Semen Production in Drone Honeybees

RIRDC Pub. No. 08/130

Semen Production in Drone Honeybees

By John W. Rhodes

October 2008 RIRDC Publication No 08/130 RIRDC Project No PRJ-000508

 2008 Rural Industries Research and Development Corporation. All rights reserved.

ISBN 1 74151 719 2 ISSN 1440-6845 Semen Production in Drone Honeybees Publication No. 08/130 Project No. PRJ-000508 The information contained in this publication is intended for general use to assist public knowledge and discussion and to help improve the development of sustainable regions. You must not rely on any information contained in this publication without taking specialist advice relevant to your particular circumstances. While reasonable care has been taken in preparing this publication to ensure that information is true and correct, the Commonwealth of Australia gives no assurance as to the accuracy of any information in this publication. The Commonwealth of Australia, the Rural Industries Research and Development Corporation (RIRDC), the authors or contributors expressly disclaim, to the maximum extent permitted by law, all responsibility and liability to any person, arising directly or indirectly from any act or omission, or for any consequences of any such act or omission, made in reliance on the contents of this publication, whether or not caused by any negligence on the part of the Commonwealth of Australia, RIRDC, the authors or contributors. The Commonwealth of Australia does not necessarily endorse the views in this publication. This publication is copyright. Apart from any use as permitted under the Copyright Act 1968, all other rights are reserved. However, wide dissemination is encouraged. Requests and inquiries concerning reproduction and rights should be addressed to the RIRDC Publications Manager on phone 02 6271 4165.

Researcher Contact Details Mr. John W. Rhodes Name: Address: NSW Department of Primary Industries, 4 Marsden Park Rd., Calala, NSW, 2340 0267 63 1206 Phone: 0267 63 1222 Fax: john.rhodes@dpi.nsw.gov.au Email: In submitting this report, the researcher has agreed to RIRDC publishing this material in its edited form. RIRDC Contact Details Rural Industries Research and Development Corporation Level 2, 15 National Circuit BARTON ACT 2600 PO Box 4776 KINGSTON ACT 2604 Phone: Fax: Email: Web: 02 6271 4100 02 6271 4199 rirdc@rirdc.gov.au. http://www.rirdc.gov.au

Published in October 2008 by Union Offset

ii

Foreword
The commercial queen bee breeding industry in eastern Australia produces queen bees between spring and autumn and mates them at queen bee mating apiaries with each mating apiary receiving a new batch of queen bees on a regulated time scale to provide maximum turnover of queen bee numbers from each mating apiary. There is a large amount of published information and experience amongst queen bee producers on selection of genetic breeding stock, grafting age and nutritional requirements of the queen bee during rearing to the mating age. There is less published information available on (i) the selection criteria required for selecting queen bees to be drone mothers for rearing large numbers of drone bees which produce high quality semen measured in terms of semen volume, number of sperm produced, sperm viability and sperm motility, and (ii) the maintenance of large numbers of drones to an age with their semen continuing at a high quality level. The objective of this project was to examine the semen quality of drone honeybees at a range of drone ages over the seasons that queen bees are reared and mated from a number of commercial breeding lines in use in eastern Australia, and compare the obtained data with published data. Commercial beekeepers that use queen bees reared in large numbers from queen mating apiaries in their honey production and pollination apiaries should benefit from this research by allowing them to have queen bees with large numbers of sperm stored in their spermathecae after mating with the beneficial effect of producing queen bees with an extended working life. In general, drone honeybees reared from commercial lines of bee breeding stock in eastern Australia were found to produce low numbers of sperm when results were compared with published data from Europe and the USA. Sperm viability and motility results were at the lower end but comparable with limited published data. One breeding line of queen bees constantly produced superior drones than to that from other breeding lines for semen volume, number of sperm, sperm viability and motility, suggesting that these characteristics may be able to be selected for in a breeding program. Recommendations from the results include the development of breeding programs for drone mother queen bees with selection criteria based on semen volume, number of sperm, sperm viability and sperm motility, and the development of programs for the management of drone producing colonies used at queen bee mating apiaries for the production of large numbers of functional drones with high longevity. This report, an addition to RIRDCs diverse range of over 1800 research publications, forms part of our Honeybee R&D program, which aims to improve the productivity and profitability of the Australian beekeeping industry. Most of our publications are available for viewing, downloading or purchasing online through our website: downloads at www.rirdc.gov.au/fullreports/index.html purchases at www.rirdc.gov.au/eshop

Peter OBrien Managing Director Rural Industries Research and Development Corporation

iii

Acknowledgments
I would like to thank the Rural Industries Research and Development Corporation, Honeybee Research and Development Program, Canberra, the University of Western Sydney, Hawkesbury Campus, and the New South Wales Department of Primary Industries for providing financial assistance and facilities. I would like to express my appreciation to Associate Professor Robert Spooner-Hart, Centre for Plant and Food Science, University of Western Sydney, Hawkesbury Campus for guidance during the progress of this study. The following people provided major support and assistance. Ms Gretchen Wheen, Richmond, NSW, for collection of samples, use of apiary and laboratory facilities, and interpretation of results. Ms Liz Kabanoff, University of Western Sydney, Hawkesbury, for support and laboratory use for sperm viability experiments. Dr Denis Anderson, CSIRO Entomology, Canberra, for disease identification experiments and use of facilities. Mr Michael Duncan, Apiary Manager, University of Western Sydney, Hawkesbury, for assistance and use of apiary facilities. Mr Steven Harden and Mr Bruce McCorkell, Biometricians, NSW DPI, Tamworth, for statistical analysis of data. Dr Bruce Tier, Animal Genetics and Breeding Unit, University of New England, Armidale, NSW, for assistance in developing experimental methodology. Dr Adrian Nicholas, NSW DPI, Tamworth, for use of laboratory facilities. Amino acid analysis of drone semen samples was carried out by the Australian Proteome Analysis Facility Ltd, Macquarie University, Sydney, NSW and fatty acid analysis by the NSW Department of Primary Industries, Oil Testing Service, Wagga Wagga, NSW and I would like to thank Dr Rod Mailer and staff. The efforts of both groups in analysing samples of small volume were appreciated. Commercial beekeepers who contributed with queen bees, use of facilities and interpretation of results, Mr Col Wilson, Kurri Kurri, NSW; Mr Ken Olley, Clifton, Qld.; Mr Linton Briggs, Glenrowan, Victoria; Mr Frank Malfroy, Richmond, NSW; Mr Greg Mulder, Murrays Run, NSW; Mr Greg Tyson, Inverell, NSW; and Mr David Weik, Tamworth, NSW. I would like to thank members of the beekeeping section, NSW DPI for support. Mr Tim Burfitt, Manager, Intensive Livestock Industry Development, Orange; Mr Bruce White, former Technical Specialist, Richmond, Dr Doug Somerville, Technical Specialist, Goulburn; and Mr Nick Annand, Livestock Officer, Bathurst.

iv

Contents
Acknowledgments................................................................................................................................. iv Executive Summary ............................................................................................................................ vii 1. Introduction ....................................................................................................................................... 1 1.1 References ..................................................................................................................................... 4 2. Determination of drone sample size for drone sperm numbers experiment................................ 5 2.1 Introduction ................................................................................................................................... 5 2.2 Method .......................................................................................................................................... 5 2.3 Statistical method .......................................................................................................................... 5 2.4 Results .......................................................................................................................................... 5 2.5 Conclusions ................................................................................................................................... 6 2.6 References ..................................................................................................................................... 6 3. Volume of semen and number of sperm per drone. ....................................................................... 7 3.1 Introduction ................................................................................................................................... 7 3.2 Methods....................................................................................................................................... 11 3.3 Statistical analysis ....................................................................................................................... 13 3.4 Results ......................................................................................................................................... 13 3.5 Discussion ................................................................................................................................... 14 3.6 Conclusions ................................................................................................................................. 16 3.7 Tables .......................................................................................................................................... 17 3.8 References ................................................................................................................................... 19 4. Sperm Viability................................................................................................................................ 21 4.1 Introduction ................................................................................................................................. 21 4.2 Methods...................................................................................................................................... 22 4.3 Statistical Analysis ...................................................................................................................... 23 4.4 Results ......................................................................................................................................... 23 4.5 Discussion ................................................................................................................................... 23 4.6 Conclusions ................................................................................................................................. 24 4.7 Tables .......................................................................................................................................... 25 4.8 References ................................................................................................................................... 27 5. Sperm Motility................................................................................................................................. 28 5.1 Introduction ................................................................................................................................. 28 5.2 Methods....................................................................................................................................... 28 5.3 Statistical Analysis ...................................................................................................................... 29 5.4 Results ......................................................................................................................................... 29 5.5 Discussion ................................................................................................................................... 29 5.6 Conclusions ................................................................................................................................. 30 5.7 Tables .......................................................................................................................................... 31 5.8 References ................................................................................................................................... 33 6. Role of genetics in drone semen production.................................................................................. 34 6.1 Introduction ................................................................................................................................. 34 6.2 Methods....................................................................................................................................... 36 Two-way cross methodology ........................................................................................................... 36 Confirmation of High and Low Lines .............................................................................................. 37 6.3 Statistical analysis ....................................................................................................................... 38 6.4 Results ......................................................................................................................................... 38 Confirmation of High Line and Low Lines...................................................................................... 38 Disease examination......................................................................................................................... 39 6.5 Discussion ................................................................................................................................... 39 Confirmation of High and Low Lines .............................................................................................. 39 Disease examination......................................................................................................................... 39 6.6 Conclusions ................................................................................................................................ 40 v

6.7 Tables ......................................................................................................................................... 41 6.8 References ................................................................................................................................... 44 7. Drones not producing semen after manual eversion initial examination................................ 45 7.1 Introduction ................................................................................................................................. 45 7.2 Method ........................................................................................................................................ 45 7.3 Results ......................................................................................................................................... 46 7.4 Discussion ................................................................................................................................... 46 7.5 Conclusions ................................................................................................................................. 46 7.6 Tables .......................................................................................................................................... 47 7.7 References ................................................................................................................................... 47 8. Manual eversion of drones comparison between two operators .............................................. 48 8.1 Introduction ................................................................................................................................. 48 8.2 Method ........................................................................................................................................ 48 8.3 Statistical analysis ....................................................................................................................... 48 8.4 Results ......................................................................................................................................... 48 8.5 Discussion ................................................................................................................................... 48 8.6 Tables .......................................................................................................................................... 49 9. Investigation of release of semen by drones after manual eversion............................................ 50 9.1 Introduction ................................................................................................................................. 50 9.2 Methods....................................................................................................................................... 51 9.3 Statistical Analysis ...................................................................................................................... 51 9.4 Results ......................................................................................................................................... 52 1. Comparison of sperm numbers in endophallus and in seminal vesicles ...................................... 52 2. Nosema disease examination........................................................................................................ 52 3. Endophallus eversion ratings ....................................................................................................... 52 4. Number of sperm produced per drone.......................................................................................... 53 9.5 Discussion ................................................................................................................................... 53 9.6 Conclusions ................................................................................................................................. 55 9.7 Tables .......................................................................................................................................... 56 9.8 References ................................................................................................................................... 64 10. Seminal vesicles v. endophallus sperm viability and motility................................................. 65 10.1 Introduction ............................................................................................................................... 65 10.2 Method ...................................................................................................................................... 65 10.3 Statistical analysis ..................................................................................................................... 65 10.4 Results ....................................................................................................................................... 66 10.5 Discussion ................................................................................................................................. 66 10.6 Conclusions ............................................................................................................................... 66 10.7 Tables ........................................................................................................................................ 67 11. Drone honeybee semen amino acid and fatty acid content..................................................... 68 11.1 Introduction .............................................................................................................................. 68 11.2 Methods.................................................................................................................................... 70 11.3 Results ...................................................................................................................................... 71 11.4 Discussion ................................................................................................................................ 71 11.5 Conclusions .............................................................................................................................. 73 11.6 Tables ....................................................................................................................................... 74 11.7 References ................................................................................................................................ 80

vi

Executive Summary
Background The importance of drone semen quality studies has its effect on the mating success of queen bees. Sufficient numbers of mature age drones, each producing a large volume of semen containing a high number of sperm are required to be present at commercial queen bee mating apiaries to result in queen bees with the maximum number of sperm present in their spermathecae after mating. Low numbers of sperm in the spermathecae of queen bees after mating contributes to early supersedure resulting in increased costs to the beekeeper from queen replacement and reduced colony production as colonies weaken in strength or become queenless during the period of queen replacement. What the report is about This project provides information to persons interested in bee breeding with particular reference to queen bee production by providing data relating to numbers of sperm produced by drones and highlighting the requirement of use of this data in the selection of queen bee drone mothers by including drone sperm production in the selection criteria and by identifying the most suitable method for collecting drone sperm data. Objectives The objective of this project was to examine the semen quality of drone honeybees at a range of drone ages over the seasons that queen bees are reared and mated from a number of commercial breeding lines in use in eastern Australia, and compare the obtained data with published data. Methods used Initial experiments were carried out during 2003 and 2004 to examine the production and quality of semen from drone honeybees Apis mellifera from different commercial breeding lines, at different ages and at different seasons during the commercial queen bee production period in eastern Australia (i.e. between spring and the following autumn) to determine whether problems can occur with breeding lines, drone age groups or times during each season when semen production and/or quality may not be satisfactory for mating requirements of queen bees. Four breeding lines of queen bees were selected to provide drones. Drones were reared during months of the year to represent spring, summer and autumn reared drones. For each breeding line and for each season, drones were sampled at three ages; 14, 21 and 35 days. Each sample comprised 30 drones or the number surviving up to 30. For initial examinations of drones for the 2003 and 2004 experiment, manual eversion was used to obtain information on the proportion of drones which released semen at the endophallus; the volume of semen produced per drone; and the number of sperm produced per drone. From the number of drones marked at emergence a high percentage of drones did not survive to 35 days of age. For the four breeding lines examined, significant differences were found between the four lines for the proportion of drones producing semen at the endophallus after manual eversion; for the volume of semen produced per drone; and for the number of sperm produced per drone. These data suggest that genetics may be involved and these three traits may be able to be selected for in a breeding program. One breeding line was consistently superior when compared with the other lines. Results In the 2003-04 experiment, 40.6% of all drones examined at ages 14 to 35 days did not produce a measurable amount of semen at the endophallus after manual eversion. Although a number of authors have made reference to the proportion of mature age drones not producing semen at the endophallus after manual eversion, it does not appear to have been identified as a subject of interest and neither has the importance of its apparent consequences been recognised (e.g. whether these drones produce semen and whether they are capable of mating with queen bees). These questions are of interest during the preparation of commercial mating apiaries for mating large numbers of queen bees.

vii

Spring reared drones produced higher volumes of semen than summer drones which were higher than autumn drones. However, autumn drones produced higher numbers of sperm than summer drones which were higher than spring drones indicating a seasonal variation in sperm concentration in semen. An age effect was found with 35 day old drones having a higher proportion of drones producing semen at the endophallus after manual eversion and producing a lower volume of semen than 14 and 21 day old drones. There was a high mortality rate amongst 35 day old drones and the higher percent of 35 day old drones producing semen at the endophallus after manual eversion, compared with 14 and 21 day old, suggests that drones able to produce semen at the endophallus after manual eversion have a higher survival rate than drones not able to produce semen and which die at a younger age. There was no significant difference between 14, 21 and 35 day old drones for the number of sperm produced. A large range was found for the volume of semen produced, mean 1.09 (ra.0.77 0.05 1.16 0.04) l which was comparable to published data from Australia, Europe and the USA. The range of sperm numbers per drone produced by drones in this experiment, mean 3.63 (range (ra.) 2.19 0.19 4.72 0.23) 106, in general, were low compared with published data from Australia, Europe and the USA. Although drone ages and seasons were identified when drone semen was of a higher quality, no drone age or season was found when drones were considered un-suitable for mating with queen bees based on semen quality. Sperm viability assessment of semen samples was carried out using a Live/Dead Sperm Viability Kit containing a membrane-permeant nucleic acid dye (SYBR 14) which stains live cells red and a conventional dead-stain cell (propidium iodide) which stains damaged or dead cells red. Drone sperm viability averaged 79.7 (ra. 59.9 - 90.5) %. A significant season effect was found with autumn drones recording higher viability than spring and summer drones; a significant age effect with 21 day old drones with highest viability followed by 35 day old and 14 day old drones with the lowest. A significant breeding line effect was found with one line showing lower viability than the other three lines. Sperm viability data from this experiment are comparable with data published by USA authors and suggest that an improvement in viability percent may be possible through a queen bee breeding program selecting for high sperm viability in drones produced from those queen bees. Sperm motility was assessed at room temperature within 5 minutes of the semen sample being collected based on a motility rating using five levels of sperm movement, 1 to 5, 1 = lowest movement. Drone sperm motility averaged 3.4 (ra. 2.1-4.5). There was a significant season effect with spring and summer drones displaying higher motility than autumn drones; a significant age effect with 35 day old drones with higher motility ratings than 21 day old which were higher than 14 day old; and a significant breeding line effect with one line displaying lower sperm motility than the other three lines. This data is at the lower end but comparable with limited USA data and suggest that sperm motility may be a selectable trait and may be improved in a bee breeding program. An examination of drones was planned to be carried out between 2004 and 2006 using a Two-way Cross methodology to determine whether genetic inheritance is involved in the production of semen at the endophallus after manual eversion, the volume of semen and the number of sperm produced in drone honeybees. Queen bees from the two breeding lines producing the highest and the lowest data from the 2003-04 experiments were superseded in 2005. Drones from replacement queen bees from the same breeding lines as the original queen bees were examined twice for the three characteristics being examined with no significant differences identified, this experiment was then suspended. Data from this experiment suggest that should the three characteristics being examined be selectable traits then recessive genes may be involved in the expression of these characteristics suggesting that identification and maintenance of these traits in queen bee breeding stock would require a continuing managed breeding program.

viii

An experiment was carried out in 2006 to examine seminal vesicles from mature age drones which did not produce semen at the endophallus after manual eversion for the presence of sperm. Of the 30 drones examined that produced semen at the endophallus, 5 drones contained small numbers of sperm in the seminal vesicles, all 30 drones not producing semen at the endophallus after manual eversion contained sperm in their seminal vesicles. In 2007, an experiment was carried out to examine the release of semen by drones after manual eversion to investigate the relationship between semen release and drone maturity and to examine the accuracy of semen volume and sperm number data using drone manual eversion as the method for sample collection. Drone honeybees from four commercial breeding lines were examined each day at ages between 3 and 25 days and at 30 days for the amount of sperm present in the endophallus and the amount remaining in the seminal vesicles after manual eversion. The level of eversion of the endophallus at each age was recorded at a rating of 0-4, 0 = no eversion and 4 = full eversion. Drone intestines were dissected from all drones and examined individually for Nosema spp. spores at four ages. The data suggests that large numbers of drones at a young age are at the lower eversion rating and large numbers of older age drones are present at the higher eversion rating, the change in numbers of drones from low to high rating is gradual and does not identify a specific age at which drones change from low release of semen to high release of semen suggesting that drone maturation occurs over a wide range of drone ages. Data from this experiment suggest that manual eversion is not a suitable method for collecting semen samples required to provide accurate semen volume and sperm number data per drone, with dissection and counting sperm numbers from the seminal vesicles providing a more reliable alternative. Compared with data from Europe and the USA, low numbers of sperm per drone were recorded from this experiment suggesting that selection of drone mother queen bees whose drones produce large numbers of sperm should be included in breeding stock selection criteria for breeding programs. Nosema disease was found not to affect sperm counts. In a 2007 experiment, sperm present in the seminal vesicles and in the endophallus after manual eversion were assessed for viability and motility to identify differences between sperm which move to the endophallus and sperm which remains in the seminal vesicles No significant interaction was found between the source of the sperm i.e. sperm remaining in the seminal vesicles or sperm which had moved to the endophallus with either sperm viability or sperm motility indicating that lower sperm quality, measured in terms of viability and motility, is not the cause of sperm remaining in the seminal vesicles of mature age drones following manual eversion. Semen from drones of different ages and sampled at different seasons was analysed to determine the amounts of amino acids and fatty acids present to examine for changes in the amounts present which may affect semen quality from drones at certain ages or at different seasons. Eighteen free amino acids were identified, a high content percent of arginine, average (av.) 52%, and proline, av. 24% of the total amino acids present in each sample, was found with lower amounts, less than 7%, of each of the remaining 16 amino acids. Thirty two fatty acids were identified from drone semen. High levels of oleic acid and elaidic acid were found with all other identified fatty acids present at less than 7% content for each semen sample. The importance of having large numbers of drones, each producing a large volume of semen containing high numbers of sperm with high viability and high motility, being available to mate with queen bees at commercial queen bee mating apiaries increases as the amount of time is reduced that a queen bee has to complete the mating process before she terminates mating flights and commences egg laying. Should chemicals present in drone seminal fluid be found to affect the amount of time a queen bee has to complete the mating process, as suggested by research results from other insect species, and then this factor translates to major changes being required in current management practices of drone rearing colonies. Management practices would require evaluation and selection of queen bees whose drones have the characteristics of high semen volume, high sperm numbers, high sperm viability levels and high sperm motility rates, to be drone mother breeding stock. In conjunction with selection of drone mother stock through breeding programs is the development of management practices for drone mother colonies to produce maximum numbers of drones surviving to a high functional age. ix

Key Findings (i) Drone ages, seasons and breeding lines were identified when drone semen was of a higher quality. No drone age, season or breeding line was found when drones were identified as not being suitable for mating with queen bees based on semen quality. The significant differences identified for age, season and breeding line suggest the question of whether there is a time period for drones during one or more seasons each year when sperm quality is at its highest level, and if so, this would question the mating quality of queen bees mating with drones out of the peak sperm quality time period. The average number of sperm produced per drone, in general, is low compared with published data from Australia, Europe and the USA, based on methodology used in this project. The average volume of semen produced per drone was comparable with published data from Australia, Europe and the USA. Of the four commercial breeding lines examined for semen volume produced per drone and number of sperm produced per drone, one line was constantly superior when compared with the other three lines suggesting that semen volume and sperm number per drone may be selectable traits. Results suggest that should volume of semen and number of sperm produced per drone be selectable traits then recessive genes may be involved. The proportion of mature age drones which did not produce semen at the endophallus after manual eversion was high which suggested further investigation into the efficiency of manual eversion as a method for obtaining semen samples. Drone survival rate was high for 14 and 21 day old drones with low numbers of drones surviving to 35 days of age. Sperm viability percent and sperm motility rating were at the lower end but comparable with published data, with results suggesting that sperm viability percent and sperm motility rating may be improved through selection in a drone mother breeding program. In a 2006 experiment, sperm was found present in the seminal vesicles of mature age drones which had not released semen after manual eversion. A 2007 experiment based on semen release after manual eversion suggested that drone maturation occurs over a wide range of ages with drones 30 days old retaining sperm in the seminal vesicles after manual eversion. In a 2007 experiment, low sperm viability and motility was not found to be a contributing factor for sperm remaining in the seminal vesicles of mature age drones after manual eversion. Manual eversion was not found to be a suitable method for collecting semen samples required to provide accurate data on semen volume and number of sperm produced per drone. Collecting semen/sperm samples from drone seminal vesicles was considered a more reliable method for providing accurate data on semen volume and sperm number per drone. x

(ii)

(iii)

(iv) (v)

(vi)

(vii) (viii)

(ix) (x)

(xi)

(xii)

(xiii)

(xiv)

Amino and Fatty Acids. Eighteen free amino acids were identified in drone semen with high levels of arginine and proline and the remaining 16 amino acids present at levels less than 7% of the total amino acids in each sample. Thirty two fatty acids were identified with high levels of oleic acid and elaidic acid and levels of less than 7% for the remaining fatty acids present for the total percent in each sample.

