of the honeybees to the various species. The nectar is secreted from the asymmetrically thick, disc-like nectary, through modified stomata. These occur on the thicker side of the nectary only. There is a correlation between the volume of the nectary and the amount of nectar secreted. The larger the nectary, the more nectar is secretes. No correlation was found between the type of vascular tissue (phloem, xylem) and the sugar concentration in the nectar. Rosmarinus officinalis and Coridothymus capitatus were the species most frequently visited by honeybees. The quantitative relationship between sucrose, glucose and fructose differed in the various species examined. Three groups of species were distinguished: 1. Melissa and Rosmarinus, in which the amount of hexoses was equal to that of sucrose; 2. Coridothymus, Salvia judaica and Stachys, in which the hexoses were dominant; 3. Phlomis, Salvia fruticosa and Satureja, in which sucrose was dominant. The concentration of the amino acids in the nectar was, in general, low. No clear correlation between the characteristics of the flowers or of the nectar and the frequency of honeybee visits could be established. 4b(11), 13b(22)

The chemical composition of Jaffa Orange (Citrus sinensis L. cv. Jaffa) blossom volatiles and their attractiveness to foraging honeybees

We report the chemical composition (GLC, GLC-MS) of volatile components of orange flower scent (OFS) and their attractiveness to honeybee workers. The OFS volatiles were separated by GLC and GLC-MS into 82 components (peaks), 31 of which were identified as terpene alcohols, terpenes, aldehydes, ketones, sesquiterpene alcohols, aromatic alcohols and esters. The OFS contained also volatile components of the worker honeybee Nassanof gland pheromone (NGP) including: (Z)-citral, (E)-citral, nerol, geraniol and (E,E)-farnesol. ± -linalool was the major component (>40%) of OFS. Twelve new compounds were identified. In a "cue" and "reward" system, using scented 30% sucrose solution (s.s.), foragers preferred (Z)-citral and (E)-citral, geraniol, nerol, (E,E)-farnesol and methyl anthranilate to a -terpineol, (± )-linalool and g -terpinene. Young, inexperienced hive bees preferred "cues" (with no "reward") of compounds that are shared in common by OFS and NGP: [(Z)-citral, (E)-citral, geraniol, nerol, and (E,E)-farnesol] to the OFS-specific volatiles. An OFS blend of 17 authentic compounds was 55% more attractive to foragers than the pure 30% s.s. Young, inexperienced hive bees were not attracted by this blend. However, both foragers and nurse bees were attracted by a blend of NGP compounds. 4b(20)

We investigated the attraction of honeybees to petri dishes containing one of several concentrations of the attractant Beeline^R , or a combination of the two. None of the observations demonstrated a preference of the foraging bees for Beeline over the tap-water control. Sucrose solutions were most attractive to honeybees. Maternal -parent cucumber plants of the F₁ hybrid cultivar Delila were sprayed with an aqueous solution of Beeline following the manufacturer's instructions. Flowers of treated and untreated control plants were visited by bees at similar frequencies, both setting the same percentage of seed per fruit of similar weight per thousands seeds. We concluded that Beeline is totally ineffective in attracting bees to the maternal parent of the cucumber cultivar Delila under summer conditions in the coastal plain of Israel. 4b (7), 13b (12).

Honeybees are attracted to nectaries only in the pistilate flowers, and to both nectaries and anthers in the staminate flowers, but ignore all other floral organs. There were diurnal fluctuations in the preference of honeybees to pistilate and staminate flowers. The number of bee visits to female flowers reached its maximal value at 09 hr, while the highest frequency of visits to male flowers was attained at noon.

Significant differences both in nectar volume and in sugar concentration, were found between flowers of various gynoecious genotypes of cucumber. In addition, marked variations of these properties occurred also within each genotype.

Bee visitation to cucumbers was greatly affected by nectar volume. Larger number of bees were attracted to flowers with high volumes of nectar, than to those with little or without nectar secretion. No correlation was found between sugar concentration or total sugar quantity in the nectar, and the frequency of bee visits to cucumber flowers of various genotypes.

