Sections of brood from colonies of the Cape honeybee ( Apis mellifera capensis), the African honeybee ( A. m. scutellata), and hybrid bees of the two races were exchanged between colonies to study the effect of different brood-origin/nurse-bee combinations on development of caste characters. When Cape larvae were raised by African workers the amount of food provided almost doubled in comparison with Cape larvae reared by their own workers. In contrast, African larvae raised by Cape workers were provided with only half the amount they received from their own workers. After the bees emerged, we found a large degree of plasticity in characters related to caste differentiation, which corresponded closely to the amount of food provided. Super-fed Cape bees had enlarged spermathecae, were heavier than normal workers and developed more rapidly, and had reduced pollen combs, all typical for a more queen-like condition. Ovariole numbers did not appear to be enhanced by additional feeding. Cape bees that behave as social parasites in African bee colonies were most queen-like in the characters studied, albeit within the range that was found for Cape bees from normal colonies, suggesting within-colony selection for characters that enhance reproduction.

Several stratospheric chemistry modules from box, 2-D or 3-D models, have been intercompared. The intercomparison was focused on the ozone loss and associated reactive species under the conditions found in the cold, wintertime Arctic and Antarctic vortices. Comparisons of both gas phase and heterogeneous chemistry modules show excellent agreement between the models under constrained conditions for photolysis and the microphysics of polar stratospheric clouds. While the mean integral ozone loss ranges from 4–80% for different 30–50 days long air parcel trajectories, the mean scatter of model results around these values is only about ±1.5%. In a case study, where the models employed their standard photolysis and microphysical schemes, the variation around the mean percentage ozone loss increases to about ±7%. This increased scatter of model results is mainly due to the different treatment of the PSC microphysics and heterogeneous chemistry in the models, whereby the most unrealistic assumptions about PSC processes consequently lead to the least representative ozone chemistry. Furthermore, for this case study the model results for the ozone mixing ratios at different altitudes were compared with a measured ozone profile to investigate the extent to which models reproduce the stratospheric ozone losses. It was found that mainly in the height range of strong ozone depletion all models underestimate the ozone loss by about a factor of two. This finding corroborates earlier studies and implies a general deficiency in our understanding of the stratospheric ozone loss chemistry rather than a specific problem related to a particular model simulation.

The fitness of a social insect colony depends greatly on the quality (i.e., mating ability, fecundity, and offspring viability) of its queen(s). In honeybees, there is marked variation in the quality of young queens that compete in a series of lethal duels to replace a colony’s previous queen. Workers interact with queens during these duels and could increase their inclusive fitness by biasing the outcomes of the duels in favor of high-quality queens. We predicted that workers will have more antagonistic interactions (chasing, grabbing, clamping) and fewer beneficent interactions (feeding, grooming) with low-quality than high-quality queens. To test this prediction, we reared queens from 0-day-old, 2-day-old, and 3-day-old worker larvae in observation colonies undergoing queen replacement, thus producing high-quality, low-quality, and very low-quality queens, respectively. Immediately after each queen emerged, we observed her for 1 h to record her interactions with the workers. Subsequent morphological measurement of the queens confirmed that initial larval age had a significant effect on queen quality. However, there was no consistent effect of queen quality on the rates of worker–queen interactions, thus falsifying our hypothesis. The mean power of our tests was high (0.599), therefore the probability of a type II error (a false negative) is low. We conclude that if workers actively select high-quality queens, then they do so prior to queen duels, during queen development. We suggest that each worker–queen interaction has a distinct adaptive significance rather than forming a suite of behavior that favors particular queens (e.g., chasing repels any queen that approaches a queen cell, thus protecting all queen cells from destruction).

The interpretation of atmospheric measurements and the forecasting of the atmospheric composition require a hierarchy of accurate chemical transport and global circulation models. Here, the results of studies using Bremens Atmospheric Photochemical Model (BRAPHO) are presented. The focus of this study is given to the calculation of the atmospheric photolysis frequencies It is shown that the spectral high resolved simulation of the O₂ Schumann–Runge bands leads to differences in the order of 10% in the calculated O₂ photolysis frequency when compared with parameterizations used in other atmospheric models. Detailed treatment of the NO absorption leads to even larger differences (in the order of 50%) compared to standard parameterizations. Refraction leads to a significant increase in the photolysis frequencies at large solar zenith angles and, under polar spring conditions, to a significant change in the nighttime mixing ratio of some trace gases, e.g., NO₃. It appears that recent changes in some important rate constants significantly alter the simulated BrO_x- and HO_x-budgets in the mid-latitude stratosphere.

Social-insect males are often regarded as being merely short-lived "flying sperm containers", which ignores their potential influence on females and paternity patterns as found in other animals. Consequently, social-insect males have received only marginal attention and sexual selection has almost never been studied in these species. Here I present a review of the mating biology of bumblebees (Bombus spp.), which are the best-studied social insects to date. I follow a male's pathway from his birth until he successfully contributes to the next generation, and show that males have evolved adaptations and behaviors to influence paternity patterns at various stages of their life, which are similar to those exhibited by males of non-social insects. By comparing the available bumblebee data with more sparse studies of male reproductive behavior in other social Hymenoptera, I argue that such male adaptations may indeed be widespread in social insects. I suggest that current paradigms on sexual selection should be challenged by using social insects as model systems, because they offer unique features, and a solid theoretical background in which clear predictions can be made and appropriate experimental tests of them can be designed.

When colonies of swarm-founding wasps lose their nests to predation or accident, the entire adult population escapes, emigrates as an absconding swarm, and renests elsewhere. Such an event causes a reduction in the adult population due to losses during the emigration itself and to adult attrition without replacement during the subsequent preemergence growth period in the new nest. We addressed the first of these sources of mortality for 27 absconding swarms of Polybia occidentalis in Costa Rica. Adult mortality over the day that included swarm emigration averaged 0.044 ± 0.039 (SD) of the original population and was a weak positive function of distance moved, but not of swarm size. A larger data set showed that emigration distance increased with swarm size. This is the first study to measure mortality rates during emigration in a swarm-founding social insect.

The contributors to this symposium have performed phenomenographic research in different projects. Their epistemology as well as their philosophy of science did more or less differ. It was the scope of this symposium to discuss whether this can be an obstacle or may lead to fruitful debates. After brief descriptions on the methodology of the research performed, 3 questions were discussed: (a) Is it appropriate to“borrow” and use phenomenography methods in non-phenomenological research contexts? (b) Is it justifiable to substitute the term ‘experience’ used in phenomenography by ‘understanding’, or ‘conception’? (c) How can we secure validity of phenomenographic research?

Many previous assessments of the global hydrogen budget have used assumed global averages of temperatures and levels of key reactants to calculate the magnitudes of the various sinks. Dry deposition is by far the largest hydrogen sink but has not been considered in detail in previous estimates of the hydrogen budget. Simulations of hydrogen using a global three-dimensional Lagrangian chemistry-transport model and two different dry deposition schemes were compared with surface measurements. An improved dry deposition scheme which included the effects of soil moisture gave better agreement between the modelled hydrogen levels and surface measurements. The seasonal variation in the hydrogen levels was also simulated much more accurately with the new dry deposition scheme. The model results at high southern latitudes were insensitive to the relative partitioning of the sources between fossil fuel combustion and biomass burning. The results indicate a global mean hydrogen dry deposition velocity of 5.3×10⁻⁴ m s⁻¹ which is lower than the previously used 7×10⁻⁴ m s⁻¹.