Implications (i) Low numbers of sperm produced per drone. This characteristic has its principal economic effect at commercial queen mating apiaries when a large number of queen bees are on mating flights at the same time over a short period of time. Low numbers of sperm present in drones can be expected to result in low numbers of sperm present in the spermathecae of queen bees on completion of mating with the flow-on effects of poor performance and early supercedure of those queen bees. Selectable traits. Should semen volume per drone and sperm numbers per drone be selectable traits as data from this project suggests, then selection of queen bees as drone mothers whose drones produce large semen volume and high sperm numbers in breeding programmes would be of economic benefit to the commercial beekeeping industry. Such drones would provide queen bees with high sperm numbers in their spermathecae after mating, with those queen bees then expected to have an increased working life period in the hive before being superseded. Low longevity of drones. Data suggests that, in general, drones have a low longevity which is of economic importance in the context of reducing numbers of mature age drones available to mate with queen bees at commercial queen bee mating apiaries. Further research is required on rearing and maintaining large numbers of drones with high sperm quality to a high age for commercial queen bee mating purposes. Improvement of sperm viability and motility levels. Data suggest that sperm viability percent and sperm motility levels are able to be improved by selecting for these characteristics in drones from drone queen mothers in breeding programs. Economic benefits to commercial beekeeping from drones producing sperm with a high viability percent and high motility levels would result from increased egg fertilisation levels from the limited number of sperm each queen has stored in her spermatheca. Drone manual eversion effectiveness. Because manual eversion does not result in all seminal fluid and sperm from the seminal vesicles being present in semen at the endophallus for all drones then it is not a suitable method for collecting semen samples where a high level of accuracy is required, e.g. for evaluating semen volume and number of sperm produced by an individual drone when evaluating these characteristics in drones from queen bees being selected as drone mother queens in breeding programs. Data suggests that dissection and counting of sperm numbers from the seminal vesicles provides a more reliable alternative than manual eversion for sperm sample collection. Fatty and amino acids. Amounts of fatty acids and amino acids in drone semen change with age and season with the implication of change in semen quality with semen from one age or season being of a higher quality than semen from drones of a different age or at a different season. As a minimum dietary requirement, results indicate that drone rearing colonies require access to the 18 amino acids and 32 fatty acids identified in this experiment with emphasis on the 2 amino acids and 2 fatty acids present at high levels.

(ii)

(iii)

(iv)

(v)

(vi)

xi

3. Recommendations (i) Standard methods require development for obtaining a semen samples from a drone for examination, and for the determination of measurable data from each drone sampled for the volume of semen produced, the number of sperm produced, sperm viability percent and sperm motility rating. Using standard methods drones from a range of commercial honeybee breeding lines in Australia to be surveyed for the four characteristics, volume of semen, number of sperm, sperm viability and sperm motility, to obtain a wider view of drone sperm quality within breeding stock available within the commercial beekeeping industry. That the four drone characteristics, semen volume, sperm number, sperm viability and sperm motility be incorporated into standard selection criteria for drone mother queen bees in breeding programs. Apiary management practices are required to be developed to promote the production and maintenance of large numbers of drones with high longevity and producing sperm of high quality to a high age, from drone mother hives at queen bee mating apiaries. Fatty and Amino acids. Research is required into drone dietary requirements and to determine benefits from supplementary feeding with arginine and proline amino acids and oleic and elaidic fatty acids, in drone rearing colonies at queen bee mating apiaries.

(ii)

(iii)

(iv)

(v)

xii

1. Introduction
The objectives of this project were to examine the production and quality of semen from drone honeybees Apis mellifera from different commercial breeding lines, at different ages and at different seasons during the commercial queen bee production period in eastern Australia between spring and the following autumn. The project examines whether problems can occur with drone breeding lines, drone age groups, or times during each season when the production and/or quality of drone semen is not satisfactory for drones being reared in drone mother colonies being managed with commercial queen bee mating apiaries for the purpose of mating with queen bees. Chemical analysis of drone semen was examined to determine changes in composition between drone age and between seasons on the basis that changes found may impact on mating performance of those drones. The reason for developing this project is based on results from two RIRDC funded projects DAN-164A (Rhodes, 1998) and DAN 182A (Rhodes, 2002) carried out to investigate introduction and early performance success of commercially reared queen bees. Failure of newly introduced queen bees to be accepted or to perform well immediately following introduction often results in a heavy economic loss for the beekeeper. A bee colony with a nonaccepted or poorly performing introduced queen may become weakened in strength and difficult to requeen resulting in a substantial loss of production from that colony. Rhodes (2002) found low numbers of sperm in the spermathecae of newly mated queen bees may be contributing to unsatisfactory performance by those queens. Physical characteristics of the queens examined for the project (e.g. queen weight, number of ovarioles and spermatheca diameter indicated that the queens were of a high standard). This suggested that low sperm counts in the spermathecae of the test queens may have resulted from a problem with drones present at the mating apiary (e.g. low numbers or low quality of drones or a combination of both factors). In a natural situation, bee colonies tend to produce virgin queens at times of swarming and supersedure. It can be expected that the production of suitable numbers of mature and high quality drones would coincide with times virgin queens are produced. Free and Williams (1975), in the northern hemisphere, found drone eggs normally first appear in a colony in spring reaching maximum numbers by early summer and cease to be laid by early autumn. During commercial queen bee production in eastern Australia, queen bees are produced by beekeepers by colony manipulation over the period between August and April. There is little available data on the numbers and quality of drones available for mating or if differences occur during this extended time period. Data from this project examine whether drones produced throughout the complete August to April season are suitable for mating purposes or whether there are times during this period when they are not suitable. Root (1890) reported the first recorded sighting of a drone mating with a queen bee, that a drone develops from an unfertilized egg and worker bees from fertilised eggs, sperm are stored in the queens spermatheca and remain viable for a number of years and that a queen bee, once mated, is able to lay fertilised or unfertilised eggs. Root discusses means of rearing large numbers of suitable drones for mating with queen bees and means of restraining undesirable drones. Beekeepers have had this basic knowledge available to them at least since Roots book was published. A publication of major practical benefit to honeybee researchers and bee breeders was Bishop (1920) who identified that the extrusion of the drones endophallus caused by artificial means, manual eversion, did not necessarily duplicate the natural act of copulation, even when it seemed to do so. This question continues to remain unanswered and casts doubt over the accuracy of data obtained 1

using manual eversion as the means for obtaining semen samples. Bishop reported that sperm developed in the testes during the drone pupal stage. The reproductive organs and mating procedure of the drone were described by Moritz (1989). The reproductive organs comprise the endophallus (copulatory organ), testes, seminal vesicles and the mucous glands. The endophallus is a long, soft membranous duct which lies inside the abdomen with three distinguishing zones the vestibulum, the cervix and the bulb. Spermatozoa start migrating from the testes to the seminal vesicles when drones are 2-3 days old. They attach their head to the gland cells of the wall and undergo a secondary physiological process of maturation. At the same time, the gland cells empty their contents into the vesicle among the spermatozoa. The mucous glands are at the rear of the seminal vesicles and open into the seminal duct. The inner wall consists of secretory epithelial cells which secrete the white mucous substance. Secretion of mucous starts immediately after emergence of the drone and is completed by about day 5. Semen consists of two components from different sources, sperm from the testes which are about 0.25 mm long, and seminal fluid from the seminal vesicles. The age at which drones mature and are able to mate with queen bees is not known with accuracy due to difficulties with obtaining this data under field conditions since queen bees mate on the wing a number of kilometres from their colony. Mating age of queen bees has been shown to be affected by climatic conditions (Ruttner, 1956) which could also be expected to affect drone mating age. Other factors such as nutritional conditions during drone rearing could be expected to affect drone mating age (van Niem Nguyen, 1995). A number of authors have provided data on the age at which drones mature, i.e. sperm are in the seminal vesicles and the mucous glands have completed development and are full, however, this physiological development in itself is not sufficient to ensure that the drone is physically capable of mating with a queen bee. Bishop (1920) considered drones to be mature after the fifth or sixth day with the added statement that maturity becomes accentuated up to the age of nine or ten days with slight morphological and histological changes after the sixth day. Mackensen and Roberts (1948) stated drones become sexually mature at 8 days of age with the maximum number of sperm accumulated in the seminal vesicles. Moritz (1989) stated drones twelve days old are mature and their sperm may be used for inseminating queen bees. van Niem Nguyen (1995) found drones in hives fed improved protein nutrition reached sexual maturity as early as ten days after emergence with all being mature by seventeen days, while drones in hives not fed supplementary pollen commenced reaching sexual maturity at twelve days with all reaching sexual maturity by eighteen days. van Niem Nguyen defined sexual maturity by the presence of semen on the tip of the endophallus following manual eversion of drones of a known age. Ruttner (1956) measured the effects of temperature, cloud cover, wind and radiation at the time of mating on mating success and concluded that such factors could delay mating flights between the first and last mating flight by 5-15 and up to 24 days. Although not stated by Ruttner and with no time period stated between when the queen first mated and last mated, this data suggests that queen bees are able to mate successfully over an extended time period with no negative effects on their ability to continue to mate. Ruttner also stated that queen bees receiving too little semen at mating were short lived queens. The effects of inbreeding on brood survival were examined by Woyke (1963) who observed that after two or three generations of brother-sister matings, queens laid fertilised eggs in worker cells with approximately 50% of the hatching larvae being diploid males. The diploid male larvae were destroyed by nurse bees with only the approximate 50% of hatching female larvae surviving. A queen bee heading such a hive would be classified as failing and would not be expected to survive over an extended time period, either being physically replaced by a beekeeper in a managed colony or superseded by worker bees in a feral colony. Practical aspects of drone/queen bee matings were summarised by Loper (1993). United States researchers in the early 1960s identified areas where drone bees congregated, termed drone congregation areas (DCAs), where drones responded to the queen sex pheromone (9-oxodec-trans-22

enoic acid) 6 to 30 metres above ground while few or no drones were present in other seemingly similar areas. Further research showed better mating success of queen bees was obtained when colonies containing drones were placed 2.5 km from the colonies containing the queen bees, rather than in the same apiary. Thus, DCAs seem to be re-orientation areas where drone numbers increase with drones flying higher and possibly seeking out more cues to follow allowing them to come into contact with queen bees on mating flights which are attracted to the same areas. This mating process has implications for sufficient numbers of drones of an age able to mate with a queen bee and containing sufficient numbers of sperm to be present at a DCA when large numbers of queen bees are present. The principal sources of data on which outcomes from this research were determined were based on differences between semen quality between the factors being examined breeding lines effects, seasonal effects and drone age effects. Measurement of semen quality per drone for this research is based on semen volume, sperm numbers produced, sperm viability, sperm motility, and a comparison of differences between the amino acids and fatty acids of semen sampled over a range of drone ages and seasons. Hunter and Birkhead (2002), when examining sperm viability and sperm competition in insects concluded that, all else being equal, males will vary in their ability to fertilize ova on the basis of sperm viability alone with their results suggesting that sperm viability is one of a suite of male adaptations to sperm competition in insects. Male fitness of honeybee colonies was examined by Kraus et al. (2003). In social insects, such as honeybees, female reproductive success is determined through the number of surviving reproductive colonies (swarms) produced and is relatively easily measured. Determination of the reproductive success of males is difficult since males are usually short lived and matings are difficult to observe. The male reproductive success of a whole colony is not able to be determined by the number of males produced as drones of one colony may out-compete others in mating efficiency. Kraus et al. (2003) found clear evidence for an extensive diversity in male mating success at the colony level with a significant positive correlation between the number of successfully mating drones per colony and the individual siring success of drones of these colonies. It has been assumed that the number of drones rather than their individual mating success was most significant for colony male success (Baudry et al., 1998, quoted in Kraus et al., 2003), however Kraus et al. consider that not only the number of mating males but also the individual siring success of a drone is determined by the colony and/or the genotype of the mother queen, and selection through the male side appears to be an extremely important factor for colony fitness. Woyke (1973) showed that the amount of semen produced per drone is highly variable and Kraus et al. (2003) stated that even when a high amount of sperm is transferred into the spermatheca of a queen bee, this does not necessarily imply that many offspring workers are sired, concluding that male reproductive success appears to be a major driver of natural selection in honeybees.

1.1 References
Baudry, E., Solignac, M., Garnery, L., Gries, M., Cornuet, J.M. and Koeniger, N. 1998 Relatedness among honeybees (Apis mellifera) of a drone congregation, Proceedings of the Royal Society of London B., 26, 2009-2014. Bishop, G.H. 1920 Fertilization in the honey-bee, The Journal of Experimental Zoology, 31 (2), 225266. Free, J.B. and Williams, I.H. 1975 Factors determining the rearing and rejection of drones by the honeybee colony, Animal Behaviour, 23, 650-675. Hunter, F.M. and Birkhead, T.R. 2002 Sperm viability and sperm competition in insects, Current Biology, 12, 121-123. Kraus, F.B., Neumann, P., Scharpenberg, H., Van Praagh, J. and Moritz, R.F.A. 2003 Male fitness of honeybee colonies (Apis mellifera L.), Journal of Evolutionary Biology, 16, 914-920. Loper, G.M. 1993 What do we really know about drone flight behaviour ?, Bee World, 73, 198-203. Mackensen. O. and Roberts, W.C. 1948 A manual for the artificial insemination of queen bees, USDA, Bureau of Entomology and Plant Quarantine, ET-250, 1-33. Moritz, R.F.A. (Ed.) 1989 The instrumental insemination of the queen bee, Apimondia Publishing House, Bucharest, Romania, 22-28. Rhodes, J.W. 1998 Introduction and early performance of queen bees:Introductory apiary status and post introduction results, RIRDC Report, Rural Industries Research and Development Corporation, Canberra, ACT, Australia, 23 pp. Rhodes, J.W. 2002 Introduction and early performance of queen bees some factors affecting success, RIRDC Report, Rural Industries Research and Development Corporation, Canberra, ACT, Australia, 49 pp. Root, A.I. 1890 ABC of Bee Culture, Medina, Ohio, USA. Ruttner, F. 1956 The mating of the honeybee, Bee World, 3, 2-15 and 23-24. van Niem Nguyen, 1995 Effects of protein nutrition and pollen supplementation of honeybee (Apis mellifera L.) colonies on characteristics of drones with particular reference to sexual maturity, MSc (Honours) Thesis, University of Western Sydney, Hawkesbury, NSW, Australia. Woyke, J. 1963 Drone larvae from fertilised eggs of the honeybee, Journal of Apicultural Research, 2, 19-24. Woyke, J. 1973 Reproductive organs of haploid and diploid drones, Journal of Apicultural Research, 12, 1-18.

2. Determination of drone sample size for drone sperm numbers experiment.

2.1 Introduction
At the commencement of this project when developing methodology for collecting data it was necessary to determine a minimum sample size of the number of drones to be examined to provide statistically sound data on the number of sperm produced per drone.

2.2 Method
A commercial beekeeper, Beekeeper A, supplied 100 drones which were marked on emergence and caught at 16 days of age when they were considered to have matured. A second commercial beekeeper, Beekeeper B, provided 7 drones which were examined at 16 days of age. The number of sperm produced by each drone was determined by manually everting the drone and collecting semen produced on the endophallus in a syringe tip. The complete volume of semen was diluted in a constant volume of buffer. The number of sperm present in each sample was determined using an Improved Neubauer Haemocytometer, depth 0.1mm, volume 1/400 mm2 , by counting the number of sperm present in the 16 small squares in each of the 4 corners and the centre group of 16 small squares (total 5x16 small squares) at one end of the haemocytometer. Fresh samples of diluted semen were placed on each end of the haemocytometer 5 times for each drone providing 10 (5x2) counts for each drone. The number of sperm/drone was then determined from the formula:
Number of sperm/drone (million) = total number of sperm counted in the 10 cells x 25 x 10 000

2.3 Statistical method

Sample sizes were calculated using Dobson (1984).

2.4 Results
Sixty drones from Beekeeper A and 6 drones from Beekeeper B provided semen. A comparison between results from drones from Beekeeper A and Beekeeper B found no significant difference, (P<0.05), between the two samples. There was a large standard error for Beekeeper B because of the small sample size. The standard errors provided a guide to the sample size required.
Beekeeper A, sample size 60 drones. Mean 2.59 x 106 sperm per drone, s.e. 0.04 Beekeeper B, sample size 6 drones. Mean 2.09 x 106 sperm per drone, s.e. 0.29

Table 1. Sample sizes with their S.E. and LSD providing a 95% confidence level.
Sample Size Number of Drones 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Standard Error 1.10 0.90 0.78 0.70 0.64 0.59 0.55 0.52 0.49 0.47 0.45 0.43 0.42 0.40 0.39 0.38 0.37 0.36 0.35 Least Significant Difference 3.47 2.71 2.30 2.03 1.84 1.69 1.57 1.48 1.40 1.33 1.27 1.22 1.17 1.13 1.10 1.06 1.03 1.00 0.98

2.5 Conclusions
A sample size of 30 drones was chosen for this project as it provided the lowest number of drones to be examined with the required level of accuracy. A sample size of 30 drones requires a difference between treatments of 1.84 million sperm for it to be significant and a sample size of 50 drones requires a difference between treatments of 1.40 million sperm.

2.6 References
Dobson, A.J. 1984 Calculating sample size, Trans. Menzies Foundation, 7, 75-79.

3. Volume of semen and number of sperm per drone.

3.1 Introduction
Honeybee sperm are 230 m long cells consisting of an 8 m head with an acrosome on its end and a long tail. The tail contains an axoneme and two different sized accompanying mitochondrial derivatives. Spermatogenesis takes place in the testes of the haploid drone during its larval and pupal life (Hoage and Kessel, 1968, quoted in Pabst and Pfeiler, 1994). Rhodes (2002) obtained the numbers of sperm present in the spermathecae of 260 Apis mellifera queen bees examined between 14 and 35 days of age over three years with 56.6% containing < 3 million, 24.4% containing 3-4.5 million, and 19.0% containing > 4.5 million sperm/queen, range 0.25-5.21 million sperm/queen. Mackensen and Roberts (1948), (quoted in Ruttner, 1956) stated that the full spermatheca of a naturally mated queen holds, on average, 5.73 million spermatozoa. Taber (quoted in Jay and Dixon, 1984) stated that queen bees with sperm counts of less than 3 million are unable to head commercial honey production colonies for one season. The two factors from Rhodes (2002) of overall low sperm counts per queen and a wide range of sperm counts between queens suggest a problem during queen bee mating. The problem could be insufficient numbers of mature age drones, inadequate quality of drones, or a combination of both factors for drones present in the mating area when the queens were mating. These results have led to this investigation of sperm production by breeding lines of A. mellifera used in commercial beekeeping practices in eastern Australia. Of interest for this work are the results from researchers examining A. mellifera drones which provide data on the number of sperm per drone and the volume of semen per drone based on (i) the sample being obtained either from manual eversion of drones, or from dissection of the seminal vesicles from drones, and (ii) the method used for determining the number of sperm present per sample. Mackensen and Roberts (1948) described their methodology for Estimating the number of sperms present either from a queen bees spermatheca or from a drones seminal vesicles by diluting each sample in either 5cc or 10cc of tap water and counting the number of sperm present in the known volume, 0.8 mm3 , contained in a counting chamber (haemocytometer). They conceded that there is considerable chance of error in making counts from such a small sample and that greater accuracy can be obtained by counting a larger sample. Mackensen (1955), when experimenting with Artificial Insemination techniques of queen bees collected sperm directly from seminal vesicles removed from drones. Mackensen reported drones 7-8 days of age produced an average of 9.89 (range 8.36-10.63) million sperm per drone, n = 6. Khler (1955, quoted in Ruttner, 1956), found the average count of spermatozoa in a drone to be 4.5 million sperm, Jaycox (1961) considered this number to be abnormally low, however the method of collecting the sperm sample was not presented. Woyke (1962) found drones yielded an average of 1.7 mm3 (range 1.5-1.75 mm3 ) of semen with each drone producing, on average, 11.0 million spermatozoa, n = 78. Woyke refers to the ejaculate which suggests the samples were obtained from manual eversion of drones. Of the above authors, Woyke (1962) did not state the methodology used for calculating sperm numbers per drone. Mackensen (1955) and Ruttner (1956) cited Mackensen and Roberts (1948) in their References although they did not state the methodology used for counting sperm numbers.

Jaycox (1961) examined the age at which drones matured by examining the effects of various foods and temperature on sexual maturity. The definition of mature remains unclear with the difference between the ability of a drone to produce semen following manual eversion and the unknown ability of that drone to mate with a queen bee under natural mating conditions. Drone maturity is investigated further in Section 9. Associated with drone maturity based on drone age and semen production has been added the influence of drone flight as a contributing factor in drone maturity. Drone flight is of interest to this work as it suggests differences in sexual maturity characteristics may occur between drones allowed free flight prior to examination for semen production compared with drones confined within a cage or within the hive prior to examination for semen production. The suggestion by Kurennoi (1953, quoted in Jaycox, 1961) that flight provides reflex or mechanical stimulation with Kurennoi concluding that the transfer of sperm to the seminal vesicles must be followed by a period of regular flights to produce maturity. Brunnich (1927, quoted in Jaycox, 1961) suggested that flight may be a factor in overcoming retarded maturation by raising the drones body temperature, he reported drone body temperatures of up to 48C compared with foraging worker bee temperatures of 44C found by Schultz-Langner (1958, quoted in Jaycox, 1961). Jaycox (1961) dissected the seminal vesicles and vasa deferentia from drones examined at one-day increases in age between 3 and 11 days of age, the number of sperm per drone was determined using the Mackenson and Roberts (1948) technique modified by placing the sperm sample in 1 ml of a 2% starch solution. Drones sampled included drones allowed free flight from emergence, and drones caged at emergence in a nursery colony. There was no significant difference between sperm counts between drones allowed free flight and confined drones for drones of specific ages, drones 8-9 days old produced average sperm counts per drone of 10.76 million for free flying drones and 11.21 million for confined drones. A continuation of the examination showed there was no decrease in numbers of sperm from confined drones up to 6 weeks of age. When comparing characteristics between Africanized and European race drones, Rinderer et al. (1985) dissected one seminal vesicle per drone and counted the number of sperm present using a haemocytometer. The result recorded for one seminal vesicle for European race drones was 5.7 0.9 million sperm, n = 80. Moritz (1986) when examining sperm competition in honeybees found there was an unequal contribution of drones to the offspring of a honeybee queen, while Moritz was unclear on the reason for this he quoted Ruttner (1976) who had reported a considerable inter drone variation concerning the number of sperms per drone suggesting that drones with a large number of sperms may be genetically more successful than others. van Niem Nguyen (1995) examined the effects of dietary supplements on drone maturity and sperm production. Drones were confined until examined. Drones were manually everted and the volume of semen produced recorded in a graduated syringe, semen was diluted in 1.0 ml Tris buffer solution and 9 ml water added. Sperm counts were made using a haemocytometer Counting Chamber B.S.748. The number of spermatozoa in 10 squares (each square contained 16 smaller squares) of the haemocytometer field was counted. The number of sperm/drone was calculated according to the formula:
Number sperm / drone = number of sperm in 10 counted squares x 25 x 10 000 Where : 25 = The volume of semen solution in 10 counted squares is 1/25 mm3 , so one mm3 of solution is multiplied by 25 10 000 = Volume (in mm3) of 10 ml of semen solution.

van Niem Nguyen (1995) found semen volumes and sperm counts from drones from colonies fed limited field pollen (n = 33) were mean 0.91 (range 0.80-1.00) l and mean 7.14 (range 5.75-8.45) million sperm per drone, compared with drones from colonies fed supplementary pollen (n = 44), mean 0.93 (range 0.82-1.00) l and mean 8.01 (range 6.75-9.50) million sperm per drone.