Crossings between non-secreting nectar gynoecious plants and nectar secreting pollinating parents resulted in hybrids with nectar secreting flowers. These findings indicate that nectar secretion is a property controlled by a dominant or a codominant factor.

The composition of sugars in the nectar of various cucumber flowers, varied with genotypes. Cucumber genotypes can be classified in accordance with the sugar composition of their nectar, as follows: a) a group dominated by sucrose; b) a group dominated by fructose. The fructose dominated nectar are more attractive to honeybees than the others. In addition, the former contain some unidentified components, which may also contribute to bee attractiveness.

No anatomical lesions or damage to nectaries were found in low-secreting genotypes. There was no direct relation between the number of open stomata on the nectary surface and the amount of secreted nectar. Therefore, the anatomical observations could not explain the difference between flowers, in regard to nectar secretion.

The number of bee visits to cucumber flowers was increased with nectar secretion. Each of these two characteristics was found to be significantly correlated with seed production. These findings should be utilized by plant breeders in order to improve both bee visitation and seed production of this crop. 4b(10)

We studied honeybee activity and environmental factors affecting the pollination of F₁ hybrid seed production of the cucumber cv. "Delila."

Our observation showed that nectar collecting forager bees have free access to the flower nectary, but while gathering nectar their body must brush the anthers area in the staminate flower and the stigma in the pistilate flower. The structure of the flowers and the strong adhesence of pollen grains to the hairs and bristles of the bee body surface indicate, that the transfer of pollen grains from staminate to pistilate flowers may effectively be carried out. The onset of the foraging activities of honeybees was observed in the early morning (6-7h), reached a maximal level in the late morning (9-12h), and it decreased early in the afternoon (14-16h). Frequency of bee visits to pistilate flowers was lower than to staminate ones, but their duration was longer in the former than in the later flowers (8-16 sec. and 4-10 sec., respectively).

Under field conditions, the nectar on the surface of the cucumber flower's nectary dried already early in the morning, leaving a solid sugar layer, which might be the cause for the low frequency of bee visits, as compared with many other plant species.

In flowers covered with gelatin capsules to prevent nectar evaporation, the amount of nectar secreted from a pistilate flower was higher than by a male flower (5-10 vrs. 1-3 microliters, respectively). However, sugar concentration of the nectar was higher in staminate flowers than in pistilate ones (40-45% vrs. 28-35%, respectively). Considerable differences in the amounts of nectar secreted by various genotypes were found under controlled conditions, indicating an inherent variation in this respect.

Neither nectar secretion nor seed production were significantly affected by application of different minerals, except for zinc treated plants, which showed a 60% increase in seed-set.

Bee hives were placed in two fields in either of two ways: a) along the fringes of the field (400x10 m.); b) along and across the beds (40x50 m.), Berurim farm and Kibbutz Shuval, respectively. At Berurim, the number of visiting bees in cucumber flowers at a distance of 50 m. from the hives was higher than at 5 or 200 m., however, no linear relation between the rate of seed set and the activity of the bees at these distances was found. At Shuval, higher uniformity in seed set per fruit was found along the row, probably because of better distribution of the bee colonies in the field.

The seed yields in beds of female plants, which were remote from source of pollen, were 30% lower than in those closer to beds of male parent. 4b(7)

We have examined, both under field and under controlled conditions, the response of the reproductive system to growth conditions; with special reference to the effect of temperature, using mainly the parental lines of F₁ cv. "Delila". In some of our experiments, additional cultivars and breeding lines were also examined.

Pollen grains collected in the summer, in the mornings of anthesis days, had a high viability of over 90%, the quantity produced however, was relatively small, 6,000 to 10,000 grains per male flower.

High temperatures were found to reduce germination ability of the pollen, both when applied to the growing plants and to the germination medium in vitro. In the latter case only 50% of the pollen grains germinated at temperature above 40° C.

We found no recognizable changes in receptivity of the stigma and style of female flowers of the gynoecious plant during anthesis day. In gynoecious plants there were, however, differences in fertility, as expressed by the percentage of fruit and seed set.

In summer, fruit and seed set were markedly affected by the differences in the environmental conditions because the three experimental fields were sown at different dates.

The position of the female flower on the main axis had a certain effect on yields. In one experiment flowers developed on higher nodes were more productive, yielded higher numbers of seeds. 4b(8)

Flower morphology, nectary structure, nectar traits and rates of honeybee foraging on pepper plants were studied. The nectary appears as swellings on the basal part of the ovary. The nectariferous cells are smaller and denser than the neighboring parenchyma. Stomata are present in the nectary epidermis, but do not appear on the other parts of the ovary epidermis. Seven pepper breeding lines were grown near a bee yard in Rehovot. Five to six fold differences in nectar volume were found between the extreme genotypes. Nectar volumes were higher during noon and afternoon hours, as compared with morning hours. High correlation coefficients between nectar volume and sugar concentration wee found. These were significant for the four high nectar yielding genotypes, ranging between r=0.65 to r-0.94. Male-fertile flowers produced more nectar and higher sugar concentration than sterile ones. Skewed distribution was observed in nectar volume of F₂ populations, but relatively low heritability values were calculated.

Pepper nectar contains fructose and glucose only. The former occupies 52-82% of the total sugar content. Pepper genotypes varied in frequency of honeybee visits and significant correlation between sugar quantity and number of honeybee visits per flower was evident. Fertile pepper flowers are not very attractive to honeybees and male-sterile flowers are even less so. The considerable variation in nectar characteristics can be exploited to increase attractiveness to honeybees, thus facilitating bee pollination in commercial production of F₁ hybrid seeds and improve fruit quality. 4b(15), 13b(35)

Poor pollination is one of the major problems in the production of inter-specific hybrid cotton seeds (Gossypium hirsutum X Gossypium barbadense). We investigated in this study: (i) the factors that attract honeybees to cotton flowers; (ii) the flower traits that are responsible for the pollination efficiency; (iii) the variance of these factors in the examined plants.

The honeybees preferred some of the genotypes more than the others: hybrid flowers attracted more bees than all the other genotypes, G. hirsutum flowers attracted less bees and G. barbadense flowers were the least attractive.

Large differences were found among the different genotypes in the nectar volume secreted from the flowers. Lines which contained large quantities of nectar in their flowers, had more honeybee visits and bore more seeds than lines that their flowers contained small quantities of nectar.

The nectary area in the flowers was in high correlation with their nectar volume. There were only small differences between the genotypes in the nectar sugar concentration of their flowers. The composition of sugars in the nectar was also the same in all of the genotypes. These factors do not affect the selective attractiveness of honeybees to cotton flowers.

The pollen grains of G. barbadense flowers repelled the bees in the hive (chemical repellence) and extensively covered the bees visiting the same flowers (physical repellence). We believe that these two factors made this type of flower less attractive to the bees despite their large nectar volume.

The genotypes were also different in flower traits which were used by the bees to distinguish between the lines: color, size, shape and smell. These differences caused the bees, in a certain degree, to limit their visits to one type of flowers.

The male sterile genotypes were found to be different in the length and diameter of their stigma, the opening angle of the corolla and the time of flower opening. A large stigma may increase the number of pollen grains attached to it by the bees. A narrow opening angle of the male sterile flowers forces the bees to touch the stigma because otherwise the bees will seldom touch it. If the male sterile flowers would open after the male fertile ones, the bees will enter the half-open male sterile flowers with their bodies covered with pollen, touching the stigma in the narrow flower (in contrast to the opposite situation). The male fertile genotypes varied in the number of their flowers, the number of stamens per flower and the percentage of cleistogamic flowers. The loss of much of the pollen grains, as a result of the bee behavior, may require an increase in the production of pollen grains by the male fertile flowers, by means of breeding the plants to improve the traits mentioned above.