Rinderer et al. (1999) collected data on sperm numbers from 12 day old drones in free flying control hives from 30 drones per sample by dissecting one seminal vesicle from each drone, macerating in 10 ml of 0.5% saline solution and using a haemocytometer to count sperm numbers. For one seminal vesicle a sperm count of 4.254 0.424 million sperm per drone was recorded. Collins and Pettis (2001) stated that a healthy drone will produce up to 1.25 l of semen with close to 10 million sperm with the majority of sperm present being alive. Drones not infested with Varroa mites (Varroa destructor) were allowed free flight until examined at 12 days of age. Sixteen drones out of the 40 examined (40%) produced semen using manual eversion, the authors stated that artificial ejaculation of drones is not completely efficient. Semen volume was determined using a Gilmont micrometer syringe with a digital scale and number of sperm by diluting a 0.1 l fraction of the semen from each drone in 250 l of a 3% saline solution and the number of sperm present counted in a counting chamber. Results from Collins and Pettis (2001) were, volume of semen per drone, mean 0.946, range 0.48-1.67 l, concentration of sperm per drone mean 9.15, range 0.5-29.25 million sperm per l, or a mean of 8.656 million sperm per drone. They considered the large variation found for concentration of sperm in the semen was influenced by the semen being viscous and sticky with the sperm cells tending to clump together after dilution providing a considerable error of measurement at this point. Duay et al. (2002) considered the number of sperm produced by a drone of paramount importance to male fitness. A. m. carnica colonies provided 68 control drones which were confined in their hives until examined at 12 days of age. Sperm numbers were determined by dissecting two seminal vesicles in ethylene glycol diluted in distilled water with sperm numbers counted using a Neubauer cell chamber. Results of number of sperm per drone were mean 7.475 2.813 million sperm. Schlns et al. (2003) compared the numbers of sperm produced by drones reared in drone cells with smaller body sized drones reared in worker bee cells. They found small drones produced significantly fewer sperm (7.5 0.5 million) than normal sized drones (11.9 1.0 million). Drones were allowed free flight during maturation, sperm counts were made by dissecting the seminal vesicles from drones. The number of sperm in one seminal vesicle was counted in 0.5 ml Hyes ringer solution further diluted in distilled water to 5 ml total volume, counting was in a Thoma counting chamber, total volume 0.064 l. Sixteen replicate fields were counted and the average for the 16 sperm counts was used as the best estimate sperm count for that drone. Part of the report by Anderson (2004) examined the effects of dietary supplements fed to dronerearing colonies on sperm numbers produced by drones. Drones were confined in a super above a queen excluder, at 20 days of age more than 95% did not produce semen after manual eversion. Drones were then examined at 27 days of age with semen collected by manual eversion. Endophalli of 10 drones per sample of everted drones were placed individually in 0.25 ml insect ringers solution and stored at -20C until examined. Sperm numbers were determined by breaking each endophallus apart in 1:80 or 1:160 distilled water and sperm numbers counted using a haemocytometer. Drones from control colonies not fed dietary supplements produced 3.19 2.37 million sperm per drone and drones from colonies fed dietary supplements produced fewer numbers of sperm. Phiancharoen et al. (2004) inseminated a number of Apis mellifera queens with sperm from four species of Apis including A. mellifera sperm and determined egg fertilization and survival. Seminal vesicles were dissected from drones collected during mating flight times and stored in Tris buffer. Sperm from the seminal vesicles were dispersed in the buffer solution and further diluted with 20 ml of distilled water. The sperm were counted with a Fuchs-Rosenthal haemocytometer. The number of sperm recorded for A. mellifera drones was 7.6 1.47 million sperm per drone, n = 10. In their paper discussing variation in sperm numbers produced by drones from different Apis species, Koeniger et al. (2005) reported sperm numbers from A. mellifera drones recorded by other authors Woyke (1960), A. m. (mellifera?), 11-12 million sperm per drone; Elbassiouny (1992), A. m. carnica, 9

10.2 0.3 (8.1-12.4) million; Moritz (1981) A. m. carnica, 8.5 million; Berg (1990, 1992), A. m. carnica, 7.1 1.9 (3.9-11.71) million; and Koeniger (2002, unpublished), A. m. carnica, 8.3 1.1 million. Koeniger et al. (2005) stated that, in general, sperm numbers of individual drones show high variance and suggested this could be due to errors in the method used. They identified two critical sources for inaccuracy (i) upon dissection some sperm may be pushed into the endophallus, (ii) (upon dilution) sperm sink to the bottom and there is a gradient of sperm concentration within the solution and suggested that thorough dispersal before taking a sample is crucial. They commented that normally the number of tested samples per drone or the number of units evaluated is not published with only average numbers of spermatozoa being given. Koeniger et al. (2005) concluded that it is premature to draw general conclusions from the differences in number of sperm of drones and that the reason for these variations have to be understood. Schlns et al. (2004) using microsatellite fingerprinting determined the contribution of sperm in the progeny of 8 queen bees inseminated with either 0.5 or 1.0 l of semen from 6 or 7 drones showed no significant effect of the insemination sequence but a strong impact of the semen volume of a drone on the frequency of his worker offspring in the colony. The importance of drones producing high numbers of sperm and the ability of researchers and bee breeders in being able to identify those drones is emphasized by Schlns et al. (2004) who found a high correlation between a drones number of sperm and the percentage of its sperm reaching the spermatheca after instrumental insemination suggesting that individual differences in number of sperm per drone have an influence on the drones paternity success.

10

3.2 Methods
(a) Field Collection of Data Field work for this experiment involved rearing and managing drones to a known age for examination and was carried out between October, 2003 and May, 2004. A 16 hive apiary was used to ensure sufficient numbers of drones and was established to investigate differences between (i) Breeding lines Due to a lack of a recorded pedigree system within commercial bee breeding programs, commercial producers of queen bees who supply the beekeeping industry in Victoria, New South Wales and Queensland were questioned with regard to the sources of their breeding stock. Four queen bee breeders were selected to provide queen bees to the project based on the unrelatedness of their breeding stock. Each queen bee breeder supplied 4 open mated sister queens from an instrumentally inseminated breeder queen mother of the Italian race and considered to be predominantly Apis mellifera ligustica. Bee hives were provided from the University of Western Sydney research apiary at Richmond, New South Wales. (ii) Drones reared during different seasons Drones were reared to represent spring drones (eggs laid 6/7.10.2003), summer drones (eggs laid 7/8.1.2004), and autumn drones (eggs laid 15/16.3.2004). (iii) Drones of different ages For each breeding line and for each season drones were sampled at 3 ages 14, 21 and 35 days of age. (iv) Nutrition To reduce differences in results from (i), (ii) and (iii) due to nutritional effects, the apiary was managed so that all hives were at the same apiary location at all times to provide consistency in foraging conditions for nectar and pollen. When nectar was observed by the apiary manager not to be in abundant supply all hives were supplied with an internal hive feeder containing sugar syrup, white sugar and water in equal volumes. Similarly, when fresh pollen being brought into the hives by foraging bees was observed not to be abundant, all hives were supplied internally with irradiated pollen sourced from Western Australia (Eucalyptus spp.) or from NSW (Canola pollen). The method used to produce numbers of drones of a known age from a known queen bee was based on methodology developed at the University of Western Sydney for previous drone related studies (Duncan, 2003). The queen bee used to produce the drones was placed on a frame (full-depth, Langstroth) of drone comb inside a frame cage constructed from wire queen excluder with a bee-proof metal lid, the cage was then placed in the middle of the brood chamber of the hive the queen came from, the two outside brood frames were removed temporarily to provide room for the cage. Hives used were 10 frame, full depth, Langstroth, containing 9 frames in the brood chamber and super. The hive was supplied with irradiated pollen and sugar syrup. Forty-eight hours after being placed in the cage the queen was removed from the cage and released into the hive, the frame of drone comb containing eggs was removed from the cage and placed in the centre of the brood chamber, the two frames removed to provide space for the cage were replaced. With the exception of periods of time when there was an abundance of nectar or pollen being brought into the hives by foraging bees, all hives were supplied continuously with pollen and sugar syrup. Eighteen days after the frame of eggs was removed from the cage the frame was replaced in the cage which was placed in the centre of the brood chamber. Drones began emerging from the sealed drone cells 2 to 3 days later. The frame of emerging drones and sealed drone cells were removed from the cage, taken to a shady area in the apiary and the newly emerged drones marked on the thorax with a water based Posca marking pen and returned to the hive in which they were reared. Marked drones were of a known age with an error of +24 hours.

11

Prior to marking, a wire queen excluder was placed between the bottom-board and the brood chamber and a plastic queen excluder was placed between the brood chamber and super of the hive to confine the marked drones within the brood chamber. New hive materials were used and the hive was strapped tightly with a metal strap to prevent spaces which allowed drones to escape. Earlier pilot experiments where drones were not confined to their hives resulted in insufficient numbers of drones being available at the ages required to complete the examinations. Collection of drone samples. Adult marked drones were collected from their colonies at the ages required using methodology used during drone collection for artificial insemination purposes by G. Wheen (2003). Hives containing drones were opened carefully early in the morning when drones were relatively inactive. The queen excluder covering the brood chamber was slid aside allowing the removal of one frame at a time, when one frame had been removed the brood chamber was covered with the queen excluder. The required number of marked drones was caught and placed in a rectangular cage constructed of wooden ends with one queen excluder side and the other side of a removable Perspex sheet with air holes. On one long end was a drilled hole of sufficient diameter to allow a drone to be passed through it during catching, the hole was covered by a sliding metal lid. The size of each cage was 16 x 10 x 3 cm. Each cage safely held 40-50 drones. A small piece of queen candy was placed in each cage to support the drones when they did not have access to nurse bees. When a cage full of drones had been caught, 40 drones, the cage with drones was placed into a previously prepared support hive to maintain the drones in a healthy condition until required for examination. The support hive was a strong double hive with one outside frame removed from the super and two middle frames from the super replaced by two frames of unsealed brood from the brood chamber. The two frames of unsealed brood were placed in the middle of the super with a space between them. Four cages of drones could then be placed between the frames of unsealed brood with nurse bees from the hive attending to the drones in the cages. The support hive maintained the drones for a number of hours until the cage of drones was removed from the hive and taken to the nearby laboratory for examination. In the laboratory the cage of drones was released into a flight cage, 40 x 30 x 30 cm with one end covered in flywire gauze and facing a table lamp. Drones were attracted to fly to the end with the light and defaecate during flight. Inside the flight box was placed a cage 1 6x 10 x 3 cm with 3mm gauze mesh on one side and containing about 100 nurse bees collected each morning from frames of unsealed brood, and a large plug of queen candy. The cage was placed gauze side up, drones released into the flight cage were attracted to settle on the gauze and were fed by the nurse bees in the cage providing active drones through the period of examination. (b) Laboratory Collection of Data Sample size. For all measurements carried out a sample size of 30 drones was used with the exception when 30 drones were not available then the number of drones available up to 30 drones was used. Semen volume. Drones were caught from the flight cage and were stimulated to ejaculate using manual eversion which involved putting vertical pressure on the thorax with the thumb and first finger of one hand and then placing horizontal pressure on the abdomen with the thumb and first finger of the second hand. Pressure commenced at the anterior end of the abdomen and proceeded towards the posterior end. Semen was collected using an Instrumental Insemination device fitted with a Harbo calibrated micrometer syringe marked with divisions of 0.2 l. The volume of semen produced by each drone was recorded to the nearest 0.1 l. Number of sperm. The number of sperm produced per drone was calculated by diluting 0.6 l semen in 0.5 ml Tris buffer, mixing, and adding a further 1.0 ml Tris buffer and mixing, i.e. the dilution rate was 0.6 l semen in 1.5 ml Tris buffer. The dilution rate was calculated as:
1.5 x 10-3 divided by 0.6 x 10-6 = 2500

12

The haemocytometer used for counting sperm numbers was an Improved Neubauer, BS.748, depth 0.1 mm, 1/400 mm2. For each drone, the number of sperm present in five squares, each square containing sixteen smaller squares, at one end of the haemocytometer was counted. This procedure was repeated five times for each drone.
Number of sperm per drone (million) = Total number of sperm counted in 25 squares x semen volume (l) x 25000 (dilution rate x 10)

Tris Buffer formula:

Sodium Chloride Glucose L (+) Arginine - HCl L (+) Lysine Tris (hydroxymethyl) aminomethane (Base 7-9) Tris (hydroxymethyl) aminomethane hydrochloride In 1000 cc distilled water (Provides pH 8.7) 11.0 g 1.0 g 0.1 g 0.1 g 4.9 g 1.5 g

3.3 Statistical analysis

Since there was a large percentage of drones with no sperm recorded, the presence of sperm was analysed using contingency tables to determine whether the proportion of drones with sperm recorded was independent of season, drone age, or breeding line. The Chi squared test for independence of factors has 2 degrees of freedom (d.f.) for drone age and season, and 3 d.f. for breeding line. Of the 848 drones examined there were 344 (41%) drones with no sperm recorded. Semen volume and number of sperm per drone for those drones with semen present were analysed as a linear model with season, drone age, breeding line and all interactions included in the model. Predictions for each factor were averaged over all other factors in the model.

3.4 Results
All drones caught from each breeding line, for each age group, for each season were examined. Drones providing no semen or a small amount of semen not sufficient to be measured after manual eversion were recorded as no semen produced. The proportion of drones which recorded a measurable volume of semen compared with the total number of drones examined for each sample is shown in Table 1, a total of 848 drones were examined, 504 (59.4%) produced semen and 344 (40.6%) did not produce semen. Line 2 had 92% of drones with semen present, compared with 40%, 39% and 65% for Lines 1, 3 and 4 respectively, Table 2. Drones aged 14, 21 and 35 days had 59%, 53% and 76% with semen present and spring, summer and autumn drones had 58%, 62% and 57% with semen present, Table 2. Chi squared tests showed the release of semen at the endophallus after manual eversion is dependent on breeding line and age but is not dependent on season. The volume of semen produced by drones was, mean 1.09 (ra. 0.77 0.05-1.16 0.04) l. Predicted semen volumes and range of volumes for drones producing a measurable volume of semen are shown in Table 3 for season, age and breeding line effects. Semen volume was higher in spring than in autumn or summer, higher at ages 14 and 21 days than 35 days, and highest for Lines 2 and 4 and lowest for Line 1. The number of sperm produced by drones was, mean 3.63 (ra. 2.19 0.19- 4.72 0.23) million sperm per drone. The predicted means of the number of sperm produced per drone and range of sperm numbers for drones producing a measurable volume of semen are shown in Table 4 for season, age, and breeding line effects. Sperm numbers were highest in autumn and lowest in spring, higher at age 21 days than 14 and 35 days, and highest for Line 2 and lowest for Line 1.

13

3.5 Discussion
The number of drones surviving to ages 14, 21 and 35 days of age is not known but it is clear from the numbers of drones marked at emergence (Table1) that a high percentage did not survive to 35 days. For example, for autumn drones no marked drones were found 35 days after emergence. At 35 days, and to lesser extents at 21 and 14 days, the population of drones surviving and able to be sampled may not be representative of the mix of each group of drones at emergence. If drone survival (or probability of being sampled) is correlated to the presence of semen at the endophallus, semen volume or number of sperm produced, and then there could be some biasing of the results from each older sample of drones. Chi squared tests of data in Table 2 showed the presence of semen at the endophallus in drones after manual eversion is dependent on drone age and breeding line but not on season. A higher proportion of drones were found with semen present at the endophallus for 35 day old drones (76%) than for 14 day old (59%) and 21 day old drones (53%), as well, there was a high mortality rate amongst the drones with insufficient drones present at 35 days to obtain a full sample of 30 drones on all but two occasions (Table 1). These data suggest that drones able to produce semen at the tip of their endophallus after manual eversion at 35 days have a higher survival rate than drones not able to produce semen after manual eversion and which die at a younger age. The proportion of drones producing semen at the endophallus after manual eversion for each sample examined, Table 2, was dependent (P 0.05) on the breeding line. Significant differences (P 0.05) were shown between the four breeding lines for semen volume, Table 3, and number of sperm produced, Table 4. In all instances drones from breeding Line 2 were superior to drones from breeding Line 1. These data suggest that genetics may be involved in the proportion of drones producing semen at the endophallus after manual eversion, volume of semen produced per drone, and number of sperm produced per drone, for breeding lines. Lines 1 and 3 had very low proportions of drones with semen present at the endophallus after manual eversion, 40% and 39% respectively, Line 4 was higher with 65% and Line 2 was highest with 92% of all drones examined producing a measurable amount of semen after manual eversion, Table 2. Line 2 queen bees heading the drone mother colonies were sourced from a honeybee breeding program which has been in operation in excess of 20 years with selection based on field evaluation combined with the use of instrumental insemination to produce the following seasons breeding stock. While not specifically selecting drones producing a large volume of semen after manual eversion for instrumental insemination during the program, such drones may have inadvertently been selected over drones producing a smaller volume of semen which has resulted in the significant difference in drone numbers producing semen after manual eversion between Line 2 and Lines 1, 3 and 4. In this experiment 40.6% of all drones examined after manual eversion at 14 to 35 days of age produced either no semen or there was a visible colouring of semen on the endophallus which was not sufficient to be collected, both were recorded as no semen present, Table 1. High percentages of drones producing no or small amounts of semen after manual eversion have been identified by Woyke et al. (2001) in Apis dorsata drones of unspecified age. Although Collins and Pettis (2001) and Anderson (2004) made reference to this subject, the proportion of mature aged drones in a colony not producing semen after manual eversion does not appear to have been identified previously as a subject of interest and neither has the importance of its apparent consequences been recognised during the mating of large numbers of queen bees at commercial mating apiaries, i.e. (i) whether these drones produce semen, and (ii) whether these drones are capable of mating with queen bees. The proportion of drones producing a measurable amount of semen compared with the number of drones examined for the three seasons, three age groups and four breeding lines is summarised in Table 2.

14

In situations where drones of an unknown age do not produce semen after manual eversion (e.g. during instrumental insemination practices), then such drones are often identified as being immature which may or may not be a correct identification. Predicted means in Tables 3 and 4 are based on the number of drones of a known age which produced a measurable amount of semen after manual eversion. The data does not include the number of drones which did not produce a measurable amount of semen after manual eversion even though those drones were considered sexually mature based on age. If the number of drones not producing semen had been included in the data then the predicted means would have been much lower. The term drone maturity and its relationship with this work are discussed in Section 9. While the ability of 14 day old drones to mate with queen bees or to release semen at the tip of the endophallus after manual eversion may be questioned, drones caught at 21 and 35 days of age can be considered to have reached maturity with physiological development necessary for successful mating completed. The predicted means for semen volume, Table 3, showed a seasonal effect with spring drones producing higher volumes of semen than summer and autumn drones (P 0.05). The higher volume of semen produced by spring drones, Table 3, but lower numbers of sperm produced, Table 4, indicates that a seasonal variation in sperm concentration (number of sperm per l of semen) occurs. Autumn drones produced higher mean numbers of sperm than summer drones (P 0.05) which produced higher mean numbers of sperm than spring drones (P 0.05), Table 4. For this experiment, the ranges of data for volume of semen per drone and number of sperm produced per drone were large, Tables 3 and 4. The volume of semen per drone, mean 1.09 (ra. 0.77 0.05 1.16 0.04) l, Table 3, was within the range provided by van Niem (1995) of 0.91 (0.8-1.0) l per drone, Woyke (1960, quoted in Rinderer, 1986) of 1.3 l per drone, and Collins and Pettis (2001) of 0.946 (0.48-1.67) l per drone. The number of sperm produced by drones, mean 3.63 (ra. 2.19 0.19 4.72 0.23) x 106 , Table 4, in this experiment were low compared with data from Schlns et al. (2003) 11.9 1.0 x 106 , Woyke (1960, quoted in Koeniger et al., 2005) 11-12 x 106 , Collins and Pettis (2001) 8.656 x 106 , van Niem (1995) 7.14 (5.75-8.45) x 106 , but were comparable with sperm counts recorded by Khler (1955, quoted in Ruttner, 1956) of an average of 4.5 x 106 and Anderson (2004) 3.19 2.37 x 106 sperm per drone. Although drone ages, breeding lines and seasons were identified when drone semen was of a higher quality, no drone age, breeding line or season was able to be identified when drones could be considered not suitable for mating with queen bees based on semen volume per drone and sperm number per drone data.

15

3.6 Conclusions
(i) Drone ages, breeding lines and seasons were identified when drone semen was identified being significantly higher in quality. No drone age, breeding line or season was identified when semen quality could be considered of low quality and drones producing that semen not suitable for mating with queen bees. The number of sperm produced per drone, mean 3.63 million, is low compared with the majority of published data. The volume of semen produced per drone, mean 1.09 l, is comparable with published data. Results suggest that volume of semen produced per drone and number of sperm produced per drone are selectable traits and may be improved by selection in a breeding program. Drone survival rates were high for 14 and 21 day old drones with low numbers of drones surviving to 35 days. Release of semen at the endophallus after manual eversion was dependent on breeding line and drone age but not on season. The significance of 41% of drones 14 to 35 days old not releasing semen at the endophallus after manual eversion requires attention.

(ii) (iii) (iv) (v) (vi) (vii)

16

3.7 Tables
Table 1. The proportion of drones producing a measurable amount of semen after manual eversion from the total number examined for four breeding lines, over three seasons and examined at three ages.
Season 2003-04 Line No. ** Number of drones with semen/ Number of drones examined (%) Age (days) 14 21 35 6/30 (20.0) 9/30 (30.0) 9/13 (69.2) 28/30 (93.3) 26/30 (86.7) 14/15 (93.3) 20/30 (66.7) 6/30 (20.0) 4/6 (66.7) 15/30 (50.0) 19/30 (63.3) 21/29 (72.4) 12/30 (40.0) 14/30 (46.7) 1/3 (33.3) 28/30 (93.3) 28/30 (93.3) 27/30 (90.0) 11/30 (36.7) 8/30 (26.7) 17/30 (56.7) 18/30 (60.0) 17/30 (56.7) 23/27 (85.2) 13/30 (43.3) 15/30 (50.0) * 28/30 (93.3) 20/22 (90.9) * 9/30 (30.0) 9/30 (30.0) * 23/30 (76.7) 6/13 (46.2) *

Spring

Summer

Autumn

1 2 3 4 1 2 3 4 1 2 3 4

601 427 190 284 219 640 750 710 129 700 460 820

* **

No drones available to be examined Number of drones marked at emergence

Table 2. The proportion of drones producing a measurable amount of semen after manual eversion from the total number examined for three seasons, three age groups and four breeding lines.
Variable Season Spring Summer Autumn Age (days) 14 21 35 Line 1 2 3 4 * Proportion of drones with semen (%) 177/303 (58.4) 204/330 (61.8) 123/215 (57.2) 211/360 (58.6) 177/335 (52.8) 116/153 (75.8) 79/196 (40.3) 199/217 (91.7) 84/216 (38.9) 142/219 (64.8) Sig.*

a a a b b a c a c b

Different letters are significantly different, P 0.05

17

Table 3. Semen volume and range of volume produced by drones for season, age and breeding line effects for drones which produced a measurable amount of semen, > 0.1 l, after manual eversion.
Variable Season Spring Summer Autumn Age (days) 14 21 35 Line 1 2 3 4 Semen volume/drone l Predicted meanss.e.* Range 1.1147 0.0418 0.9752 0.0508 0.8728 0.0511 1.0179 0.0365 1.1368 0.0405 0.8062 0.0533 0.7666 0.0547 1.1642 0.0356 0.9316 0.0545 1.0898 0.0405 0.2-3.6 0.1-3.0 0.1-1.6 0.1-2.2 0.2-3.6 0.2-3.0 0.2-1.8 0.1-3.0 0.1-2.6 0.2-3.6 Sig.+

a ab b a a b c a b a

* +

Data are fitted means averaged over other variables Different letters are significantly different, P 0.05

Table 4. Sperm numbers and range of sperm numbers produced by drones for season, age, and breeding line effects for drones which produced a measurable amount of semen, > 0.1 l, after manual eversion.
Variable Season Spring Summer Autumn Age (days) 14 21 35 Line 1 2 3 4 * + Number sperm/drone 106 Predicted meanss.e.* Range 2.19 0.19 3.48 0.23 4.72 0.23 3.27 0.16 3.76 0.18 3.35 0.24 2.39 0.25 4.61 0.16 3.27 0.24 3.58 0.18 0.12-9.43 0.12-19.13 0.04-11.86 0.12-11.52 0.18-13.52 0.12-19.13 0.13-7.78 0.12-19.13 0.04-7.16 0.02-13.52 Sig.+

c b a a a a c a b b

Data are fitted means averaged over other variables Different letters are significantly different, P 0.05

18

3.8 References
Anderson, D. 2004 Improving queen bee production, Publication No. 04/153, Rural Industries Research and Development Corporation, Canberra, ACT, Australia. Berg, S. and Koeniger, N.1990 Larger drones (Apis mellifera) have more offspring, in: Proc. German Zoological Society, 83rd Meeting, Frankfurt am Main, Gustav Fischer Verlag, 614. Berg, S. 1992 Der reproduktionserfolg von drohnen (Apis mellifera L.) unterschiedlicher Gre, PhD Thesis, Johann-Wolfgang Goethe University, Frankfurt am Main. Brunnich, K. 1927 The temperature of the honey-bee. 1. The male sexual organs:their histological structure and physiological functioning, Journal of Experimental Zoology, 31 (2), 225-265. Collins, A.M. and Pettis, J.S. 2001 Effect of Varroa infestation on semen quality, American Bee Journal, 141 (8), 590-593. Duay, P., De Jong, D. and Engels, W. 2002 Decreased flight performance and sperm production in drones of the honeybee (Apis mellifera) slightly infested by Varroa destructor mites during pupal development, Genetic Molecular Research, 1 (3), 227-232. Duncan, M. 2003 Controlled rearing of drone honeybees, University Western Sydney, Richmond, NSW, Australia. (Pers. comm..). Elbassiouny, A.M.M. 1992 Comparative studies on the Carniolan honeybee race and its crosses, PhD Thesis, Ain Shams University, Alexandria, Egypt, 53. Hoage, T.R. and Kessel, R.G. 1968 Journal of Ultrastructure Research, 24, 6-32. Jay, C. and Dixon, D. 1984 Infertile and Nosema infected honeybees shipped to western Canada, Journal of Apicultural Research, 23, 40-44. Jaycox, E.R. 1961 Effects of foods and temperatures on maturity of drone bees, Annals of the Entomological Society of America, 54, 519-523. Koeniger, G., Koeniger, N., Tingek, S. and Phiancharoen, M. 2005 Variance in spermatozoa number among Apis dorsata drones and among Apis mellifera drones, Apidologie, 36, 279-284. Khler, F. 1955 Untersuchungen zum Problem der knstlichen Begattung der Bienenknigin (Apis mellifica L.), Wrzburg : Inaugral-Dissertation. Kurennoi, N.M. 1953 When are drones sexually mature?, Pchelovodstvo, 11, 28-32 (in Russian). Translation in Marie Simpson, 1954, Some recent Russian researches on bees and beekeeping, Bee Research Association, London, 36 pp. Mackensen, O. 1955 Experiments in the technique of artificial insemination of queen bees, Journal of Economic Entomology, 48 (4), 418-421. Mackensen, O. and Roberts, W.C. 1948 A manual for the artificial insemination of queen bees, USDA No. ET 250. Moritz, R.F.A. 1981 Der einfluss der inzucht auf die fitness der drohnen von Apis mellifera carnica, Apidologie, 12, 41-55.

19

Moritz, R.F.A. 1986 Intracolonial worker relationship and sperm competition in the honeybee (Apis mellifera L.), Experienta 42, 445-448. Pabst, M.A. and Pfeiler, G. 1994 The sperms of young drones of Apis mellifera, International Congress on Electron Microscopy 13, Paris, 689-690. Phiancharoen, M., Wongsiri, S., Koeniger, N. and Koeniger, G. 2004 Instrumental insemination of Apis mellifera queens with hetero- and conspecific spermatozoa results in different sperm survival, Apidologie, 35, 503-511. Rhodes, J.W. 2002 Introduction and early performance of queen bees some factors affecting success, RIRDC Report, Canberra, Australia, 49 pages. Rinderer, T.E., Collins, A.M. and Pesante, D. 1985 A comparison of Africanized and European drones:weights, mucus gland and seminal vesicle weights, and counts of spermatozoa, Apidologie, 16 (4), 407-412. Rinderer, T.E. (Ed.) 1986 Bee Genetics and Breeding, Academic Press Inc., Orlando, Florida, USA. Rinderer, T.E, Guzman, L.I., Lancaster, V.A., Delatte, G.T. and Stelzer, J.A. 1999 Varroa in the mating yard : 1. The effects of Varroa jacobsoni and Apistan on drone honeybees, American Bee Journal, 139 (5), 134-139. Ruttner, F. 1956 The mating of the honeybee, Bee World, 37 (1), 3-15. Ruttner, F. 1976 The instrumental insemination of the queen bee, Apimondia, Bucharest. Schlns, H., Schlns, E.A., van Praagh, J. and Moritz, R.F.A. 2003 Sperm numbers in drone honeybees (Apis mellifera) depend on body size, Apidologie, 34, 577-584. Schlns, H., Koeniger, G., Koeniger, N. and Moritz, R.F.A. 2004 Sperm utilization pattern in the honeybee (Apis mellifera), Behavioral Ecology and Sociobiology, 56, 458-463. Schulz-Langner, E. 1958 Zeitschr. Bienenforsch, 4 (4), 67-86. van Niem Nguyen, 1995 Effects of protein nutrition and pollen supplementation of honeybee (Apis mellifera L.) colonies on characteristics of drones with particular reference to sexual maturity, Msc. Thesis, University of Western Sydney, Hawkesbury, New South Wales, Australia. Wheen, G. 2003 Collection and management of drones prepared for semen collection, Richmond, NSW, Australia. (Pers. comm..). Woyke, J. 1960 Natural and artificial insemination of queen honeybees, Pszcz. Zesz. Nauk., 4, 183275. Woyke, J. 1962 Natural and artificial insemination of queen bees, Bee World, 43, 21-25. Woyke, J., Wilde, J. and Wilde, M. 2001 Apis dorsata drone flights, collection of semen from everted endophalli and instrumental insemination of queens, Apidologie, 32, 407-416.

20

4. Sperm Viability
4.1 Introduction
Evidence showing that sperm competition selects for higher sperm quality in insects was provided by Garca-Gonzlez and Simmons (2005) who found that paternity success was determined by the proportion of live sperm in a males ejaculate and were able to predict the paternity patterns observed on the basis of the males relative representation of viable sperm in the females sperm-storage organ. Their results provided direct experimental evidence that sperm quality in the cricket Teleogryllus oceanicus plays an important role in determining which male has the advantage when males compete for fertilisation, and supported the hypothesis that selection should maximise sperm quality. Hunter and Birkhead (2002) examined the importance of sperm quality in determining which males sperm had the advantage when sperm from two or more males were competing to fertilise a females ova, finding that, all else being equal, males vary in their ability to fertilise ova on the basis of sperm viability alone suggesting that sperm viability is one of a suite of male adaptations to sperm competition in insects. Kraus et al. (2003) investigated male fitness of drone honeybees finding that selection through the male side appears to be an extremely important factor for colony fitness as the number of mating drones and the individual siring success of each drone is determined by the colony and/or the genotype of the mother queen. Interest in honeybee semen quality increased with the development of instrumental insemination of queen bees using semen collected by various means from drones of various ages and with the semen stored for different amounts of time under different conditions prior to use. Attention to drone semen quality was raised by Vesely (1970, quoted in Locke and Peng, 1993) who studied the retention of semen in the lateral oviducts of artificially inseminated queen bees which he suggested was due to factors affecting semen quality and which often resulted in infertility and death of the queen bee. Woyke and Jasinski (1978) examined the effects of drone age on semen quality finding that as the age of drones increased, mostly, a lower number of spermatozoa entered the spermatheca of queens, and the percentage of queens with semen residue in their oviducts increased. Effects of drone age on semen quality were examined by Locke and Peng (1993). They found that sperm viability, identified by supravital staining, was adversely affected by drone age. They suggested the decline in viability may be an indicator of a natural ageing process with sperm in the seminal vesicles reaching some age at which sperm senescence begins, with the final stage being sperm membrane disruption and death. Viability assessment of honeybee sperm was investigated and improved by Collins and Donoghue (1999) who developed and validated the use of the living:dead fluorescent stains SYBR-14 and propidium iodide (PI) with honeybee sperm. In an experiment examining the relationship between semen quality, i.e. percent viable sperm, and queen performance using instrumentally inseminated queen bees inseminated with known ratios of live:dead sperm, Collins (2000a) determined that sperm survival levels of 50% were sufficient for the queen to produce only fertilised eggs, at least early in their lives. Lodesani et al. (2004) when examining the effect of time on the viability of sperm in a queen bees spermatheca found a relatively low percentage of dead sperm in the spermatheca (20.5%) of two month old inseminated queen bees suggesting to them that sperm in the spermatheca undergo an initial selection for quality. However, after inseminating queen bees with only dead sperm, Collins (2000b) found dead sperm in the vaginal area of inseminated queen bees but none in the spermathecae suggesting that activity of the sperm themselves is critical in their migration to the spermatheca. Dead

21

sperm were found in the spermathecae of queen bees inseminated with a mixture of live and dead sperm allowing Collins (2000b) to suggest that live sperm drag dead sperm along with them when moving towards the spermatheca. Bresslau (1905, quoted in Collins, 2000b), identified a muscular sperm pump contained in the spermaduct which was considered to support migration of sperm into the spermatheca as well as the release of spermathecal contents for fertilisation. Sperm viability, using the Live:Dead SYBR-14:propidium iodide stains, was used as a measure of semen quality by Collins and Pettis (2001), they found that some sperm were killed during collection and mixing so that 100% fresh semen did not always provide 100% live sperm at examination. Damiens et al. (2002) using the Live:Dead SYBR-14:propidium iodide stains examined the viability of sperm of Eupelmus orientalis and Dinarmus basalis (Hymenoptera: Chalcidoidea) and found for both species that 40% of the sperm in the seminal vesicles was non-viable. They considered the high proportion of non-viable sperm could not be explained by the experimental procedure as special care was taken and suggested the results may be explained, in part, by the high initial dilution rate used during flushing procedures, 100 l saline for E. orientalis and 30 l saline for D. basalis. Centrifuging pooled semen at conventional speeds was found to result in unacceptably low viability (34.1%, unpublished data) by Collins (2003) using the Live:Dead dual fluorescent staining technique, the paper recommended best viability results were obtained with speeds of 82 or 250 g at 20-30 or 10-20 minutes respectively. Collins (2004) reported that the temperature of the assay buffer in which the semen was collected had less effect on viability than the way the semen was collected, washing semen into the buffer had less effect than collecting semen in a syringe with an average of 15% increase in viability, and highest viability levels, average of 21% higher, were found when semen was collected directly from the seminal vesicles rather than from ejaculation. Honeybee sperm viability data varies between authors. Locke and Peng (1993) reported sperm viability decreasing with increasing drone age with drones 28 and 42 days old , 81.4 1.62% and 80.1 2.01% respectively, having significantly lower sperm viability than 14 day old drones, 86.2 1.12%. Collins (2000a) observed good survival of sperm, 70-80%, for sperm stored at room temperature to 6 weeks. Collins and Pettis (2001) found a mean of 99.2 (ra. 82.3-100.0) % viability in a sample of 16 drones a minimum of 12 days old. Collins (2004) examined 20 drones collecting semen in a syringe and using room temperature buffer found a mean percent live sperm of 78.1 10.1 (ra. 58.5-91.0) %.

4.2 Methods
Viability assessment of semen samples were carried out using a Live/Dead Sperm Viability Kit developed at the University of Nevada, Reno, which contains a membrane-permeant nucleic acid stain developed at Molecular Probes (SYBR 14 dye) and the conventional dead-cell stain propidium iodide (Live/Dead Sperm Viability Kit (L-7011), 2001). DNA with intact cell membranes fluoresce bright green and cells with damaged cell membranes fluoresce red. Both dyes can be excited with visible-wavelength light, when bound to DNA the fluorescence emission maxima of these dyes are 516 nm and 617 nm respectively. From 3.2 (b) Laboratory Collection of Data, the volume of semen remaining from each drone after the amount required for sperm number and sperm motility examination had been removed was diluted in 40 l Tris buffer in a clean, labelled, small Effendorf tube. The two stains SYBR 14 dye and the propidium iodide dye were brought to room temperature. 1.5 l of SYBR 14 was added to the semen + buffer mixture, shaken and allowed to stand 3-5 minutes. 1.5 l of propidium iodide dye was then added, the mixture shaken and allowed to stand a further 3-5 minutes. A slide was made from each drone sample by placing 1.5 l of Tris buffer on the slide, adding 1.5 l of stained semen to the buffer and mixing. A slide cover was placed on the stained semen which was examined with an Olympus BX60 compound microscope with a blue fluorescence filter, excitation bandpass filter 460-490 nm, dichroic mirror 505 nm, barrier filter 515 nm suitable for examination of material stained with both SYBR 14 dye and propidium iodide dye at a magnification of 200. Six to eight fields of view of each

22

slide were photographed and saved for counting the numbers of live:dead sperm present on each field of view. Counting the number of live and dead sperm for five photographed field of views for each drone was carried out using image processing software Image Pro Plus, v 5.1, Media Cybernetics, Maryland, USA.

4.3 Statistical Analysis

Linear mixed model with breeding line, season, age, and all interactions included. ANOVA was not used due to unbalanced data, e.g. autumn 35 day old drone data missing.

4.4 Results
From Table 1. Drone sperm viability averaged 79.7 (ra. 59.9 90.5) %. Of the 848 drones examined, 425 (50.1%) produced a sufficient volume of semen for examination for sperm viability. The numbers of drones providing a sample of semen compared with the number of drones examined for each Line were, Line 1, 62/196 drones (31.6%); Line 2, 178/217 (82.0%); Line 3, 81/216 (37.5%); and Line 4, 117/219 (53.4%): for each Age were, 14 days of age,170/360 (47.2%); 21 days of age, 153/335 (45.7%); and 35 days of age, 102/153 (66.7%): and for each Season were, Spring, 159/303 drones (52.5%); Summer 162/330 (49.1%); and Autumn 104/215 (48.4%). No 35 day old autumn drones survived to be examined. From Table 2. A Seasonal effect was identified with autumn drones having a significantly higher, P 0.05, sperm viability than spring and summer drones. An Age effect was identified, P 0.05, with 21 day old drones having the highest sperm viability followed by 35 day old and 14 day old with the lowest viability. A Line effect, P 0.05, was shown with Line 1 drones having a significantly lower sperm viability than Lines 2, 3 and 4.

4.5 Discussion
The release of semen at the endophallus after manual eversion was dependent on breeding line and age but not on season, Table 1. A significantly higher season effect with autumn drones showing higher sperm viability than spring and summer drones, Table 2, when interpreted in conjunction with data showing no autumn drones survived to 35 days of age for examination, Table 1, provides two possible explanations of the results, (i) sperm viability is higher in autumn drones than in spring and summer drones with a possible reason for this being autumn reared queen bees mating with autumn drones are required to survive winter conditions and to be able to breed populous colonies the following spring, i.e. there is a colony survival benefit in autumn mated queen bees being well mated during the autumn mating period ; or (ii) the results also suggest that drones with low sperm viability die at a younger age with fewer drones with low sperm viability available to be sampled at the older age. The significant differences in sperm viability between drones sampled at different ages with 14 day old drones having the lowest viability, 78%, and 21 day old drones the highest viability, 82%, Table 2, requires explanation. Locke and Peng (1993) found sperm viability decreasing with increase in drone age with 28 and 42 day old drones having significantly lower sperm viability, 81% and 80% respectively, than 14 day old drones, 86%. If a batch of drones reared at the same time from the same drone mother and maintained under the same conditions is sampled at increasing ages for sperm viability, then the expected result would be similar to Locke and Pengs (1993) result of decreasing sperm viability with increasing age. The data obtained in this experiment of younger drones having the lowest sperm viability may be explained by drones with low sperm viability dying at a younger age and not being available to be sampled at the older ages.

23

The line effect of Line 1 having the significantly different lowest sperm viability, 75%, compared with 82%, 81% and 81% for Lines 2, 3 and 4 respectively suggests that sperm viability in drones may be a selectable trait when selecting queen bees to be drone mother queens for breeding purposes. Sperm viability data recorded from this experiment, average 79.7, range 59.9-90.5 %, are similar to data published by other authors e.g. Locke and Peng (1993) found 28 day old drones with 81.4 1.62% live sperm, 42 day old drones with 80.1 2.01%, and 14 day old drones with 86.2 1.12%. For drones a minimum of 12 days old, Collins and Pettis (2001) found a mean of 99.2%, range 82.3100.0%. Collins (2004) collecting semen in a syringe and using room temperature buffer found a mean live percent sperm of 78.1 10.1%, range 58.5-91.0% live sperm. These results indicate that viability levels of sperm from drone honeybees reared from queen bees from commercial breeding lines in eastern Australia are on the lower end, but are comparable with published data and suggest that an improvement in viability percent may be possible through a queen bee breeding program selecting for high sperm viability in drones produced from those queen bees. Collins (2004) compared methods for semen collection reporting the highest viability levels were found when semen was collected directly from the seminal vesicles rather than from ejaculation. This statement presents a strong argument for the collection of semen samples requiring a high level of accuracy to be collected from drone seminal vesicles.

4.6 Conclusions
(i) Significant differences were identified for sperm viability percent for drone age, season and breeding line. No combination of factors involving drone age, season or breeding line was identified which suggested that drones were not suitable for mating with queen bees based on sperm viability. Drone sperm viability averaged 79.7 (ra. 59.9-90.5) % and is on the lower end but comparable with limited published data. Results suggest that drone sperm viability is a selectable trait able to be improved by a breeding program.

(ii) (iii)

24

4.7 Tables
Table 1. Number of drones providing a semen sample after manual eversion, and assessment of mean percent sperm viability for drones over three seasons from four breeding lines for three ages.
Season Line Age (days) 14 14 14 14 21 21 21 21 35 35 35 35 14 14 14 14 21 21 21 21 35 35 35 35 14 14 14 14 21 21 21 21 35 35 35 35 No. drones examined 30 30 30 30 30 30 30 30 13 15 6 29 30 30 30 30 30 30 30 30 3 30 30 27 30 30 30 30 30 22 30 13 0 0 0 0 No. drones providing semen 5 25 17 13 8 24 6 16 9 12 4 20 8 21 6 14 10 26 6 14 0 24 16 17 10 26 6 19 12 20 7 4 0 0 0 0 Sperm viability mean % 88.7 77.2 77.1 85.8 59.9 76.5 86.6 78.1 71.7 79.6 64.5 81.9 68.7 71.7 66.8 70.3 79.4 89.9 90.5 81.0 83.2 80.0 79.3 78.6 89.3 90.5 83.7 83.4 87.1 88.0 84.2 -

Spring

Summer

Autumn

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

25

Table 2. Predicted sperm viability, live percent, for sperm produced by drones for season, age, and breeding line for drones which produced a measurable amount of semen, > 0.1 l after manual version.
Variable Season Spring Summer Autumn Age (days) 14 21 35 Line 1 2 3 4 * 75.47 1.70 82.33 1.06 80.84 1.69 81.25 1.25 Different letters are significantly different, P 0.05 b a a a 77.85 1.06 81.92 1.11 80.16 1.58 b a ab 77.00 1.09 78.06 1.09 84.87 1.54 b b a Sperm Viability, Live % Predicted mean s.e.

Sig.*

26

4.8 References
Bresslau, E. 1905 Der samenblasengang der bienenknigin, Zoologischer Anzeiger, 29, 299-325. Collins, A.M. 2000a Survival of honeybee (Hymenoptera: Apidae) spermatozoa stored at abovefreezing temperatures, Journal of Economic Entomology, 93 (3), 568-571. Collins, A.M. 2000b Relationship between semen quality and performance of instrumentally inseminated honeybee queens, Apidologie, 31, 421-429. Collins, A.M. 2003 A scientific note on the effect of centrifugation on pooled honeybee semen, Apidologie, 34, 469-470. Collins, A.M. 2004 Sources of variation in the viability of honeybee, Apis mellifera L., semen collected for artificial insemination, Invertebrate Reproduction and Development, 45 (3), 231-237. Collins, A.M. and Donoghue, A.M. 1999 Viability assessment of honeybee Apis mellifera sperm using dual fluorescent staining, Theriogenology, 51, 1513-1523. Collins, A.M. and Pettis, J.S. 2001 Effect of varroa infestation on semen quality, American Bee Journal, 141 (8), 590-593. Damiens, D., Bressac, C., Brillard, J-P. and Chevrier, C. 2002 Qualitative aspects of sperm stock in males and females from Eupelmus orientalis and Dinarmus basalis (Hymenoptera:Chalcidoidea) as revealed by dual fluorescence, Physiological Entomology, 27 (2), 97-102. Garca-Gonzlez, F. and Simmons, L.W. 2005 Sperm viability matters in insect sperm competition, Current Biology, 15, 271-275. Hunter, F.M. and Birkhead, T.R. 2002 Sperm viability and sperm competition in insects, Current Biology, 12, 121-123. Kraus, F.B., Neumann, P., Scharpenberg, H., van Praagh, J. and Moritz, R.F.A, 2003 Male fitness of honeybee colonies (Apis mellifera L.), Journal of Evolutionary Biology, 16, 914-920. Live/Dead Sperm Viability Kit 2001 Molecular Probes Inc., 4849 Pitchford Avenue, Eugene, OR 97402-9165, USA. Locke, S.J. and Peng, Y-S. 1993 The effects of drone age, semen storage and contamination on semen quality in the honeybee (Apis mellifera), Physiological Entomology, 18, 144-148. Lodesani, M., Balduzzi, D. and Galli, A. 2004 A study on spermatozoa viability over time in honeybee (Apis mellifera ligustica) queen spermathecae, Journal of Apicultural Research, 43 (1), 27-28. Vesely, V. 1970 Retention of semen in the lateral oviducts of artificially inseminated honeybee queens A. mellifera, Acta Entomologica Bohemoslovaca, 67, 83-92. Woyke, J. and Jasinski, Z. 1978 Influence of age of drones on the results of instrumental insemination of honeybee queens, Apidologie, 9, 203-212.

27

5. Sperm Motility
5.1 Introduction
Sperm, generally, are motile cells with sperm motility being critical at the time of fertilisation which allows sperm motility to be used as a measure of sperm quality. Tourmente et al. (2007) considered semen quality analysis a powerful tool for evaluation of fertility potential of males and stated that sperm motility, evaluated as the sperm velocity and the percentage of motile spermatozoa, is positively correlated with fertilisation success in several species. Investigations into diluents suitable for honeybee sperm storage by Verma (1978) found sperm survived longer in Tris buffer diluent at pH 7.19 than at pH 8.4. Locke and Peng (1993) examined honeybee sperm motility patterns to determine how drone age, semen storage time and contamination contribute to semen quality. Drone bees were marked on emergence and allowed free flight until examined, semen samples were collected by manual ejaculation. A 10 l subsample of semen suspended in modified Kieve solution was prepared for motility assessment. Motility was scored on a scale from 0 - 4 (0 = no movement, 4 = most movement), 3 minutes after the slide was placed on the stage at 400 magnification. In this experiment there was a lag time of 2 - 5 hours between semen collection and motility assessment. No significant differences, P < 0.05, were found for sperm motility for sperm from drones 14, 28 and 42 days of age with motility scores of 2.85 0.13, 2.55 0.15 and 2.65 0.17 respectively. Kaftanoglu and Peng (1984) found honeybee sperm motility rates were higher at pH 6.35-8.40 in hypertonic than in hypotonic solutions and no vigorous sperm motility was observed in saline and trisbuffer diluents, both of which were hypotonic to seminal plasma. Dietrich et al. (2005) assessed motility in rainbow trout (Oncorhynchus mykiss) semen and found that post-mortem storage ( 40-60 min.) lowered sperm motility and also significantly influenced sperm motility parameters such as sperm velocities, percentage of motile sperm and sperm trajectory parameters, demonstrating that sperm storage time after collection significantly alters sperm motility characteristics. Examining three species of deer Martinez-Pastor et al. (2005) stated in their introduction that the influence of season on sperm production and quality has been largely considered as a factor of high impact however they recorded that there were great differences between species. For chamois deer they found that sperm samples studied during the breeding season provided good motility results with almost half of the samples collected during the non-breeding season having almost no motility. Gizejewski (2004, quoted in Martinez-Pastor et al., 2005) found for red deer that sperm motility was better in the first part of the mating period than in the rest of the mating period. Abdelwahab et al. (2006) found the mean sperm motility in dairy goats was significantly higher, P 0.05, in spring than in autumn. Ravimurugan et al. (2006) found a seasonal effect on motility of buffalo sperm with sperm motility being significantly highest, P 0.01, in winter for the first ejaculate and the mean sperm motility highest in summer for the second ejaculation.

5.2 Methods
Drone honeybee sperm motility was assessed based on the method described in Locke and Peng (1993), at room temperature. From Section 3.2 (b), a 10 l sample of semen diluted in Tris buffer was assessed for motility within 5 minutes of each semen sample being collected. Motility was assessed by preparing the two ends of two haemocytometer slides, Improved Neubauer BS.748, depth 0.1 mm and 1/400 mm2 , covered with a coverslip and five fields of view examined, one each from three ends of the two haemocytometers and two different fields of view on the fourth end of one haemocytometer. A light microscope at 400 magnifications was used to record motility which was scored at five levels of movement 1 = no movement; 2 = < 50% of sperm vibrating; 3 = > 50% sperm vibrating but no circular or progressive movement; 4 = > 50% vibrating but < 50% exhibiting circular and progressive movement and; 5 = > 50% exhibited circular and progressive movement. 28

5.3 Statistical Analysis

Linear mixed model with breeding line, season, age and all interactions included.

5.4 Results
From Table 1. Drone sperm motility averaged 3.4 (ra.2.1-4.5). Of the 848 drones examined, 459 produced a sufficient volume of semen for examination for sperm motility. The number of drones providing a sample of semen compared with the number of drones examined, for each Line were, Line 1, 54/196 drones (27.6%); Line 2, 188/202 (93.1%); Line 3, 80/216 (37.0%); Line 4, 137/219 (62.6%); for each Age were, age 14 days, 201/360 (55.8%); 21 days, 155/335 (46.3%); age 35 days, 103/153 (67.3%); for each Season were, spring, 187/303 (61.7%); summer, 167/330 (50.6%); autumn, 105/215 (48.8%). No 35 day old drones survived to be examined. From Table 2. A seasonal effect was identified with sperm from spring and summer drones having significantly higher, P 0.05, motility rating than autumn drones. An age effect was present with significant differences at the P 0.05 level with motility rates for 35 day old drones higher than 21 day old which were higher than 14 day old drones. A Line effect was found with Lines 2, 3 and 4 having a significantly higher, P 0.05, sperm motility rate than Line 1.