Pollination by honeybees in the afternoon was less successful than in the morning, despite the small reduction in the viability of the pollen grains. This was caused by the reduction in the activity of the bees and the amount of pollen grains in the stamens, by decreased stigma receptiveness and by the lack of contact between the bees and the stigma in the fully opened male sterile flower.

Infra-specific hybridization usually bore more seeds then inter-specific hybridization, but one of the male sterile lines bore the same amount of seeds in both of the hybridizations.

Examining the inheritance of flower-traits in the hybrids, a significant heterosis was found in the following traits: number of stamens per flower, stigma length, and calyx diameter. The width of the nectaries strip was intermediate between that of the two parents.

For the promotion of hybrid cotton seed production, it is necessary to breed both parent lines for flowers that would be attractive to bees, and that the bees would not be able to distinguish between them (high nectar volume and a similar flower type that does not repel the bees), and for traits that will enable the stigma to be well covered with pollen grains (narrow flower, big stigma, right time of the flower opening and a male fertile line with a high production potential of pollen grains). 4b(16)

We compared short- and long-term exposure periods of fluvalinate to honeybee colonies for effects on population levels, honey production and fluvalinate residues in wax and honey. We conducted the study in three climatic zones of Israel (Hula and Jordan Valleys, northern and southern coastal area) to examine the possibility of reducing the frequency of fluvalinate treatments to two per year.

When fluvalinate inserts were placed in the brood-nest center of bee colonies for either 6 weeks or 6-8 months, the resulting bee populations were similar. However, the Varroa mite populations remained low only when exposed to fluvalinate for a 6 to 8 month treatment period. Long-term treatment reduced honey yield and increased fluvalinate residues in wax and honey.

Fluvalinate inserts were placed at the hive entrance and in the brood-nest. Hive-entrance inserts effectively controlled Varroa and lowered fluvalinate residues in wax and honey when compared to inserts placed in the brood center. Entrance placement also reduced time and labor costs. 13b(29)

Origanum oil and thymol, naturally occurring compounds of plant origin, were evaluated in bee colonies containing brood year-round in a subtropical climate as alternate acaricides of an integrated control program for Varroa jacobsoni.

Treating colonies in March-April with 20 and 33% origanum oil impregnated in cardboards resulted in 85 and 91% mortality of Varroa , respectively. The use of pure origanum oil during summer was harmful to bee colonies. Application of 10% of thymol (in cardboards) to bee colonies resulted in 54-85% mortality of Varroa, but 30% thymol was harmful to the bee colonies. The constraints on the use of essential oils to control Varroa in a subtropical climate are discussed. 13b(33)

We investigated the effect of Mavrik inserts on Varroa in Israel and on both Varrroa and Tropilaelaps in Thailand. No mortality or damage to either developing or adult bees or any effects on the egg-laying rate of queen bees could be detected.

In Israel, 14 to 16 days after placing Mavrik inserts inside brood-nest beneath queen excluders , the population level of Varroa was reduced to less than 10 mites per bee colony. The inserts were removed after 14-16 days to prevent contamination of honey with fluvalinate.

In Thailand, similar results were obtained in A. mellifera bee colonies infested by both Tropilaelaps and Varroa mites.

No detectable traces of fluvalinate were found in honey samples removed from supers above queen excluders. Honey removed from combs within an immediate contact with the inserts in the brood-nest contained 0.057 ppm fluvalinate on the average.

As a result of this study the control of Varroa in Israel is based now on Mavrik inserts, which is less labor intensive than Amitraz fumigations. 13b(21)

We controlled Varroa jacobsoni and Tropilaelaps clareae with fluvalinate in 231 Italian colonies in three commercial beeyards south of Chiang-Mai. Hive entrance placement of fluvalinate for 21 days effectively controlled Varroa and Tropilaelaps (94% mortality). We found no detectable fluvalinate residues either in honey or in wax samples obtained from brood-chambers of colonies treated with fluvalinate. 13b(48).