5.5 Discussion
Limited data are available for comparison with sperm motility results from this experiment. Locke and Peng (1993), when their scoring scale is converted from 0-4 to 1-5 used in this experiment, found sperm motility of drones aged 14 to 42 days ranging between 3.55 and 3.85 after a lag time of 2-5 hours. Dietrich et al. (2005) found post-mortem storage in trout sperm of 40-60 minutes lowered sperm motility. Sperm motility ratings from this experiment for drones aged 14-35 day, over three seasons, averaged 3.4 (ra. 2.1-4.5) with a time of approx. 5 minutes between sperm collection and examination. This data suggests that sperm motility of drones from commercial breeding lines of queen bees from eastern Australia is lower than data from Locke and Peng (1993) due to the increased lag time in their experiment. Differences were found for the percent of drones producing semen compared with the number of drones able to be examined for sperm motility. The release of semen at the endophallus after manual eversion was dependent on breeding line and age but not on season, Table 1. A seasonal effect was identified, Table 2, with spring and summer drone sperm motility higher than autumn motility, P 0.05. Results from motility studies on mammal sperm identified the importance of seasonal effect on sperm production and quality, however, great differences in results have been found between mammalian species and differences between animal and insect genera are not reported. The lower motility of autumn drone sperm compared with spring and summer drone sperm, combined with mammalian sperm data identifying distinct seasonal effects, suggest the question of whether there is a time period for Apis mellifera during one or more seasons each year when sperm production is at its highest level of quality. Queen bees for the commercial beekeeping industry are reared and mated continuously between early spring and late autumn each year, if one or more time periods of peak sperm quality were identified for honeybees then this would question the mating quality of queen bees mating with drones out of the peak sperm quality time period. An age effect on drone sperm motility was identified, Table 2, with 35 day old drones having the highest motility levels followed by 21 day old drones and 14 day old drones with the lowest levels, P 0.05. Locke and Peng (1993) did not find significant differences between sperm motility from drones 14, 28 and 42 days of age. Should an age effect be present, as shown by data in this experiment, then an expected result would have been the 14 day old drones with the highest motility 29

and the 35 day old drones with the lowest motility. An explanation for the 35 day old drones having the highest motility and the 14 day old the lowest may result from the number of drones available for examination at each age, 201 for 14 day old, 155 for 21 day old, and 103 for 35 day old, i.e. the decreasing number of drones available for examination with increasing age providing highest motility for the oldest aged drones suggests that drones with low motility levels die at a younger age and are not available for sampling at the older age. A breeding line effect was identified, Table 2, with Line 1 drone sperm motility being lower, P 0.05, than Lines 2, 3 and 4, Table 2. These data suggest that drone sperm motility may be a selectable trait able to be improved in a bee breeding program by selection of queen bees whose drones produce sperm with high motility rates. Although significant differences for sperm motility were identified for season, age and line, Table 2, differences were not considered sufficient to determine that one set of circumstances produced sperm motility of low quality not suitable for those drones to mate with queen bees.

5.6 Conclusions
(i) (ii) (iii) Drone sperm motility rating is lower than limited published data, av. 3.4 (ra. 2.1-4.4) on a scale of 1-5. Results suggest that drone sperm motility is a selectable trait able to be improved by a breeding program. A significant seasonal effect suggests that a higher quality sperm, based on motility rating, may be produced by drones during certain periods each year.

30

5.7 Tables
Table 1. Number of drones providing semen after manual eversion for assessment of mean sperm motility for drones over three seasons from four lines for three ages.
Season Line Age (days) 14 14 14 14 21 21 21 21 35 35 35 35 14 14 14 14 21 21 21 21 35 35 35 35 14 14 14 14 21 21 21 21 35 35 35 35 No. drones examined 30 30 30 30 30 30 30 30 13 15 6 29 30 30 30 30 30 30 30 30 3 30 30 27 30 30 30 30 30 22 30 13 0 0 0 0 No. drones providing semen 0 30 26 30 8 24 6 17 7 14 4 21 7 23 8 15 10 26 7 14 0 24 16 17 10 27 6 19 12 20 7 4 0 0 0 0 Sperm motility Grade 1-5 Mean 3.1 3.1 2.3 2.6 4.1 4.2 4.3 4.4 4.2 4.2 4.5 4.3 3.4 3.6 4.0 3.4 4.0 3.5 4.0 4.0 3.6 3.0 2.1 3.0 2.3 3.2 2.2 3.1 2.6 2.8 -

Spring

Summer

Autumn

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

31

Table 2. Predicted sperm motility, Grade 1-5, for sperm produced by drones for season, age and breeding line for drones which produced a measurable amount of semen, > 0.1 l, after manual eversion.
Variable Sperm Motility Grade 1-5 Predicted mean s.e. 3.66 0.09 3.62 0.09 3.25 0.13 3.08 0.09 3.46 0.09 4.00 0.13 3.10 0.15 3.71 0.09 3.57 0.14 3.66 0.10 Sig.*

Season Spring Summer Autumn Age (days) 14 21 35 Line 1 2 3 4 *

a a b c b a b a a a

Different letters are significantly different, P 0.05

32

5.8 References
Abdelwahab, D., Rivera, M.M., Rodrguez-Gil, J.E. and Rigau, T. 2006 Seasonality effects on sperm motility kinematic parameters of Murciano-Granadina bucks, Reproduction of Domestic Animals, 41, Supplement 2, 103. (Abstract). Dietrich, G., Kowalski, R., Wojtczak, M., Dobosz, S., Goryczko, K. and Ciereszko, A. 2005 Motility parameters of rainbow trout (Oncorhynchus mykiss) spermatozoa in relation to sequential collection of milt, time of post-mortem storage and anaesthesia, Fish Physiology and Biochemistry, 31 (1), 1-9. Gizejewski, Z. 2004 Effect of season on characteristics of red deer (Cervus elaphus L.) semen collected using modified artificial vagina, Reproductive Biology, 4, 51-66. Kaftanoglu, O. and Peng, Y-S. 1984 Preservation of honeybee spermatozoa in liquid nitrogen, Journal of Apicultural Research, 23, 157-163. Locke, S.J. and Peng, Y-S. 1993 The effects of drone age, semen storage and contamination on semen quality in the honeybee (Apis mellifera), Physiological Entomology, 18, 144-148. Martinez-Pastor, F., Guerra, C., Kaabi, M., Garcia-Macias, V., de Paz, P., Alvarez, M., Herraez, P. and Anel, L. 2005 Season effect on genitalia and epididymal sperm from Iberian red deer, roe deer and Cantabrian chamois, Theriogenology, 63 (7), 1857-1875. Ravimurugan, T., Kanakaraj, P. and Thangaraju, P. 2006 Effect of seasons on semen production traits in murrah buffalo bulls, Tamilnadu Journal of Veterinary and Animal Sciences, 2 (3), 109-111. Tourmente, M., Cardozo, G.A., Guidobaldi, H.A., Giojalas, L.C., Bertona, M. and Chiaraviglio, M. 2007 Sperm motility parameters to evaluate the seminal quality of Boa constrictor occidentalis, a threatened snake species, Research in Veterinary Science, 82 (1), 93-98. Verma, L.R. 1978 Biology of honeybee (Apis mellifera L.) spermatozoa 1. Effect of different diluents on motility and survival, Apidologie, 9 (3), 167-174.

33

6. Role of genetics in drone semen production

6.1 Introduction.
Data from Section 3 showed significant differences, P 0.05, between the four breeding lines examined for the proportion of drones producing semen at the endophallus after manual eversion, for the volume of semen produced per drone and for the number of sperm produced per drone, with a high producing line and a low producing line able to be identified. The aims of this experiment were to examine whether (i) the ability of mature age drone honeybees to release semen at their endophallus after manual eversion, (ii) the volume of semen produced per drone, and (iii) the number of sperm produced per drone, are selectable traits. There is no readily available data on the subject of the ability of drone bees to release semen following manual eversion which raises the questions whether the absence of semen at the endophallus after manual eversion, Section 3, Table 1, indicates that the drone: (i) did not produce semen in the seminal vesicles (ii) did produce semen in the seminal vesicles but did not release the semen due to (a) physiological or behavioural reasons (b) physical reasons based on the ability of the operator carrying out the manual eversion. As significant differences were found between breeding lines for (i) the proportion of drones producing semen after manual eversion, (ii) the volume of semen produced per drone, and (iii) the number of sperm produced per drone, with the manual eversions of all drones being carried out by one operator and all measurements being determined by a second operator, then the human factor of operator effect may be removed from the list of reasons significant differences were found. An investigation into the proportion of drones producing semen after manual eversion, Section 3, Tables 1 and 2, and the volume of semen and number of sperm produced per drone, Section 3, Tables 3 and 4, resulted in significant differences, P 0.05, between the breeding lines. These data suggest that genetics may be involved in the proportion of drones releasing semen at their endophallus, the volume of semen produced per drone, and the number of sperm produced per drone for each breeding line, based on semen samples being collected by manual eversion of drones. Page and Fondrk (1995), using two-way selection for quantities of stored pollen, were able to produce high and low pollen hoarding strains of honeybees. Page et al. (1995) concluded that the amount of pollen stored in honeybee colonies is a selectable trait with five generations of two-way selection resulting in high and low strains which differed more than six-fold in quantities of stored pollen. Further comparisons with hybrid crosses suggested that colony-level, high pollen-hoarding behaviour is inherited as a recessive trait. Page et al. (1995) quoted numerous references demonstrating genotypic variability of individual bees within a colony for a number of traits of foraging behaviour, recruitment and lifetime productivity (i) the decision to forage for pollen or nectar, (ii) behavioural plasticity associated with switching foraging resources, (iii) nectar and pollen load size, (iv) round trip time and foraging activity measured by numbers of foraging round trips, (v) the age at which individuals initiate foraging, (vi) attendance of recruitment dances, (vii) foraging distance and (viii) dance dialects. Hunt et al. (1995) identified two genomic regions that affect the amount of pollen stored in honeybee colonies and influence whether foragers will collect pollen or nectar.

34

Honeybee foragers were tested for their proboscis extension response, PER, to water and a range of concentrations of sucrose by Page et al. (1998). Using the PER assay they examined returning pollen and nectar foragers at the hive entrance from non-selected strains of bees, and also examined bees from artificially selected high- and low- pollen-hoarding strains. Both sources of bees provided similar results with pollen collecting foragers having a significantly higher probability of responding to water and to lower concentrations of sucrose. These results demonstrated a genotypic effect on PERs of returning foragers with the PERs of departing high- and low-strain foragers consistent with those of returning foragers. Page et al. (1998) identified a previously mapped genomic region that contained a locus that appears to influence sucrose response thresholds which they considered demonstrated a gene-brain-behaviour pathway that can be altered as a consequence of colony-level selection for quantities of stored food. Bees derived from artificially selected high- and low-pollen hoarding strains for their proboscis extension reflex, PER, response to water and varying sucrose concentrations were also examined by Pankiw and Page (1999). Statements in their Introduction included that (i) genotype affects the probability that a forager will collect pollen or nectar with the variation in response thresholds to sucrose being a direct or indirect result of variation in genotype, (ii) conversely genotype could set response thresholds which may then have a causal effect on foraging behaviour, and (iii) a third alternative is that genotype affects both the response threshold and foraging behaviour with no causal association between them. Pankiw and Page (1999) examined the hypothesis that observed variation in response thresholds between workers of the high and low pollen hoarding strains are not a consequence of experiences associated with foraging behaviour with their results showing that PER thresholds were strongly correlated with genotype and since the genotypic effects on PER preceded foraging, then they concluded that the foraging role did not cause differences between high and low strains. Humphries et al. (2003), examined how genotypic differences between high and low pollen hoarding strains of honeybees correlate with changes at the level of proteins involved in neuronal function finding the two genetic strains of honeybees showed characteristic differences in the regulation of protein expression which may have contributed to the behavioural differences between them. Moura et al. (1997) conducted a two-way selection experiment in a composite rabbit population to investigate responses to selection for post weaning average daily gain and feed conversion. They found that selection for average daily gain was effective in improving average daily gain and feed conversion demonstrating that daily average gain is a selectable trait. The above examples of selectable traits involving foraging behaviour in honeybees and average weight gain in rabbits demonstrate genotypic variability for some aspects of behaviour and physiology present in some animals. On this basis it may be that different breeding lines of drones display differences in behaviour when manually everted with some lines releasing semen at the endophallus in a higher proportion of drones than other lines. As well, some breeding lines of drones may display differences in their physiology by having higher proportions of drones producing higher volumes of semen and higher numbers of sperm in their seminal vesicles than other lines. Drones were examined for the presence of adult bee diseases Nosema spp. and virus infections to identify their presence or absence and, if present, examine their possible effects on the results obtained.

35

6.2 Methods
Examination of drones for this experiment was carried out between November 2004 and March 2006. A Two-way Cross Methodology was developed (Tier, 2004) to determine whether genetic inheritance is involved in the release of semen at the endophallus after manual eversion, in semen volume produced per drone and in the number of sperm produced per drone.

Two-way cross methodology

1. Select one line to represent the High Line and one line to represent the Low Line from data developed in Section 3. From Section 3, Tables 2, 3 and 4, Line 2 was selected as a High Line and Line 1 selected as a Low Line for the proportion of drones producing a measurable amount of semen at the endophallus after manual eversion, for the volume of semen produced per drone and for number of sperm produced per drone. - Rear 50 drones from each of 5 queens from each line. Examine drones from each queen for proportion of drones producing a measurable amount of semen at the endophallus after manual eversion, the volume of semen produced and the number of sperm produced. - Examine data from drones from each line to confirm the two lines are significantly different from each other for the three characteristics measured. - Select the queen bee with the highest data from the High Line (Selected High Line queen mother) and the queen bee with the lowest data from the Low Line (Selected Low Line queen mother) to produce daughter queen bees. - The remaining queen bees identified in each group are used to rear High Line and Low Line drones for the experiment (High and Low Line drone mothers). 2. Rearing of High Line and Low Line queen bees and drone bees for the Line Crossing experiment.
High Line Selected High Line queen mother Raise 10 daughter queens - divide into 5 (a) queens and 5 (b) queens Raise drones - homogenise semen Raise 10 daughter queens - divide into 5 (c) queens and 5 (d) queens Raise drones - homogenise semen H

High Line

Remaining High Line identified queens Selected Low Line queen mother

Low Line

Low Line

Remaining Low Line identified queens

36

3. Crossing the four groups of queen bees and drone bees using artificial insemination to ensure correct matings.
Drones H H L L H 5(a) 5(c) 5(d) L 5(b)

Queens

4. Produce 2 queens from each of the 5 queens in each of the four groups (Point 3), all queens to be open mated (allowed to mate on the wing with unknown drones). - provides a total of 40 queens (10 (a), 10 (b), 10 (c), and 10 (d) queens. 5. Rear 30 drones from each of the 40 queens, in Point 4. Examine each drone at 21 days of age for the proportion of drones producing a measurable amount of semen after manual eversion, the volume of semen produced and the number of sperm produced. 6. Select data from the 3 highest and 3 lowest producing queens from each of the four groups of 10 queens (Point 4). A total of 24 queen bees to provide data. 7. Analyse data from Point 6 for variation between and within groups with the purpose of identifying variations in the three characteristics measured proportion of drones producing semen, semen volume per drone, and number of sperm per drone. 8. Analyse data for correlations between High and Low Lines.

Confirmation of High and Low Lines

Prior to commencing the two-way cross experiment, drones from the identified High Line and Low Line were examined three times for differences between the High and Low Lines. Test 1. Drones were reared in hives headed by queen bees used in the 2003-04 experiments, Section 3, in December 2004. Forty drones from each hive were examined at 21 days of age in January, 2005. Drones were reared and examined for (i) proportion of drones producing a measurable volume of semen after manual eversion, (ii) volume of semen produced per drone, and (iii) number of sperm produced per drone, from five High Line queen bees, hive numbers 1-5, and three hives with Low Line queen bees, hive numbers 6-8. Results are shown in Tables 1 and 2. Test 2. The 2-way crossing experiment was planned to be completed in the beekeeping season of spring 2005 - autumn 2006. When the experiment recommenced in spring, 2005, it was found that the identified High and Low Line queen bees had been superseded. Young queen bees from the same breeding lines as the High and Low Line queens were obtained from the two beekeepers who supplied the original queen bees. The replacement queen bees were obtained in October, 2005. Evaluation of drones from the two lines to determine whether they satisfied requirements for classification into High and Low Line, based on the three characteristics examined, commenced in November, 2005. Thirty drones reared from 4 High Line and 3 Low Line replacement queen bees were examined at 21 days of age for (i) the proportion of drones producing a measurable amount of semen after manual eversion, (ii) the volume of semen produced per drone, and (iii) the number of sperm produced per drone. Results are shown in Tables 3 and 4. Test 3. Replacement High and Low Line queen bees providing drones for Test 2 were examined a second time in March 2006 for characteristics (i), (ii) and (iii) in Test 2. The number of drones available up to 30 drones per hive at 21 days of age was examined from 3 High and 3 Low Line hives. Results are shown in Tables 5 and 6. 37

Disease identification Nosema and virus diseases. Forty drones from Lines 1 and 40 drones from Line 2 remaining from the semen volume and sperm number examinations, January 2005, were held at -20C until examined at the CSIRO Entomology Laboratories, Canberra under the direction of Dr. D. Anderson, March 14-17, 2005. Drones were examined for Nosema disease, Nosema spp., and for the virus diseases, Kashmir bee virus, chronic bee paralysis virus, black queen cell virus, bee virus X, bee virus Y, cloudy wing virus and sacbrood virus. Nosema disease examination. The drone body was placed in distilled water on a watch glass and the intestine dissected from the abdomen. The intestine was macerated in 1 ml distilled water and placed on a vortex for 10 seconds. One drop of solution was placed on a slide, covered with a coverslip, and examined with a light microscope at 400 magnifications. Three fields of view were examined for the presence of spores. For drones where spores were present, the number of spores per drone was calculated from the number of spores present in 10 squares, each of 16 small squares, of a haemocytometer, Improved Neubauer, depth 1 mm, 1/400 mm2 , using Cantwell (1970) calculation tables. Virus examination. Each drones head and thorax were stored in100 l insect Ringers solution in a labelled Eppendorf tube until examined. For examination, the drone head and thorax were placed in an Eppendorf tube with 900 l phosphate buffer and 100 l chloroform, macerated, placed on a vortex for 30-45 seconds, and then centrifuged at 8000 rpm for 2 minutes. The drone sample was placed in a fume cabinet and the clear liquid removed and placed in a new, labelled Eppendorf tube. Gel-diffusion plates were prepared using 0.25% Agarose gel with a diluted antiserum of antiserum/insect Ringers:1/20. Antiserums used were Kashmir strain K23 (ACT local strain), Kashmir strain 2 (Fiji), chronic bee paralysis, black queen cell virus, bee virus X, bee virus Y, cloudy wing virus and sacbrood virus. Acute bee paralysis virus was not examined for since it has not been identified in Australia. The gel-diffusion plates containing the antiserums and drone extracts were held in a damp chamber overnight and examined the following morning.

6.3 Statistical analysis

Two-sample test for equality of proportions with continuity correction.

6.4 Results
Confirmation of High Line and Low Lines
Drones from test queen bees were examined at a mature age for the three characteristics (i) the proportion of drones releasing semen at their endophallus after manual eversion, (ii) the volume of semen produced per drone, (iii) the number of sperm produced per drone. Test 1. When results from drones from all 5 High Line hives were compared against data from drones from all 3 Low Line hives, a significant difference, P 0.05, was found for the number of sperm produced per drone, and no significant differences were found for the proportion of drones releasing semen after manual eversion, or for the volume of semen produced per drone. When data from the 2 highest producing High Line hives were compared against data from the 2 lowest producing Low Line hives, significant differences, P 0.05, were found for all three characteristics examined, Table 2. Test 2. No significant differences were found for the three characteristics examined when data from drones from the 4 replacement High Line queen bees were compared against data from drones from the 3 replacement Low Line queen bees, Table 4. 38

Test 3. No significant differences were found for the three characteristics examined when data from drones 3 replacement High Line queen bees were compared against data from drones from 3 replacement Low line queen bees in a second examination, Table 6.

Disease examination
Virus disease. For the 40 High Line drones examined no drones were found positive for virus infection. For the 40 Low Line drones examined, one drone (L-11) showed a positive reaction for sacbrood virus. No other virus infection was identified. Nosema disease. (i) High Line - small numbers of spores were found in 10/40 High Line drones. Numbers of spores present in the 10 drones with spores present were, av. 0.148 (ra. 0.05-0.225) x 106 spores per drone. (ii) Low Line Large numbers of spores were present per drone in 2 drones, drone L-27 19.53 x 106 and drone L-29 3.4 x 106, and small numbers in 2 drones, drone L-35 0.075 x 106 and drone L-38 0.025 x 106 spores per drone.

6.5 Discussion
Confirmation of High and Low Lines
This experiment was suspended after Test 3 showed that a High Line and a Low Line were not present for the three characteristics being examined in drones reared from queen bees from the two replacement lines. The loss of the differences between the two lines of bees for the characteristics of (i) the proportion of drones producing semen at the endophallus after manual eversion, (ii) the volume of semen produced per drone, and (iii) the number of sperm produced per drone, suggests that if these characteristics are selectable traits, then recessive genes may be involved. Should recessive genes be involved in the expression of those three characteristics, then identification and maintenance of these traits in queen bee breeding stock would require a continuing managed breeding program.

Disease examination
Nosema - small numbers of spores in 10 of the 40 High Line drones examined may have resulted from feeding behaviour, or may suggest the drones have been or were about to become infected. Large numbers of spores in 2 of the 40 drones examined from the Low Line suggests that a heavy infection may have been present in all drones in that line. A more sensitive test than the one carried out may have shown this. Virus one Low Line drone out of the 80 examined was positive for sacbrood virus. A gel-diffusion test requires large numbers of particles present, more than one million, to provide a reaction. Combined, the results suggest that the Low Line drones are affected by the pathogens Nosema spp.and sacbrood virus. A high concentration of sacbrood virus in one drone, in conjunction with the Nosema levels from the Low Line is sufficient to cast doubt on all the drones examined i.e. a basic health condition may exist. These results indicate that further examinations of drones to obtain data on semen production and sperm production should include Nosema disease and virus disease examinations of those drones to identify effects of diseases on drone semen and sperm production.

39

6.6 Conclusions
(i) Results suggest that recessive genes are involved in the expression of (i) the proportion of drones releasing semen at the endophallus after manual eversion, (ii) the volume of semen produced per drone, and (iii) the number of sperm produced per drone, should these three characteristics be found to be selectable traits. Levels of diseases identified in drones indicate that disease monitoring of test drones is required in association with experiments on drones for semen quality.

(ii)

40

6.7 Tables
Table 1. Data from drones from the original High and Low Lines hive comparison experiment (Test 1) showing the proportion of drones producing semen after manual eversion, semen volume and sperm numbers for each hive sample for drones from each of the queen bees being examined.
Hive Line* Number of drones examined Drones with semen % 47.5 65.0 72.5 69.6 87.5 60.0 35.0 77.5 Semen volume per drone average (range) l 1.25 (0.2-1.8) 1.00 (0.6-1.8) 1.01 (0.6-1.4) 0.98 (0.6-1.6) 1.17 (0.6-1.6) 0.67 (0.1-1.2) 0.75 (0.2-1.1) 0.82 (0.2-1.2) Number of sperm per drone average (range) 6 10 4.35 (0.23-8.91) 3.24 (1.24-5.78) 3.95 (2.04-8.04) 4.33 (1.02-8.22) 4.07 (1.88-7.63) 2.08 (0.17-4.15) 2.43 (0.62-4.65) 2.70 (0.30-4.98)

number 1 2 3 4 5 6 7 8 *

H H H H H L L L

40 40 40 40 40 40 40 40

H = High Line, L = Low Line

Table 2. Analysis of data from Table 1 showing differences between drones from High and Low Line queen bees for all High Line hives v. all Low Line hives, and the two representative hives for High Line v. Low Line.
Hive numbers Line Proportion of drones producing semen n.s.* Semen volume/drone l Number of sperm/dron e 106

1,2,3,4,5 6,7,8 3 and 5 6 and 7 * **

H v. L H v. L

n.s.

s.

s.**

s.

s.

n.s. = not significant, P 0.05 s. = significant, P 0.05

41

Table 3. Data from drones from the High and Low Line replacement queen bees (Test 2) showing the proportion of drones producing semen after manual eversion, semen volume and sperm numbers for each hive sample.
Hive Line* Number of drones examined Drones with semen % 43.3 53.3 50.0 60.0 63.3 40.0 70.0 Semen volume per drone average (range) l 1.1 (0.4-1.6) 1.21 (1.0-1.6) 0.99 (0.2-1.5) 1.17 (0.6-1.5) 0.84 (0.4-1.2) 0.89 (0.6-1.4) 0.82 (0.4-1.2) Number of sperm per drone average (range) 106 7.17 (2.70-14.50) 4.73 (1.35-10.48) 3.68 (0.24-7.23) 5.98 (3.24-10.8) 4.32 (1.79-6.75) 3.49 (0.93-5.32) 3.47 (1.3-6.93)

number 1 2 3 4 5 6 7 *

H H H H L L L

30 30 30 30 30 30 30

H = High Line, L = Low Line

Table 4. Analysis of data from Table 3 showing differences between drones from replacement High and Low Line queen bees for all High Line hives v. all Low Line hives.
Hive numbers Line Proportion of drones producing semen n.s.* Semen volume/drone l Number of sperm/drone 106

1,2,3,4 5,6,7 *

H v. L

n.s.

n.s.

n.s. = not significant, P 0.05

Table 5. Data from drones from the High and Low Line replacement queen bees (Test 3) showing the proportion of drones producing semen after manual eversion, semen volume and sperm numbers for the combined High Line hives and combined Low Line hives.
Line Number of drones examined 90 59 Drones with semen % 33.3 35.6 Semen volume per drone mean s.d. l 0.76 0.08 0.81 0.09 Number of sperm per drone mean s.d. 106 4.0 0.7 4.2 0.8

High Low

42

Table 6. Analysis of data from Table 5 showing differences between drones from High and Low Line queen bees for all High Line hives v. all Low Line hives.
Line Proportion of drones producing semen n.s.* Semen volume/drone l Number of sperm/drone 106

H v. L *

n.s.

n.s.

n.s. = not significant, P 0.05

43

6.8 References
Cantwell, G.E. 1970 Standard methods for counting Nosema spores, American Bee Journal, 110, 222223. Humphries, M.A., Mller, U. and Fondrk, M.K. 2003 PKA and PKC content in the honeybee central brain differs in genotypic strains with distinct foraging behaviour, Journal of Comparative Physiology, 189, 555-562. Hunt, G.J., Page, R.E. Jr., Fondrk, M.K. and Dullum, C.J. 1995 Major quantitative trait loci affecting honeybee foraging behaviour, Genetics, 141, 1537-1545. Moura, A.S.A.M.T., Kaps, M., Vogt, D.W. and Lamberson, W.R. 1997 Two-way selection for daily gain and feed conversion in a composite rabbit population, Journal of Animal Science, 75, 2344-2349. Page, R.E. Jr. and Fondrk, M.K. 1995 The effects of colony-level selection on the social organisation of honeybee (Apis mellifera L.) colonies:colony-level components of pollen hoarding, Behavioral Ecology and Sociobiology, 36, 135-144. Page, R.E.Jr., Waddington, K.D., Hunt, G.J. and Fondrk, M.K. 1995 Genetic determinants of honeybee foraging behaviour, Animal Behaviour, 50, 1617-1625. Page, R.E. Jr., Erber, J. and Fondrk, M.K. 1998 The effect of genotype on response thresholds to sucrose and foraging behaviour of honeybees (Apis mellifera L.), Journal of Comparative Physiology A, 182, 489-500. Pankiw, T. and Page, R.E.Jr. 1999 The effect of genotype, age, sex, and caste on response thresholds to sucrose and foraging behaviour of honeybees (Apis mellifera L.), Journal of Comparative Physiology A, 185, 207-213. Tier, B. 2004 Development of an experimental program to determine if genetic inheritance is involved in the volume of semen and number of sperm produced by drone honeybees, AGBU, UNE Armidale, New South Wales, Australia. (Pers. comm.).

44

7. Drones not producing semen after manual eversion initial examination

7.1 Introduction
Section 3.4 data showed 41% of drones aged 14-35 days for the four breeding lines, not releasing semen at their endophallus after manual eversion. This experiment was developed to provide initial data on the presence of sperm in the seminal vesicles of mature age drones not releasing semen after manual eversion.

7.2 Method
Drones were reared in six hives in September 2006 and confined in their hives until 21-22 days of age following Section 3 Methodology. Drones sampled from the six hives were combined for examination: (i) Drones were selected at random and everted manually; the proportion of drones producing semen at the endophallus was recorded. (ii) Thirty drones which produced semen at their endophallus after manual eversion were numbered, the volume of semen on the endophallus measured and recorded, the seminal vesicles were dissected from the drones abdomen and the number of sperm present in the seminal vesicles counted and recorded. The intestines were removed from the abdomen and the number of Nosema spp. spores present counted and recorded. (iii) Thirty drones which did not produce semen at the endophallus after manual eversion were numbered, the seminal vesicles dissected from the abdomen and the number of sperm present in the seminal vesicles counted and recorded. The intestine was removed and the number of Nosema spp. spores counted and recorded. Sperm counts and Nosema spore counts. The seminal vesicles or drone intestine, respectively, was placed in 0.5 ml water in a mortar and pestle and crushed. The numbers of sperm, or Nosema spores, were counted in the 5 groups of 16 small squares at the 4 corners and centre of each end of a haemocytometer, total of 160 small squares counted/sample. The haemocytometer was an Improved Neubauer, depth 0.1 mm, 1/400 mm2. The number of sperm present was calculated using Section 3 Methodology, and the number of Nosema spores calculated according to Cantwell (1970).

45

7.3 Results
(i) (ii) Of the 250 drones examined by manual eversion, 173 (69.2%) produced semen at the endophallus. The 30 drones examined which produced semen at the endophallus after manual eversion - volume of semen produced, av. 0.9 (ra. 0.5 1.2) l - 5/30 (16.7%) had sperm present in the seminal vesicles - number of sperm present in the seminal vesicles, av. 0.2 (ra. 0.003 0.09) million - 2/30 (6.7%) contained Nosema spores, 2.28 and 14.05 million spores per drone The 30 drones examined which did not produce semen at the endophallus after manual eversion - 30/30 (100%) had sperm present in the seminal vesicles - number of sperm present in the seminal vesicles, av. 1.38 (ra. 0.01 4.60) million - 7/30 (23.3%) of drones contained Nosema spores, av. 0.59 (ra. 0.03 3.53) million per drone

(iii)

Data for individual drones are shown in Table 1.

7.4 Discussion
When mature age drones do not release semen at the endophallus after manual eversion there is a question of whether the drone did not produce sperm, or did produce sperm and did not release it. Data from Table 1 show that drones that released semen released most of the sperm from their seminal vesicles with only small numbers of sperm found in the seminal vesicles of 5 drones, and all drones which did not release semen after manual eversion contained sperm in their seminal vesicles. The high proportion of drones not releasing semen after manual eversion, 30.8% in this experiment, suggests that factors are involved other than the effect of the operator carrying out the manual eversion (e.g. physiological factors within the drone). Section 3.4 data showing a significant breeding line effect for the number of drones producing semen after manual eversion suggest that this characteristic may be a selectable trait. The Nosema spore levels recorded for 9 of the 60 (15.0%) drones examined, Table 1, does not appear to have had a significant impact on the number of sperm produced by each of the infected drones when compared with the number of sperm present in drones with no Nosema spores present. This may be explained by sperm formation occurring during the pupal stage of drone reproduction, prior to an infection from Nosema disease. The Nosema infection may have affected the drones in areas not examined in this experiment (e.g. reduced ability to fly resulting from lower nutrition uptake).

7.5 Conclusions
(i) (ii) (iii) The majority of drones which released semen at the endophallus after manual eversion released a high percentage of the sperm present in their seminal vesicles. The presence of sperm in the seminal vesicles, by itself, does not ensure the release of semen at the endophallus after manual eversion of mature age drones. Nosema spp. disease did not have an impact on the number of sperm produced by drones.

46

7.6 Tables
Table 1. After manual eversion, the volume of semen at the endophallus, the number of sperm present in the seminal vesicles, and the number of Nosema spores in the intestines of drones producing semen at the endophallus; and the number of sperm in the seminal vesicles and number of Nosema spores in the intestines of drones not producing semen at the endophallus.
No. Drones producing semen at the endophallus Number of Volume of Number of Nosema semen at sperm in spores endophallus seminal 106 /drone l/drone vesicles 106 /drone 1.0 0.0 0.0 0.75 0.0 14.05 0.5 0.0 0.0 1.0 0.0 0.0 0.75 0.0 0.0 1.2 0.005 0.0 1.0 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 1.2 0.0 0.0 1.0 0.0 0.0 1.0 0.005 2.28 1.0 0.0 0.0 1.0 0.0 0.0 0.5 0.0 0.0 0.75 0.0 0.0 1.2 0.0 0.0 1.0 0.0 0.0 0.75 0.003 0.0 0.75 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 0.5 0.0 0.0 1.0 0.005 0.0 0.75 0.0 0.0 1.0 0.0 0.0 0.75 0.09 0.0 0.75 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 Drones not producing semen at the endophallus Number of Number of Nosema sperm in spores seminal 106 /drone vesicles 106 /drone 0.09 0.0 0.08 0.0 2.35 0.0 2.13 0.0 1.19 0.0 0.98 0.0 2.70 3.53 2.11 0.0 0.69 0.0 1.03 0.0 1.25 0.0 1.81 0.0 0.01 0.03 2.16 0.0 0.03 0.0 3.03 0.0 4.60 0.43 0.01 0.03 4.08 0.0 0.88 0.05 2.35 0.03 0.79 0.0 1.53 0.0 0.10 0.0 0.75 0.05 0.55 0.0 2.21 0.0 1.06 0.0 0.03 0.0 0.90 0.0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

7.7 References
Cantwell, G.E. 1970 Standard methods for counting Nosema spores, American Bee Journal, 110, 222223.

47

8. Manual eversion of drones comparison between two operators

8.1 Introduction
Section 3 of this project involved the collection of semen from drones for measurement of semen volume and number of sperm produced by individual drones. One result from this experiment was that for each of the four breeding lines of drones examined at ages 14-35 days of age, only a proportion of drones released a sufficient volume of semen, > 0.1 l, at the endophallus after manual eversion, av. 59 (ra. 39 92)%. The manual eversion of drones for all data collected for Section 3 was carried out by the same operator, Operator 1, a skilled commercial honeybee instrumental insemination expert. This experiment was carried out to compare success rates between Operator 1 and a second person, Operator 2, when manually everting drones.

8.2 Method
In March-April, 2007, drones were reared from one drone mother to a mature age in one hive in a Richmond, NSW apiary using drone rearing methodology described in Section 3. In April, 128 drones, 21-22 days of age were collected and held in a flight box in the laboratory. Operator 1 and Operator 2 each selected 64 drones at random from the flight box and manually everted each drone. Results of each eversion were recorded with eversion ranked on a scale of 0-4; 0 = no eversion; 1 = one quarter eversion; 2 = one half eversion; 3 = three quarters eversion; and 4 = full eversion.

8.3 Statistical analysis

Data were analysed using logistic regression, a generalised linear modelling (GLM) technique.

8.4 Results
Results for the number of drones everted to a scale of 0 4 obtained by each operator is shown in Table 1. Parameters for Operator 1 were significantly different to zero but not significantly different to those of Operator 2 (P = 0.929 and 0.923). There was no significant difference, P 0.05, between the two operators.

8.5 Discussion
As no significant difference was found between the two operators in their ability to manually evert drones, Operator 2 was able to continue experiments involving manual eversion of drones on the basis that drone eversion data obtained from Operator 2 would be comparable to that obtained by Operator 1.

48

8.6 Tables
Table 1. Manual eversion of drones comparison data, number of drone eversions for each scale level recorded by Operator 1 and Operator 2.
Eversion scale 0 1 2 3 4 Total Number of times recorded Operator 1 Operator 2 2 1 0 3 20 11 7 21 35 28 64 64

49

9. Investigation of release of semen by drones after manual eversion.

9.1 Introduction
Data from this project have shown that manual eversion of 848 drones examined at 14-35 days old resulted in 40.6% not releasing semen at the endophallus (Section 3.4), and 250 drones examined at 21 days old resulted in 30.8% not releasing semen at the endophallus (Section 7.3). Thirty drones, 21 days old (Section 7.3) that did not release semen were dissected. All 30 were found with sperm present in their seminal vesicles. The aim of this experiment was to collect data on semen release to provide further information on drone maturation. The age at which drones become mature may be defined as the age a drone is capable of mating with a queen bee under natural conditions. Release of semen after manual eversion is not an indicator of drone maturation as the same values stimulating semen release in a natural mating situation do not apply. A second and more important product of this experiment is examining the accuracy of semen volume and sperm number data obtained from semen samples collected from drones using manual eversion when evaluating drone semen quality for the purpose of selecting queen bees as drone mother queens in bee breeding programs. Bishop (1920) considered that sperm and seminal fluid organ development was complete in drones about 9-12 days after emergence. He examined the response of drones to manual eversion over a range of ages finding , 3 day old, no secretion or only mucous with full eversion only if strong stimulus applied; 5 day old, organ may be caused to extrude with ejaculation of mucous and inactive sperm; 9 day old, organ extruded more readily with ejaculation of active sperm first and mucous second; 12 day old, apparatus apparently mature after 9th day with reaction more easily produced up to the 12th day; 21 day old, same as for 12 day old drones. Mackensen and Roberts (1948) reported that after initial eversion of the endophallus by anaesthesia of the drones, complete eversion was obtained by manual eversion with a great variation in the degree of eversion, the distribution of semen and mucous, and the amount of semen ejaculated. The effects of food and temperature on drone maturity was studied by Jaycox (1961) who assessed drone sexual maturity on the movement of sperm from the testes to the seminal vesicles and vasa deferentia which was completed at age 8-11 days. Drones have completed their physiological development by 12 days of age, Moritz (1989), with sperm completing their migration to the vas deferens at about 2-3 days old and completing their second physiological process during a second stage of maturation. Moritz stated that 12 day old drones may be used for inseminating queen bees. Mackensen and Roberts (1948) commented on the great variation in the degree of drone eversion recorded and the amount of semen ejaculated. Bishop (1920), Mackensen and Roberts (1948), Jaycox (1961), Moritz (1989), Collins and Pettis (2001) and Anderson (2004) commented on drones at the age each specified as reaching maturity not producing semen after manual eversion. Reference was made to the high variation in sperm numbers found between drones by Ruttner (1976), Collins and Pettis (2001), Woyke et al. (2001) and Koeniger et al. (2005). Colonello and Hartfelder (2003) analysed mucous gland protein content and pattern at different ages for drone Africanized honeybees as a means of describing proteinaceous male accessory gland products during sexual maturation. They found protein content peaked at day 5, then decreased reaching a stable level at day 8 which they commented coincided with the age drones initiate first flight activities. Moors et al. (2005) examined the development of mucous glands in drone honeybees each day between days 2-6 and days 9 and 12 after emergence finding accumulation of mucous commenced at the onset of adult life and reached its maximum by day 6.

50

9.2 Methods
Two open-mated sister queen bees reared from unrelated instrumentally inseminated breeder queens of the type Apis mellifera ligustica were obtained from four commercial queen bee producers in eastern Australia. The eight queen bees were introduced into colonies of similar strength at the New South Wales DPI Research Station, Tamworth, New South Wales. Drones were reared from the eight queen bees according to methodology described in Section 3.2 in December 2006 with drones emerging and being marked at one day of age in January, 2007. Ten drones from each breeding line were caught each day for examination when the drones were 3 to 25 days old, and also when the drones were aged 30 days old, a total of 24 times. Line 3 drones were present in their hives and able to be sampled between 3 and 18 days of age only. Drones were caught early in the morning before drone flight commenced and held in a small flight cage, approx. 30 x 30 x 50 cm supported by a cage holding approx. 50 young worker bees with access to queen candy, which fed and supported the drones until examined. Examination comprised manual eversion of the drone, described in Section 3.2 (b), with the level of eversion of the endophallus recorded as a rating of 0-4, 0 = no eversion; 1 = one quarter eversion; 2 = one half eversion; 3 = three quarters eversion and 4 = full eversion. The amount of the endophallus which had everted was dissected from the drone and the number of sperm present determined. The drones abdomen was opened dorsally and the seminal vesicles, vas deferentia and remaining portion of the endophallus dissected from the drone using forceps and the number of sperm present in the removed organs determined. At four intervals when drones were 4, 9, 18 and 23 days old, during dissection of the seminal vesicles the drones intestine was removed and the number of Nosema spp. spores recorded for individual drones. For each drone the three parts, everted portion of the endophallus, seminal vesicles plus attached remaining organs, and the intestine were examined separately. The number of sperm in each part of the drones reproductive system was determined by mashing the dissected organ in a mortar and pestle in 0.5 ml water, and counting the number of sperm present in 10 squares (each square containing 16 small squares) of a haemocytometer, an Improved Neubauer, BS.748, depth 0.1 mm, 1/400 mm2. The total number of sperm present in the organ being examined was calculated using the formula in van Niem Nguyen (1995), modified for the reduced dilution rate. The presence of Nosema spores, Nosema spp., was determined by mashing the drone intestine in 0.5 ml water and using a light microscope at 400 magnifications to identify spores present in the sample.

9.3 Statistical Analysis

Data were analysed with a Generalised Linear Model (GLM), the variables being analysed examined the effects of breeding line, source of sperm (seminal vesicles and endophallus) and the age of the drone (days). A Chi-square analysis was used to test the association between drone age and eversion rating.

51

9.4 Results
1. Comparison of sperm numbers in endophallus and in seminal vesicles
Line effect No significant difference, P = 0.263, between the four drone lines was found for total sperm, Table 1. Source effect The total number of sperm produced per drone, for all drones for all ages, is a factor of the source of the sperm (seminal vesicles or endophallus). A sperm source effect was identified, P < 0.001, with total sperm numbers from the seminal vesicles significantly higher than total sperm numbers from the endophallus, Table 2. Age effect There was a significant linear relationship, P = 0.006, between age and total sperm, for the number of sperm present from both sources (seminal vesicles and endophallus), for all drones, Table 3 and Figure 1. Source.Age effect There was a significant source.age effect, P < 0.001, for the number of sperm present in the seminal vesicles or in the endophallus for all drones, Table 4 and Figure 2. Line.Source effect A significant line.source effect was identified, P = 0.003, for all drones from each line for the number of sperm produced from each source (seminal vesicles or endophallus), for all ages, Table 5. Line.Age effect No significant difference, P = 0.456, was found between the four breeding lines and the 24 ages drones were examined, for sperm produced from both sources (seminal vesicles and endophallus), Table 6. Line.Source.Age effect A significant three way interaction effect was found, P = 0.022, for the source of sperm produced in the seminal vesicles or in the endophallus, for drones from each line examined for each age, Table 7.

2. Nosema disease examination

No Nosema spp. spores were observed for Lines 1, 2, 3 and 4 examined at 4, 9 and 18 days of age, and for Lines 1, 2 and 3 examined at 23 days of age, a total of 148 drones were examined.

3. Endophallus eversion ratings

Eversion effect A significant eversion effect, P < 0.001, was found for the total number of sperm produced by drones for each eversion rating, Table 8. Line.Eversion effect A significant line.eversion effect, P = 0.018, was found for the total number of sperm produced by drones from each line for each eversion rating, Table 9.

52

Age.Eversion effect A significant age.eversion effect, P < 0.001, was found for the total number of drones of each age for each eversion rating, Table 10.

4. Number of sperm produced per drone

The number of sperm produced per drone from the seminal vesicles and the endophallus from all drones examined was, mean 2.12 (ra. 0.00-6.94) million sperm per drone.

9.5 Discussion
Although significant differences, P 0.05, were found between breeding lines for the number of sperm produced by individual drones, Section 3 Table 4, no significant differences were found between the total numbers of sperm produced by each of the four breeding lines of drones examined in this experiment, Table 1. A significant sperm source effect, P < 0.001, was identified with total sperm numbers from the seminal vesicles significantly higher than total sperm numbers from the endophallus for all drones, Table 2. From the results, the ratio of sperm numbers in the seminal vesicles:endophallus is 3.1:1 which is higher than would be expected for the seminal vesicles if manual eversion was an efficient method for retrieving sperm and/or drones matured at a young age (e.g. < 15 days), Table 2. The significant linear relationship, P = 0.006, Table 3 and Figure 1, between drone age and the total number of sperm recorded from both sources, seminal vesicles and endophallus, describes the relationship between each drone producing a fixed number of sperm in the testes which move to the seminal vesicles and remain there until the drone mates with a queen bee or dies. The data show an increase in sperm numbers to about age 11 days and then a gradual decline to age 30 days, Table 3 and Figure 1. Sperm is released from the testes to the seminal vesicles during the first few days after adult drone emergence, this is generally considered to be completed by about day 3 (Moritz, 1989) after emergence. Data from this experiment suggest that movement of sperm from the testes to the seminal vesicles may not be complete until day 5 to 11 after adult drone emergence. The significant sperm source.age effect, P < 0.001, Table 4 and Figure 2, showing the numbers of sperm for all drones present at each age in the seminal vesicles and numbers present in the endophallus resulted in higher numbers of sperm in the seminal vesicles of younger drones and moving towards convergence with increase in drone age. If manual eversion was an efficient method for retrieving sperm from mature age drones and if drones matured at the age generally accepted, 8-12 days, (Bishop, 1920; Jaycox, 1961 and Moritz, 1989), then the seminal vesicles data and the endophallus data in Figure 2 would be expected to show high readings for the seminal vesicles, and low readings for the endophallus, until about days 8-12. At the age when the drones matured the two lines would be expected to cross over with the remainder of Figure 2 showing low sperm numbers for the seminal vesicles and high sperm numbers for the endophallus. Figure 2 shows the two lines continuing to converge past drone age 30 days suggesting that either a proportion of drones matured at different ages, or manual eversion is not an efficient method for collecting semen samples from drones, or a combination of both. A significant line.source effect, P = 0.003, compared the total number of sperm from all drones for all ages from each line present in the seminal vesicles or in the endophallus, Table 5. There were no significant differences between sperm numbers from the endophallus for each line with all endophallus sperm numbers being significantly lower than all seminal vesicle numbers from each line. Line 4 seminal vesicle sperm number is significantly lower than Lines 1, 2 and 3. Regardless of the numbers of sperm in the seminal vesicles, all lines had about the same numbers of sperm in the endophallus and all have more sperm in the seminal vesicles than in the endophallus across all lines. The lower numbers of sperm from the endophallus for all lines for drones aged to 30 days further suggests that manual eversion has a low overall efficiency as a method for retrieving sperm from drones.

53

No significant line.age effect was identified, P = 0.456, when the total sperm from the seminal vesicles and the endophallus from each line for each age was examined, Table 6, with the effect of increasing age being the same for each line. A significant three way interaction effect was found for Line.Source.Age, P = 0.002, for the source of sperm produced in the seminal vesicles or in the endophallus for drones from each line examined for each age, Table 7. For the 148 drones examined for Nosema disease at 4 ages, no Nosema spp. spores were observed indicating that Nosema disease was not a factor in affecting sperm count numbers recorded in this experiment. A significant eversion effect, P < 0.001, for the total number of sperm produced by drones for each eversion rating, Table 8, shows the highest eversion rating, rating 4, has the lowest number of sperm associated with it, as well, the lowest eversion rating, rating 0, has the highest number of sperm associated with it. This result may be explained, in part, by the range of ages of drones contributing data, 3 to 30 days of age. A significant line.eversion effect, P = 0.018, was found for the total number of sperm produced by drones from each line for each eversion rating, Table 9. The inflated prediction for line 4 drones for eversion rating 4 may have resulted from line 4 drones not being available for sampling after age 18 days. A significant age.eversion effect, P < 0.001, was found for the total number of drones of each age for each eversion rating, Table 10. The data suggest that large numbers of drones at a young age are at the lower eversion rating and large numbers of older age drones are present at the higher eversion rating. These data support drone maturation at an age after emergence however the change in numbers of drones in the low eversion ratings and drones in the high eversion ratings is gradual and do not identify a specific age at which drones change from low release of semen to high release of semen suggesting that drone maturation occurs over a wide range of drone ages. The significant sperm source effect, Table 2; linear relationship for Age effect, Table 3; and the Line.Source effect, Table 5, all strongly suggest that either manual eversion of drones is not an efficient method for effecting movement of sperm from the seminal vesicles to the tip of the endophallus in semen, or, manual eversion is an efficient method and the lack of movement of sperm is due to physiological factors within the drone, i.e. drones which have not reached sexual maturity are restricted in the amount of sperm they release with manual eversion. Data from the Source.Age effect, Table 4 and Figure 2, showing sperm remaining in the seminal vesicles of drones up to 30 days of age, despite inefficiencies associated with manual eversion, suggest that drone maturation occurs over a wide range of ages for different drones. Data from this experiment suggest that manual eversion is not a suitable method for collecting semen samples required to provide accurate semen volume and sperm number data for individual drones, with dissection and counting from the seminal vesicles providing a more reliable alternative. Although no significant line effect, Table 1, was found for the total number of sperm produced from each drone line in this experiment, a significant effect, P 0.05, was found for the volume of semen and number of sperm produced per drone for drones from different breeding lines, Section 3, Tables 3 and 4. The importance of the proportion of semen released at the endophallus after manual eversion is in the accuracy of data on semen volume and sperm numbers produced per drone when these data are used for the selection of queen bees as drone mother stock for bee breeding programs when selecting for drones which produce high volumes of semen and high numbers of sperm. Low numbers of sperm per drone were recorded, mean 2.12 (ra. 0.00 6.94) million sperm per drone from drones aged between 3 and 30 days and did not take into account sperm which may have been present in the testes of the younger aged drones. These data suggest that selection of drone mother queen bees whose drones produce large numbers of sperm should be included in breeding stock selection criteria for breeding programs. 54

9.6 Conclusions
(i) (ii) (iii) Movement of sperm from the testes to the seminal vesicles is not completed until about drone age day 5 to 11. Older aged drones have higher numbers of sperm released to the endophallus after manual eversion than younger age drones. Release of sperm from the seminal vesicles to the endophallus after manual eversion occurs over a range of ages with no specific drone age identified when drone maturation, sperm released from a significant number of drones, occurred. Manual eversion of drones was identified as not an efficient method for obtaining accurate samples of semen for determination of semen volume and number of sperm produced per drone. Older aged drones have a higher endophallus eversion rating after manual eversion than younger aged drones. No specific drone age was identified when a change from low eversion rating to high eversion rating occurred. For drones 3 to 30 days of age, the number of sperm present in the seminal vesicles and/or endophallus was low, mean 2.12 (ra. 0.00-6.94) million sperm per drone.

(iv)

(v)

(vi)

55

9.7 Tables
Table 1. Line effect. Predicted effect of breeding line on the number of sperm present from both sources (seminal vesicles and endophallus) for all drones for all ages.
Line 1 2 3 4 s.e.d. Predicted value* 777.18 833.15 761.67 733.03 Mean 45.38 min. 38.85 max. 51.72

Data were square root transformed

Table 2. Sperm Source effect. Predicted values of sperm source, seminal vesicles v. endophallus, for all drones for all ages.
Source Seminal vesicles Endophallus s.e.d. Predicted value* 1177.92 374.60 Mean 31.51 min. 31.51 max. 31.51

Data were square root transformed

56

Table 3. Age effect. Predicted effect of age on the number of sperm present in both sources (seminal vesicles and endophallus) for all drones for each age.
Age (days) 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 s.e.d. Predicted value* 379.78 475.18 561.23 633.79 691.89 735.66 765.39 782.73 789.96 789.36 782.90 772.36 759.51 745.95 733.12 722.43 714.83 710.76 710.56 712.96 714.13 710.02 698.07 549.51 Mean 54.68 min. 9.97 max. 115.11

Data were square root transformed

57

Table 4. Source.Age effect. Predicted effect of age on the number of sperm present in the seminal vesicles or in the endophallus for all drones for each age.
Age (days) Predicted value* Seminal vesicles 939.15 1019.72 1090.93 1148.65 1191.92 1220.86 1235.75 1238.25 1230.65 1215.21 1193.91 1168.54 1140.85 1112.45 1084.78 1059.25 1036.83 1017.91 1002.88 990.44 976.77 957.83 931.04 708.30 Mean 66.40 Endophallus -179.60 -69.36 31.52 118.92 191.86 250.47 295.04 327.21 349.28 363.51 371.89 376.19 378.18 379.45 381.45 385.60 392.84 403.60 418.24 435.48 451.48 462.21 465.09 390.72 min. 10.31 max. 136.41.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 s.e.d.

Data were square root transformed

Table 5. Line.Source effect. Predicted effect of line on the number of sperm present in the seminal vesicles and in the endophallus for drones for all ages.
Line Predicted value* Seminal Endophallus vesicles 1269.98 284.37 1251.00 415.31 1118.97 404.37 1071.73 394.33 Mean min. 54.57 63.17 max. 82.24

1 2 3 4 s.e.d.

Data were square root transformed

58

Table 6. Line.Age effect.Predicted effect of drones from four breeding lines on the age drones were examined for sperm produced from both sources (seminal vesicles and endophallus).
Age (days) 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 s.e.d. 1 339.71 438.97 528.87 605.29 667.25 714.88 748.46 769.66 780.74 784.00 781.39 774.71 765.71 756.01 747.03 740.19 736.45 736.23 739.89 746.14 751.16 750.91 742.81 613.53 mean 84.81 Predicted value* Line 2 3 454.24 316.31 547.99 416.32 632.38 506.96 703.29 584.12 759.74 646.82 801.86 695.19 829.94 729.52 845.62 751.45 851.20 763.28 848.95 767.28 840.83 765.42 828.64 759.48 814.14 751.23 798.93 742.26 784.44 734.03 773.00 727.93 762.85 724.94 757.12 725.46 755.27 729.86 756.02 736.86 755.53 742.62 749.78 743.12 736.17 735.76 579.35 610.20 min. max. 10.55 266.34

4 408.84 497.45 576.69 642.45 693.75 730.72 753.65 764.19 764.62 757.22 743.96 726.62 706.97 686.61 666.98 649.49 635.09 624.22 617.22 612.82 607.19 596.28 577.53 394.98

Data were square root transformed

59

Table 7. Line.Source.Age effect. Predicted three way interaction effect between line, source and age.
Age (days) Sem. Ves. 922.29 1013.11 1094.57 1162.54 1216.05 1255.23 1280.37 1293.12 1295.76 1290.57 1279.52 1264.39 1246.95 1228.80 1211.37 1196.09 1183.91 1175.24 1170.45 1168.26 1164.84 1156.14 1139.59 968.09 mean 111.91 Line 1 Endo. -242.87 -135.16 -36.82 48.04 118.45 174.52 216.55 246.19 265.73 277.42 283.26 285.02 284.48 283.22 282.68 284.29 289.00 297.22 309.32 324.02 337.49 345.69 346.03 258.98 min. 11.18 Sem. Ves. 1093.85 1166.75 1230.28 1280.32 1315.91 1337.17 1344.38 1339.20 1323.92 1300.80 1271.82 1238.77 1203.41 1167.33 1131.98 1098.77 1068.66 1042.07 1019.36 999.24 977.89 951.27 916.80 655.67 max. 350.11 2 Endo. -185.38 -70.76 34.49 126.26 203.58 266.56 315.50 352.04 378.49 397.10 409.84 418.52 424.88 430.53 436.90 445.42 457.04 472.17 491.19 512.80 533.17 548.28 555.53 503.03 Sem. Ves. 747.92 840.94 924.59 994.76 1050.47 1091.85 1119.18 1134.13 1138.97 1135.98 1127.12 1114.19 1098.95 1083.00 1067.77 1054.69 1044.70 1038.24 1035.65 1035.66 1034.43 1027.93 1013.58 853.07 3 Endo. -115.29 -8.30 89.34 173.48 243.18 298.54 339.85 368.78 387.60 398.59 403.71 404.76 403.50 401.53 400.28 401.18 405.17 412.69 424.08 438.07 450.82 458.30 457.93 367.32 Sem. Ves. 992.54 1058.10 1114.29 1157.00 1185.26 1199.18 1199.07 1186.56 1163.94 1133.49 1097.18 1056.79 1014.10 970.69 928.01 887.47 850.03 816.11 786.06 758.61 729.93 695.98 654.18 356.39 4 Endo. -174.86 -63.20 39.09 127.89 202.25 262.26 308.24 341.83 365.31 380.95 390.74 396.45 399.85 402.53 405.95 411.50 420.16 432.33 448.38 467.03 484.44 496.59 500.88 433.56

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 s.e.d.

Data were square root transformed

60

Table 8. Eversion effect. Predicted values for the total number of sperm produced for each eversion rating.
Eversion rating 0 1 2 3 4 s.e.d. Predicted value ** 1623 1443 1349 1272 786 Mean 105.2 min. 67.7, max. 145.8 Sig.* a ab ab b c

* **

Different letters are significantly different, P 0.05 Data were square root transformed

Table 9. Line.Eversion effect. Predicted values for the total number of sperm produced by drones from each line for each eversion rating.
Line 0 1570 1720 1595 1607 Mean 205.6 Predicted value* Eversion rating 1 2 3 1470 1272 1505 1445 1424 1545 1333 1298 1098 1524 1402 940 min. max. 89.1 433.5

1 2 3 4 s.e.d.

4 224 728 847 1344

Data were square root transformed

61

Table 10. Age. Eversion effect. The number of drones recorded at each age for each eversion rating.
Age (days) 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 0 1 1 5 10 11 14 11 10 13 7 9 10 10 10 5 9 0 8 1 4 0 1 0 1 1 17 10 7 6 4 5 6 5 6 4 9 7 7 3 7 3 4 2 2 4 0 4 0 1 Number of drones Eversion rating 2 3 11 0 29 0 28 0 24 0 16 8 13 8 14 7 13 11 10 9 15 12 10 9 12 8 12 6 16 4 15 4 17 2 13 3 8 2 9 3 11 0 9 0 6 1 2 3 8 0

4 1 0 0 0 1 0 2 1 2 2 2 3 5 7 9 7 10 10 15 11 21 18 5 20

62

Figure 1. Age effect. Cubic smoothing spline describing change in sperm count with age of the drone.

Figure 2. Source.Age effect. Cubic smoothing spline describing changes in sperm numbers with drone age (days) for source of sperm (seminal vesicle or endophallus).

63

9.8 References
Anderson, D. 2004 Improving queen bee production, Publication No. 04/153, Rural Industries Research and Development Corporation, Canberra, ACT, Australia. Bishop, G.H. 1920 Fertilization in the honeybee, Journal of Experimental Zoology, 31, 225-266. Cantwell, G.E. 1970 Standard methods for counting Nosema spores, American Bee Journal, 110, 222223. Collins, A.M. and Pettis, J.S. 2001 Effect of Varroa infestation on semen quality, American Bee Journal, 141 (8), 590-593. Colonello, N.A. and Hartfelder, K. 2003 Protein content and pattern during mucus gland maturation and its ecdysteroid control in honeybee drones, Apidologie, 34, 257-267. Jaycox, E.R. 1961 The effects of various foods and temperatures on sexual maturity of the drone honeybee (Apis mellifera), Annals of the Entomological Society of America, 54 (4), 519-523. Koeniger, G., Koeniger, N., Tingek, S. and Phiancharoen, M. 2005 Variance in spermatozoa number among Apis dorsata drones and among Apis mellifera drones, Apidologie, 36, 279-284. Mackensen, O. and Roberts, W.C. 1948 A manual for the artificial insemination of queen bees, USDA ET-250, Bureau of Entomology and Plant Quarantine, USA. Moors, L., Spaas, O., Koeniger, G. and Billen. J. 2005 Morphological and ultrastructural changes in the mucus glands of Apis mellifera drones during pupal development and sexual maturation, Apidologie, 36, 245-254. Moritz, R.F.A. 1989 The instrumental insemination of the queen bee, Apimondia Publishing House, Bucharest, Romania. Ruttner, F. 1976 The instrumental insemination of the queen bee, Apimondia, Bucharest. van Niem Nguyen 1995 Effect of protein nutrition and pollen supplementation of honeybee (Apis mellifera L.) colonies on characteristics of drones with particular reference to sexual maturity, MSc Thesis, University of Western Sydney, Hawkesbury, NSW, Australia. Woyke, J., Wilde, J. and Wilde, M. 2001 Apis dorsata drone flights, collection of semen from everted endophalli and instrumental insemination of queens, Apidologie, 32, 407-416.

64

10. Seminal vesicles v. endophallus sperm viability and motility.

10.1 Introduction
An investigation of the movement of sperm and seminal fluid after manual eversion of drones 3 to 30 days of age showed a significant sperm Source.Age effect, P < 0.001, with higher numbers of sperm in the seminal vesicles of younger aged drones and higher numbers of sperm at the endophallus in older age drones (Section 9, Table 4 and Figure 2). For all ages between 3 and 30 days, there was a proportion of drones with sperm remaining in the seminal vesicles after manual eversion suggesting that the problem of sperm retention in the seminal vesicles of older age drones may result from either manual eversion being an inefficient method for retrieving sperm, and/or there is a problem with the sperm (e.g. low viability and/or low motility). This experiment was carried out in 2007 and examined the viability and motility of sperm present in the seminal vesicles and in the endophallus of individual mature age drones after manual eversion.

10.2 Method
Drones were reared to 28 days of age from the same drone mother, a commercially reared open-mated queen of the type Apis mellifera ligustica produced from an instrumentally inseminated breeder queen. Drones were reared according to the method described in Section 3. Twenty four drones were examined. Drones were caught early in the morning and held in a flight cage supported by young worker bees with access to queen candy until examined. Examination comprised manual eversion of the drone, described in Section 3.2 (b). For each drone (i) the amount of endophallus which everted was dissected from the drone, placed in 50 l buffer, formulae described in Section 3.2 (b), mashed, and divided into 40l for sperm viability assessment and 10 l for sperm motility assessment, and (ii) the seminal vesicles were dissected from the drone, placed in 50 l buffer, mashed, and divided into 40l for sperm viability assessment and 10 l for sperm motility assessment. Sperm viability assessment was carried out using a Live/Dead Sperm Viability Kit containing a membrane-permeant nucleic acid stain and a dead cell stain, described in Section 4.2. Sperm motility was assessed by scoring sperm movement at room temperature at five levels of movement, described in Section 5.2.

10.3 Statistical analysis

Sperm viability. Data were analysed with a Generalised Linear Model (GLM) to test the effects of sperm source (endophallus or seminal vesicles), condition (dead or alive), and the interaction between them. Data were square root transformed to satisfy some of the assumptions underlying the analysis. Sperm motility. Data were analysed with a Generalised Linear Model (GLM) to test the effects of sperm source (endophallus or seminal vesicles) with motility ratings as the response variable.

65

10.4 Results
1. Viability Source effect. A significant source effect was identified, P < 0.001, with total sperm numbers present in the seminal vesicles significantly higher than total sperm numbers present in the endophallus, Table 1. Condition effect. A significant condition effect was identified, P < 0.01, with a significantly higher number of sperm alive than dead, Table 2. Source.Condition effect. No significant interaction between source and condition was found, P = 0.346, Table 3. 2. Motility Source effect. No significant interaction between source and motility was found, P = 0.516, Table 4.

10.5 Discussion
Sperm viability analysis results indicate there are significantly more sperm in the seminal vesicles than in the endophallus, Table 1. Significantly more sperm were found alive than dead, Table 2. Data in Table 2 represents 32.7% of all sperm dead which, although high, is within the range of sperm viability, av. 79.7 (ra. 59.9-90.5) % recorded in this project, Section 4.4. Drones for this experiment were reared from one drone mother queen indicating that individual queen bees are producing drones with unacceptably high numbers of dead sperm, further suggesting the importance of sperm viability as a selection criterion when selecting drone mother queen bees for breeding programs. There was no significant interaction between sperm source and condition with similar proportions of dead sperm present in the seminal vesicles as in the endophallus. This indicates that sperm in the seminal vesicles after manual eversion does not remain there due to low viability and suggests that drone physiology i.e. drone immaturity, or ineffectiveness of the manual eversion method for retrieving sperm may be the cause. No significant difference, P = 0.05, was found between the motility of sperm from the seminal vesicles and sperm from the endophallus. No significant interaction between the source of the sperm i.e. sperm remaining in the seminal vesicles or sperm which had moved to the endophallus with either sperm viability or sperm motility, indicates that lower sperm quality, measured in terms of viability and motility, is not the cause of sperm remaining in the seminal vesicles of mature age drones following manual eversion.

10.6 Conclusions
(i) Lower sperm quality, measured in terms of sperm viability and motility, are not the cause of sperm remaining in the seminal vesicles of mature age drones after manual eversion. Individual queen bees are able to produce drones with unacceptably high levels of dead sperm in their seminal vesicles.

(ii)

66

10.7 Tables
Table 1. Predicted values of sperm source, for total numbers of sperm present in the seminal vesicles v. endophallus.
Source Seminal vesicles Endophallus s.e.d. * ** Predicted value** 2.80 1.55 0.12 Sig.* a b

Different letters are significantly different, P < 0.001 Data were square root transformed

Table 2. Predicted values of sperm condition (alive or dead).

Condition Dead Alive s.e.d. * ** Predicted value** 2.57 1.78 1.78 Sig.* a b

Different letters are significantly different, P < 0.001 Data were square root transformed

Table 3. Predicted effect of sperm condition (alive or dead) for sperm present in both sources (seminal vesicles and endophallus).
Predicted value* Alive Dead Seminal vesicles 3.25 2.35 Endophallus 1.89 1.22 s.e.d. Mean min. max. 0.17 0.17 0.17 * Data were square root transformed Source Sig. a a

Table 4. Predicted effect of sperm source (seminal vesicles and endophallus) for motility rating.
Source Seminal vesicles Endophallus s.e.d. Predicted value 1.65 1.90 0.37 Sig. a a

67

11. Drone honeybee semen amino acid and fatty acid content.
11.1 Introduction
The objectives of this experiment were to examine changes in amino acid and fatty acid content of drone semen with drone age and season to identify differences which may occur and affect drone semen quality. Woyke and Jasinski (1978) investigated the effects of increasing drone age on the number of sperm entering a queen bees spermatheca and the amount of residual semen in the oviducts after instrumental insemination. They found a difference in semen composition with age with semen from drones 10-14 days old being liquid, yellow-cream in colour and easily taken into the tip of a syringe, semen from drones 28 days and older was thick, cream-brown in colour and more difficult to take into a syringe tip. They reported that as the age of the drone increased, mostly, a lower number of sperm entered the spermatheca of queen bees and the percentage of queen bees with residue in their oviducts increased from 0-14% of queens inseminated with semen from drones 14 days old to 43-67% of queen bees inseminated with semen from drones 28 days old, variations were found for different seasons. An early investigations into the composition of honeybee drone semen involved separation of seminal fluid from the sperm by centrifugation and determining the free and bound amino acids present in each, e.g. Novak et al. (1960) identified 15 amino acids at different levels in both the seminal fluid and sperm. Levels of arginine (39.2 mg/g), lysine (12.6 mg/g) and leucine (12.5 mg/g) in sperm and arginine (1.37 mg/g) and glutamic acid (1.03 mg/g) in seminal fluid were high in comparison with other amino acids present in sperm and seminal fluid respectively. Crailsheim and Leonard (1997) identified amino acids in worker honeybee haemolymph. They found proline was the predominant amino acid comprising from 50% in newly emerged bees to up to 80% of total amino acids from the third day and decreasing in older bees. The bulk of amino acids with lower concentrations showed either no change in concentration or were higher in newly emerged bees and decreased during the lifespan of the bees. Eighteen amino acids were identified with ten described by de Groot (1953, quoted in Crailsheim and Leonard, 1997) as essential for the growth of honeybees in diet experiments, arginine, histidine, lysine, tryptophane, phenylalanine, methionine, threonine, leucine, isoleucine and valine. The remaining eight identified amino acids were aspargine, glutamine, serine, glycine, alanine, proline, cystine and tyrosine. Amino acids present in drone honeybee haemolymph were examined by Leonard and Crailsheim (1999) finding concentrations of free amino acids higher than in worker bee haemolymph and with different relative proportions of individual amino acids. The overall concentration of free amino acids reached its highest level at the fifth day after adult drone emergence with only minor changes in concentration occurring after the ninth day, with the ninth day being about the age at which drones mature. The most abundant amino acid was proline with the average content rising from the first day to a maximum at seven days, then continuously decreasing with age to 25 day old drones. Essential free amino acids described by de Groot (1953) were at their highest levels in one and three day old drones, then decreasing to seven day old drones and not changing significantly after seven days. A comparison of the free amino acids in five day old drones and five day old worker bees showed significant differences in the levels of all amino acids except tryptophane and tyrosine, with the concentrations of amino acids in drone haemolymph generally higher than levels in worker bees. Somervile (2004) analysed 182 pollen samples from 61 identified plant species commonly foraged by honeybees in New South Wales and identified 17 amino acids present (amino acids described in Table

68

1 with the exceptions of asparagine, glutamine and tryptophan). He found isoleucine the most frequently limited amino acid with 66 samples with levels of isoleucine, 11 samples with valine and 2 samples with methionine below desirable levels in de Groot (1953). Somerville (2004) concluded that results indicated that most of the essential amino acids in honeybee collected pollen samples were at levels sufficient to meet honeybee nutritional requirements outlined by de Groot (1953). Blum et al. (1962) studied chemistry of the drone reproductive system with an emphasis on carbohydrates. They found three sugars, fructose, glucose and trehalose present in all the anatomical areas of the reproductive system as well as in the ejaculated semen, occurring almost exclusively in the seminal fluid. Several metal ions including iron, magnesium, copper, calcium, manganese and sodium were present in the ejaculated semen. Blum et al. (1967) examined fatty acids in drone semen. Studies on the oxidation of free sugars in seminal fluid (Blum et al. 1962) indicated that honeybee sperm utilize intracellular compounds as an energy source for sperm motility, this suggested to Blum et al. (1967) that fatty acids may be the source of the endogenous energy. An initial investigation to determine the fatty acid composition of semen showed that phospholipids constituted the major fraction present in the lipids of honeybee semen. They found the phospholipids contain no appreciable amounts of short-chain fatty acids with C16:0 (23%) and C18:1 (56%) accounting for 80% of the total fatty acids. A review of the importance of fatty acids for honeybees (Manning, 2001a) reported that lipids are important primarily as a source of energy with some components of lipids involved in the synthesis of reserve fat and glycogen, and in the membrane structure of cells. Fatty acids and sterols were found important in honeybee development, nutrition and reproduction. He found linoleic fatty acid, C18:2, the dominant fatty acid in Eucalyptus spp. pollens, comprising 44% of total fatty acids present. Manning (2001b) presented data showing the main fatty acids present in Eucalyptus spp. pollens are C14:0, C16:0, C18:0, C18:1, C18:2, C18:3, C20:0 and smaller amounts of C22:0. Blum et al. (1962) concluded that the seminal fluid is primarily derived from the seminal vesicles and penis bulb and probably contains traces of mucous gland secretion. The roles of chemicals present in the male accessory glands are of major interest for understanding the importance of the amount of time, measured in a small number of days, a queen bee has available to complete the mating process successfully, i.e. to achieve the maximum number of sperm stored in her spermatheca, and the subsequent importance on this factor of the availability of sufficient numbers of mature age drones able to mate with the queen during that time frame. As well, there is the further consequence of the combination of these factors for successful mating of large numbers of queen bees over the same time period under commercial queen bee rearing conditions. Gillott (1996) reported on the functions of male accessory glands in insects categorizing available information into functions of structural, biochemical, physiological and behavioural with some overlap between these. Among the physiological roles was facilitation of sperm transfer to the spermatheca. Gilliot stated that for representatives from some Orders, e.g. Orthoptera, Diptera and Lepidoptera, male accessory glands material transferred during mating significantly modifies the mated females behaviour. These molecules, usually proteins or peptides, serve to promote egg production or render the female unwilling to remate, the same molecule appearing to have both roles in some Diptera. Of importance to this experiment is the role these chemicals have in triggering oviposition. Rice (1996) experimenting with Drosophila melanogaster found indirect evidence indicating that seminal fluid reduces the competitive ability of sperm from another male, reduces a females propensity to remate, and increases her egg laying rate. Tram and Wolfner (1998) with D. melanogaster examined postmating changes in female insects triggered by seminal fluid proteins from the males accessory gland proteins and by sperm. Their results demonstrated that copulation, without male accessory gland fluids or sperm, was sufficient to produce a short term decrease in attractiveness of mated females, with the maintenance of decreased attractiveness requiring the presence of sperm. 69

Chemicals present in the mating plug transferred by male bumblebees Bombus terrestris into the queen bees sexual tract shortly after sperm transfer were identified by Baer et al. (2000) with the main compounds being four fatty acids, palmitic, linoleic, oleic and stearic acids, and a cyclic peptide, cyclopropylproline. Further investigations by Baer et al. (2001) found linoleic acid the only substance in the mating plug that decreased female remating behaviour, even when they did not receive sperm. No effects were detected from the remaining three fatty acids and the peptide. Gillott (2002) stated that the Accessory Reproductive Glands (ARG) of male insects produce secretions essential for the transfer of sperm to the female. In species where sperm are transferred in a liquid medium, such as in honeybees, the seminal fluid is also a product of the ARG although components secreted by other regions of the male reproductive tract, e.g. the ejaculatory duct, may be added. One reported role of the ARG is the acceleration of egg production. The clearest demonstration of involvement of ARG products in stimulating egg development is in D. melanogaster, oocytes gradually develop during sexual maturation but unless mating occurs a large proportion of the oocytes do not mature. Sperm is also important in regulating oocyte development, females mated with spermless males showed oocyte progression significantly lower in the absence of sperm.

11.2 Methods
Semen collection. Drones were reared to provide semen at the required age and season from daughter queen bees from instrumentally inseminated breeding queen bee stock of the type Apis mellifea ligustica produced by commercial queen bee producers in eastern Australia. Drone rearing and semen collection methods were similar to those described in Section 3.2. Data for this experiment are based on semen samples collected Drone rearing
1 2

Date sample collected

9.11.2003 17.11.2003 6.1.2004 13.1.2004 27.1.2004 18.5.2004 25.5.2004 31.10.2006 13.11.2006 26.4.2006 3.5.2006 17.5.2006

Drone age (days)

14 21 14 21 35 14 21 21 35 14 21 35

Season
Spring :: Summer :: :: Autumn :: Spring :: Autumn :: ::

3 4 5

Between the dates of semen collection and forwarding to the analytical laboratory, samples were maintained at -20C. Semen samples were forwarded to the respective analytical laboratory in individual glass capillary tubes prepared with Vaseline at tip, air space, buffer, air space, semen, and reverse procedure. Samples were packed in dry ice and transported by an overnight freight service. Buffer composition was:
Sodium chloride 11.0 g; glucose 1.0 g; L (+) Arginine-HCl 0.1 g; L (+) Lysine 0.1 g; Tris (hydroxymethyl) aminomethane (Base 7-9) 4.9 g; Tris (hydroxymethyl) aminomethane hydrochloride 1.5 g; in 1000 cc distilled water, Ph 8.7.

70

Fatty acid profile. The fatty acid profile was determined as fatty acid methyl esters by gas chromatography described in Mailer et al. (2002). Separation of the fatty acid methyl esters was performed on a Varian 3800 Gas Chromatograph using a Supelco BPX 70 chromatography column (30 m by 0.22 mm; 0.25 m film) and flame ionisation detector (FID). The column temperature was programmed at 185oC for 8 min, then increased at 10oC per min up to 220oC. It was held for 3 min before cooling. The injector (split mode) temperature was set at 240oC with a split ratio of 1:50. Detector temperature was 250oC. Data was analysed using Star Workstation Chromatography software (Version 4.51). Fatty acid analysis was carried out by the Department of Primary Industries Agricultural Institute, Wagga Wagga, New South Wales, Australia. Free Amino acid profile. The 2003-04 sample analysis was performed using the Waters AccQTag chemistry, data is presented as free amino acids. Samples were analysed in duplicate and results expressed as an average. The 2006 samples underwent 24 hr liquid hydrolysis in 6M HCl at 110C. After hydrolysis, all amino acids were analysed using the Waters AccQTag chemistry, data are presented as free amino acids plus water. With some exceptions, samples were analysed in duplicate with results expressed as an average. Amino acid analysis was carried out by the Australian Proteome Analysis Facility Ltd, Macquarie University, Sydney, NSW, Australia.

11.3 Results
Amino acids Free amino acids identified and amounts present in each sample of drone honeybee semen were recorded for drone ages 14, 21 and 35 days, and three seasons, spring, summer and autumn. Results for samples collected for 2003-04 are shown in Table 1, and for 2006, amounts presented as free amino acids plus water, are shown in Table 2. Eighteen amino acids were identified with the methodology used for the 2003-04 sample analysis, and 16 amino acids identified with the methodology used for the 2006 sample analysis. Amino acids identified and the percent of each amino acid of the total amino acids identified in each semen sample are shown in Figures 1 and 2. Fatty acids The fatty acids identified and the percent of each present in each sample for three drone ages, 14, 21 and 35 days, and three seasons, spring, summer and autumn, for the years 2003-04 are shown in Table 3, and for three ages and two seasons for the year 2006 in Table 4. The combined data from 2003-04 and 2006 showing the changes in the percent of each fatty acid for the three ages and three seasons are shown in Figure 3.

11.4 Discussion
Eighteen free amino acids were identified in semen samples collected during 2003-04 by the methodology used for analysis of these samples, Table 1 and Figure 1. A high content percent of arginine, av. 51.2 (ra. 43.2-57.5)%, and proline, av. 20.0 (ra. 17.5-23.5)%, was found with lower percent amounts, less than 7%, of total amounts of each of the remaining 16 amino acids for the 6 ages and seasons samples were analysed, Table 1. Possible contamination of the semen samples with arginine-HCl and lysine from buffer in the capilliary tubes during storage should be taken into consideration when interpreting the high arginine levels and the lysine levels identified in the semen samples. The low range of variations which occurred between samples from each age and season for each amino acid may be explained by pollens from different plant species being foraged and consumed during the time of rearing and sampling the adult drones. The results suggest that the 18 amino acids identified are important for the production of drone semen with emphasis on the larger amounts of arginine and proline present.

71

Sixteen free amino acids plus water were identified in semen samples collected during 2006 by the methodology used for analysis of these samples, Table 2 and Figure 2. Results show, in general, a low range of differences between samples from the 8 ages and seasons sampled which may be explained by different plant species pollen being foraged and consumed at the time the drones were sampled. The 16 amino acids identified in the 2006 samples support the 18 amino acids identified in the 200304 samples. Novak et al. (1960) identified 15 amino acids present at different levels with higher amounts of arginine, lysine, leucine and glutamic acid, this experiment found high levels of arginine and proline only. Crailsheim and Leonard (1997) identified 18 amino acids in worker bee haemolymph and reported the 10 amino acids de Groot (1953) described as essential for the growth of the honeybee. This experiment which examined amino acids present in drone semen identified two extra amino acids, aspartic acid and glutamic acid, and found proline, which was not included in de Groots ten essential amino acids for honeybee growth, to be the second highest in amounts of each amino acid present in each semen sample. Leonard and Crailsheim (1999) found essential amino acid levels described by de Groot (1953) in drone haemolymph not changing significantly after 7 days of age, and with proline the most abundant amino acid, this experiment found low levels of change of amounts of amino acids in semen from drones aged 14 to 35 days, and with proline the second most abundant amino acid in drone semen. No differences, in general, were identified between the amino acid levels present in semen from drones reared from queen bees supplied by two queen bee breeders, A and B. Fatty acids Identification of fatty acids in drone semen for the 2003-04 and 2006 years provided a total of 32 fatty acids identified, and unidentified peaks present at 9 points, Tables 3 and 4. Each fatty acid, as a percent of the total fatty acids identified in each sample, in general, varied for drone age, season and year sampled. Oleic acid (C18:1n9c) was present in the highest amounts and varied considerably between years, 2003-04 av. 65.5 (ra. 48.5 80.1)% and 2006 av. 20.8 (ra. 10.0 32.4)%, elaidic acid (C18:1n9t) was present in low amounts in 2003-04 av. 2.6 (ra. 1.1 4.2)% and higher amounts in 2006 av. 37.1 (ra. 8.1 58.3)%, all other identified fatty acids peaked at less than 7% for all data for 2003-04 and 2006, Figure 3. Changes in amounts between each fatty acid for drone age, season and year may result from differences in plant species pollen being foraged and consumed during the time of rearing and sampling the adult drones. These results identify fatty acids important for drone semen content and emphasise two fatty acids, oleic and elaidic acids, present in drone semen in larger amounts than other identified fatty acids. The changes in amounts of fatty acids present at different ages supports findings by Woyke and Jasinski (1978) that semen composition changes with increase in drone age. Blum et al. (1967) found C16:0 (23%) and C18:1 (56%) were highly represented and this experiment found C18 fatty acids the most prominent. The significance of high amounts of arginine and proline amino acids and oleic and elaidic fatty acids in drone semen is not understood. Research is required to identify benefits for drones in drone rearing colonies at queen bee mating apiaries from supplementary feeding with the four identified substances. The importance of a sufficient number of mature age drones each with high numbers of sperm and high sperm viability and motility at queen bee mating apiaries is emphasised by research data on Drosophila melanogaster which indicates that chemicals present in the seminal fluid and in male accessory glands material transferred to the female during mating modifies the mated females behaviour by reducing the females propensity to remate and acceleration of egg production (Gillott 1996, 2002; Rice, 1996 and Baer et al. 2001). Should chemicals present in the seminal fluid and accessory glands of drone bees have similar effects on queen bees at mating, then it may be that the trigger mechanism which stops a queen from continuing mating and begin egg laying may commence functioning only after the queen has mated for the first time. An effect of the chemicals of acceleration 72

of egg production after mating the first time may result in the queen bee having only a pre-determined and limited amount of time, e.g. a small number of days, in which to complete the mating process once she has mated with the first drone. The process of a limited amount of time in which to complete mating after mating with the first drone explains results from Rhodes (2002) in which a high proportion of queen bees 14 to 35 days of age had low numbers of sperm present in their spermathecae.

11.5 Conclusions
(i) (ii) (iii) (iv) The significance of higher levels of arginine and proline amino acids, and oleic and elaidic fatty acids in drone semen require identification and attention. Dietary requirements of drone honeybees reared for breeding purposes require attention in relation to the eighteen amino acids and thirty two fatty acids identified in drone semen. Changes in amounts of amino acids and fatty acids in drone semen with drone age and season suggest changes occur in semen quality with drone age and season. The amount of time, hours, available to a queen bee to complete the mating process between mating for the first time and terminating mating flights requires identification. The less time available to a queen bee to complete the mating process corresponds to an increase in the importance of sufficient numbers of drones of adequate quality present at the mating site at the times the queen is on its mating flights.

73

11.6 Tables
Table 1. Free amino acids identified and percent of total present in each sample of drone honeybee semen for drone age and season shown for the years 2003-04.
Amino acid Aspartic Glutamic Ac. Serine Asparagine Glycine Glutamine Histidine Threonine Arginine Alanine Proline Tyrosine Valine Cysteine Methionine Isoleucine Leucine Lysine Phenylalanin e Tryptophan 14/spring 4.4 4.9 2.5 0.0 1.2 3.4 2.8 2.1 47.2 1.7 22.3 0.9 1.9 0.0 1.2 1.9 2.5 2.5 1.3 0.3 % of Total Amino Acids Drone age (days) / season 14/summer 14/autumn 21/spring 21/summer 1.0 2.5 2.7 1.7 3.1 0.7 0.8 0.3 2.7 2.0 1.9 1.5 0.0 0.0 0.0 0.0 1.5 1.2 1.2 1.0 6.6 4.7 4.1 3.4 2.7 2.8 2.6 2.4 2.6 1.7 1.7 1.3 43.2 51.0 50.8 57.5 2.6 1.6 1.2 1.4 17.5 19.4 17.5 19.6 2.0 0.9 1.2 0.8 2.2 1.7 1.5 1.3 0.0 0.0 0.0 0.0 1.9 1.3 1.3 1.0 3.1 2.1 2.2 1.6 3.7 2.3 2.7 1.9 3.0 2.5 2.3 2.2 2.8 1.5 1.9 1.2 0.7 0.3 2.7 0.3

35/summer 1.1 0.2 1.3 0.0 0.9 3.0 2.0 1.1 57.3 1.3 23.5 0.5 1.1 0.0 0.7 1.4 1.7 2.3 0.8 0.2

74

Table 2. Free amino acids plus water identified and percent of total present in each sample of drone honeybee semen for drone age and season shown for the year 2006.
Amino acid 21 spring 35 spring % of Total Amino Acids Drone age (days) / season 14 14 21 21 Autum Autum Autum autumn n n n B A* B* A 10.1 10.2 10.3 10.4 11.6 11.7 11.8 11.8 5.9 5.8 5.9 5.8 5.4 5.0 5.3 5.2 2.2 2.3 2.3 2.3 5.0 5.0 5.0 5.0 9.3 9.1 9.5 9.2 4.7 4.7 4.8 4.8 6.6 6.5 6.5 6.4 3.7 3.9 3.5 3.8 5.3 5.3 5.4 5.4 2.9 2.9 2.6 2.9 6.3 6.3 6.3 6.3 8.1 7.9 8.0 7.9 8.2 8.9 8.5 8.3 4.6 4.4 4.4 4.5 -

35 autumn A 10.5 12.1 6.1 5.4 1.9 5.1 8.5 4.9 6.2 3.6 5.5 2.1 6.4 8.1 8.7 4.9 -

35 autumn B 9.7 11.4 6.3 7.2 2.1 4.9 9.1 4.6 6.2 4.0 5.2 2.9 6.2 8.0 7.6 4.6 -

Aspartic Glutamic Ac. Serine Asparagine Glycine Glutamine Histidine Threonine Arginine Alanine Proline Tyrosine Valine Cysteine Methionine Isoleucine Leucine Lysine Phenylalanin e Tryptophan *

10.2 11.8 5.7 5.0 2.3 4.9 9.3 4.7 6.7 4.0 5.2 2.5 6.2 8.1 8.9 4.5 -

10.3 12.0 6.0 5.4 2.0 5.0 8.9 4.7 6.2 4.3 5.3 2.8 6.4 8.3 7.8 4.7 -

Semen samples from drones from A = queen breeder A, and B = queen breeder B

75

Table 3. Fatty acids identified and percent of total fatty acids present in drone honeybee semen for drone age and season shown for the years 2003-04. Fatty acid 14/sprin g 0.4 0.7 1.3 0.3 1.6 2.8 0.6 2.9 0.4 0.6 4.6 0.4 6.1 2.9 65.0 4.5 0.3 4.3 0.0 0.0 0.0 0.0 0.2 21/sprin g 0.0 0.0 0.3 0.0 0.3 0.8 0.4 1.3 1.8 1.6 1.6 0.5 2.7 1.1 73.0 3.3 2.9 6.0 0.0 0.4 0.0 0.0 0.2 % of Total Fatty Acids Drone age (days)/season 14/sum 21/sum 35/sum mer mer mer 0.0 0.8 0.4 0.6 0.4 0.3 0.0 1.3 0.5 0.0 0.3 0.0 0.9 1.8 0.7 1.2 3.1 1.2 0.0 0.9 0.3 1.7 2.8 1.9 0.0 0.0 1.3 0.7 0.7 1.3 2.0 5.2 2.4 0.0 0.4 0.7 1.2 4.9 4.5 3.3 4.2 1.8 80.1 63.1 60.3 3.0 2.8 2.7 0.7 0.8 5.1 4.4 4.9 4.5 1.7 4.1 2.9 0.0 0.0 0.6 2.2 1.9 0.0 0.0 3.6 0.0 0.0 0.0 0.7 -

C6 : 0 C8 : 0 C10 : 0 C11 : 0 C12 : 0 C13 : 0 C14 : 0 C14 : 1 C15 : 0 C15 : 1 A C16 : 0 B C16 : 1 C C17 : 0 C17 : 1 D C18 : 0 C18 : 1n9t C18 : 1n9c C18 : 2 C18 : 2n6t C18 : 2n6c E F C18 : 3 C18 :3n6 C18 : 3n3 C20 : 0 C20 : 1 C20 : 2 C20 : 4n6 G H C21 : 0 C22 : 0 C22 : 1 J C24 : 0 C24 : 1

14/autu mn 0.0 0.5 0.9 0.0 1.2 2.3 0.7 2.4 0.0 0.5 3.5 0.4 4.8 2.5 68.3 6.1 0.5 4.6 0.0 0.0 0.0 0.0 0.0 -

21/autu mn 0.4 0.9 1.2 0.0 1.8 3.5 0.7 3.3 1.3 1.9 5.2 0.5 7.0 2.1 48.5 4.6 8.5 4.0 0.0 0.5 0.0 0.0 0.7 -

76

Table 4. Fatty acids identified and percent of total fatty acids present in drone honeybee semen for drone age and season shown for the year 2006.
Fatty acid 21 spring C6 : 0 C8 : 0 C10 ; 0 C11 : 0 C12 : 0 C13 : 0 C14 : 0 C14 : 1 C15 : 0 C15 : 1 A C16 : 0 B C16 : 1 C C17 : 0 C17 : 1 D C18 : 0 C18 : 1n9t C18 : 1n9c C18 : 2 C18 : 2n6t C18 : 2n6c E F C18 : 3 C18 : 3n6 C18 : 3n3 C20 : 0 C20 : 1 C20 : 2 C20 : 4n6 G H C21 : 0 C22 : 0 C22 : 1 J C24 : 0 C24 : 1 * 3.1 2.6 1.8 1.0 0.3 0.1 0.3 1.9 2.4 1.7 1.6 0.3 0.1 0.1 0.1 1.8 58.3 14.5 1.4 0.9 0.1 2.7 0.7 1.7 0.7 0.1 0.1 0.1 0.1 0.1 0.3 35 spring 2.3 1.5 1.3 0.8 0.3 0.1 0.2 1.1 2.4 2.0 1.9 0.5 0.3 0.1 0.1 1.8 54.7 20.9 1.0 1.0 0.1 2.3 1.1 1.4 0.9 0.1 0.1 0.1 0.1 0.4 0.1 % of Total Fatty Acids Drone age (days) and season 14 14 21 21 autumn autumn autumn autumn A* B* A B 1.2 1.6 6.3 3.9 1.4 0.9 5.6 2.7 0.6 0.8 3.7 1.7 1.3 0.7 4.4 1.8 0.1 0.1 0.1 0.3 0.3 0.1 0.1 0.1 0.1 0.4 0.5 0.3 5.9 1.0 14.2 5.5 3.1 1.4 1.3 1.7 0.5 0.8 2.4 1.2 2.3 1.3 1.8 1.4 1.0 1.0 2.4 0.4 0.5 0.5 2.9 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 2.7 1.5 5.3 1.6 36.7 48.8 17.2 40.8 32.4 30.2 15.1 29.4 2.3 0.9 5.4 0.9 2.2 1.8 4.7 1.2 0.1 0.1 0.8 0.1 1.1 1.6 0.8 1.3 1.3 1.2 1.2 1.1 0.2 1.5 0.7 1.1 0.3 0.9 1.1 0.9 0.3 0.3 0.5 0.1 0.5 0.2 0.1 0.1 0.4 0.1 0.1 0.1 0.5 0.4 1.3 0.6 0.2 0.1 0.4 0.1 0.1 0.2 0.1 0.1

35 autumn A 5.1 6.5 5.1 2.9 0.5 0.1 0.6 3.0 5.6 2.0 2.9 0.6 0.3 0.1 0.1 1.4 32.1 13.5 1.9 1.4 0.1 2.5 2.1 5.0 4.5 0.2 0.1 0.2 0.1 0.1 0.1

35 autumn B 6.7 4.6 3.4 3.2 0.2 0.3 0.5 33.0 1.0 1.6 2.5 1.1 2.4 0.2 0.2 3.9 8.1 10.0 5.5 6.3 0.1 0.8 0.6 0.2 0.7 0.2 1.1 0.1 1.3 0.1 0.4

Semen samples from drones from A = queen breeder A, and B = queen breeder B

77

Figure 1. Free amino acids, percent of total in each sample, 2003-4.

70

60

50

40 %

30

14Sp 14Su 14Au 21Sp 21Su 35Su

20

10

0 Aspartic Asparagine Histidine Alanine Amino Acids Valine Isoleucine Phenylalanine

Figure 2. Free amino Acids 2006. Percent of total present in each sample.

14

12

10

21/Sp 35/Sp 14/Au L 14/Au H 21/Au L 21/Au H 35/Au L 35/Au H

% 6 4 2 0

Amino Acids

78

Va lin e C ys te in e M et hi on in e Is ol eu ci ne Le uc in e Ly si Ph ne en yl al an in Tr e yp to ph an

As pa rti c G lu ta m ic Se rin As e pa ra gi ne G ly ci ne G lu ta m in e H is tid in e Th re on in e Ar gi ni ne

ni ne Pr ol in e Ty ro si ne

Al a

Figure 3. Fatty acids. Percent of each fatty acid present for each sample for all ages and seasons for 2003-4 and 2006.
90 14Sp1 80 70 60 50 % 40 30 20 10 0 14Sum1 14 Aut1 14Aut2 14Aut3 21Sp1 21Sp2 21Sum1 21Aut1 21Aut2 21Aut3 35 Sp1 35Sum1 35Aut1 35Aut2

C D 1 C 8: 18 0 C :1n 18 9 :1 t n C 9c C 18 18 : 2 C :2 18 n6 :2 t n6 c

C F C 18 18 :3 C : 3n 18 6 :3 C n3 20 C :0 20 C :1 2 C 0: 20 2 :4 n6

C C 17 C :0 17 :1

C 6 : C 0 8 C :0 10 C :0 11 C :0 12 C :0 13 C :0 14 C :0 14 C :1 15 C :0 15 :1

C H 21 C :0 22 C :0 22 :1

C A 16 :0

C B 16 :1

Fatty Acids

79

C J 24 C :0 24 :1

11.7 References
Baer, B., Maile, R., Schmid-Hempel, P., Morgan, E.D, and Jones, G.R. 2000 Chemistry of a mating plug in bumblebees, Journal of Chemical Ecology, 26 (8), 1869-1875. Baer, B., Morgan, E.D. and Schmid-Hempel, P. 2001 A non-specific fatty acid within the bumblebee mating plug prevents females from remating, Proceedings of the National Academy of Sciences of the United States of America, 98 (7), 3926-3928. Blum, M.S., Glowska, Z. and Taber, S. III 1962 Chemistry of the drone honeybee reproductive system. II. Carbohydrates in the reproductive organs and semen, Annals of the Entomological Society of America, 55, 135-139. Blum, M.S., Bumgarner, J.E. and Taber, S. III 1967 Composition and possible significance of fatty acids in the lipid classes in honeybee semen, Journal of Insect Physiology, 13, 1301-1308. Crailsheim, K. and Leonard, B. 1997 Amino acids in honeybee worker haemolymph, Amino Acids, 13, 141-153. de Groot, A.P. 1953 Protein and amino acid requirements of the honeybee (Apis mellifera L.), Physiology Comp. Oecology, 3, 197-285. Gillott, C. 1996 Male insect accessory glands:functions and control of secretory activity, Invertebrate Reproduction and Development, 30:1-3, 199-205. Gillott, C. 2002 Male accessory gland secretions:modulators of female reproductive physiology and behaviour, Annual Review of Entomology, 48, 163-184. Leonard, B. and Crailsheim, K. 1999 Amino acids and osmolarity in honeybee drone haemolymph, Amino Acids, 17, 195-205. Mailer, R.J., Ayton, J. and Conlan, D. 2002 Comparison and evaluation of the quality of thirty eight commercial Australian and New Zealand olive oils, Advances in Horticultural Sciences, 16 (3-4), 256259. Manning, R. 2001a Fatty acids in pollen:a review of their importance for honeybees, Bee World, 82 (2), 60-75. Manning, R. 2001b Pollen analysis of Eucalypts in Western Australia, RIRDC Publication No., 01/53, Rural Industries Research and Development Corporation, Canberra, ACT, Australia. Novak, A.F., Blum, M.S., Taber, S. III, and Liuzzo, J.A. 1960 Separation and determination of seminal plasma and sperm amino acids of the honeybee, Apis mellifera, Annals of the Entomological Society of America, 53, 841-843. Rhodes, J.W. 2002 Introduction and early performance of queen bees some factors affecting success, RIRDC Report, Rural Industries Research and Development Corporation, Canberra, ACT, Australia. Rice, W.R. 1996 Sexually antagonistic male adaptation triggered by experimental arrest of female evolution, Nature, 381, 232-234. (Abstract). Somerville, D. 2004 The floral resources of New South Wales of primary importance to beekeeping, PhD Thesis, Australian National University, Canberra, Australia. Tram, U. and Wolfner, M.F. 1998 Seminal fluid regulation of female sexual attractiveness in Drosophila melanogaster, Proceedings of the National Academy of Sciences of the United States of America, 95, 4051-4054. Woyke, J. and Jasinski, Z. 1978 Influence of age of drones on the results of instrumental insemination of honeybee queens, Apidologie, 9 (3), 203-212.

80

Semen Production in Drone Honeybees

RIRDC Pub. No. 08/130

This report looks at the importance of drone semen quality studies and its effect on the mating success of queen bees. Sufficient numbers of mature age drones, each producing a large volume of semen containing a high number of sperm are required to be present at commercial queen bee mating apiaries to result in queen bees with the maximum number of sperm present in their spermathecae after mating. Low numbers of sperm in the spermathecae of queen bees after mating contributes to early supersedure resulting in increased costs to the beekeeper from queen replacement and reduced colony production as colonies weaken in strength or become queenless during the period of queen replacement. This project provides information to persons interested in bee breeding with particular reference to queen bee production by

providing data relating to numbers of sperm produced by drones and highlighting the requirement of use of this data in the selection of queen bee drone mothers by including drone sperm production in the selection criteria and by identifying the most suitable method for collecting drone sperm data. The Rural Industries Research and Development Corporation (RIRDC) manages and funds priority research and translates results into practical outcomes for industry. Our business is about new products and services and better ways of producing them. Most of the information we produce can be downloaded for free from our website: www.rirdc.gov.au. RIRDC books can be purchased by phoning 02 6271 4160 or online at: www.rirdc.gov.au/eshop.

Contact RIRDC: Level 2 15 National Circuit Barton ACT 2600

This publication can be viewed and purchased from our webstie:

PO Box 4776 Kingston ACT 2604 Ph: 02 6271 4100 Fax: 02 6271 4199 Email: rirdc@rirdc.gov.au web: www.rirdc.gov.au

www.rirdc.gov.au