Optokinetic speed control and estimation of travel distance in walking honeybees

Honeybees returning from foraging trips were video-filmed while they walked through a narrow transparent gangway to reach the hive entrance. On their way they were presented with black-and-white gratings viewed underneath as well as to both sides of the gangway. Bees could exit the gangway through one of two or three side exits installed at different distances from the gangway entrance. In one set of experiments, the substrate on which the bees walked was moved either in the bee's direction or against it. In another set of experiments, the substrate was stationary, but the pattern was moved in one or the other direction. The bee's walking speed (WS) was evaluated from the video tapes. When the substrate moved against, or the pattern in the bee's direction, in either case decreasing the speed of pattern flow (PFS), the bees increased WS, and, at the same time, they preferred the more distant exit. When the substrate moved in, or the pattern against the bee's direction, thus increasing PFS, WS decreased and the bees preferred the nearer exit. These results suggest that the speed of optic flow controls the speed of locomotion and might therefore also serve for assessing the distance travelled.Abbreviations PFS pattern-flow speed - WS walking speed Dedicated to Hans-Jochen Autrum, editor emeritus, for help in giving birth to this and many other papers over very many years, often with criticism but nevertheless with encouragement and sympathy.     

Integration time for short broad band clicks in echolocating FM-bats (Eptesicus fuscus)

Vespertilionid FM-bats (four Eptesicus fuscus and one Vespertilio murinus) were trained in an electronic phantom target simulator to detect synthetic echoes consisting of either one or two clicks. The threshold sound pressure for single clicks was around 47 dB peSPL for all five bats corresponding to a threshold energy of -95 dB re 1 Pa² * s. By varying the interclick interval, T, for double clicks it was shown that the threshold intensity was around — 3 dB relative to the threshold for single clicks at T up to 2.4 ms, indicating perfect power summation of both clicks. A threshold shift of -13.5 dB for a 1 ms train of 20 clicks (0.05 ms interclick interval) confirmed that the bats integrated the power of the stimuli. At T longer than around 2.5 ms the threshold for double clicks was the same as for single clicks. Thus, the bats performed like perfect energy detectors with an integration time of approximately 2.4 ms. This integration time is an order of magnitude shorter than that reported for bats listening passively for pure tones. In our setup the bats emitted sonar signals with durations of 2–3 ms. Hence, the results may indicate that while echolocating the bats integration time is adapted to the duration of the sonar emissions.Abbreviations AGC automatic gain control - FM frequency modulated - peSPL peak equivalent sound pressure level - rms root mean square - SD standard deviation - SE standard error of mean - T interclick interval     

Modelling the effect of treatment and behavioral change in HIV transmission dynamics

In this paper we analyze a model for the HIV-infection transmission in a male homosexual population. In the model we consider two types of infected individuals. Those that are infected but do not know their serological status and/or are not under any sort of clinical /therapeutical treatment, and those who are. The two groups of infectives differ in their incubation time, contact rate with susceptible individuals, and probability of disease transmission. The aim of this article is to study the roles played by detection and changes in sexual behavior in the incidence and prevalence of HIV. The analytical results show that there exists a unique endemic equilibrium which is globally asymptotically stable under a range of parameter values whenever a detection /treatment rate and an indirect measure of the level of infection risk are sufficiently large. However, any level of detection/ treatment rate coupled with a decrease of the transmission probability lowers the incidence rate and prevalence level in the population. In general, only significant reductions in the transmission probability (achieved through, for example, the adoption of safe sexual practices) can contain effectively the spread of the disease.     

Use of recombinant interferon-α in Human Immunodeficiency Virus (HIV)-infected individuals

Rationale and objective Interferon alpha (IFN-) has anti-retroviral activity and is a possible HIV infection-limiting factor. The aim of this work is to prevent or delay disease progression in asymptomatic Human Immunodeficiency Virus (HIV) carriers.Design and interventions Recombinant IFN alpha-2b (3×10⁶ IU 3 times weekly) was compared. to no treatment (control) in a randomized trial. Endpoints were: (i) appearance of any CDC group IV symptoms and (ii) disease progression (which excluded shifts to group IVC2 or reversible IVA, or IVB). The trial lasted from October 1987 to February 1992.Setting The trial was performed at the Santiago de las Vegas sanatorium, a specialized institution for the care of HIV-infected and AIDS patients.Population Subjects were anti-HIV-1 seropositive, Western blot-confirmed, asymptomatic (CDC group II), or with generalized lymphadenopathies (CDC group III). The groups had 79 (control) and 71 (IFN) patients.Main results Long-term IFN- treatments significantly reduced the proportion of patients who shifted to any group IV (control: 46/79; IFN: 14/71;p<0.001) or developed AIDS (control: 27/79; IFN: 12/71;p<0.05). IFN also delayed progression to AIDS (95% confidence interval for 0.5 probability of progression) from 67–83 to 116–180 months after infection. The IFN group had significantly less opportunistic infections and non-infectious complications. CD4 cell count and hemoglobin decreased in the control but not in the IFN group. Fewer IFN-treated patients developed positive serum HIV antigen detection.Conclusion IFN alpha treatment during the early stages of infection seems to be beneficial to the patients.Abbreviations CI confidence interval - AIDS Acquired Immunodeficiency syndrome - HIV Human Immunodeficiency Virus - IFN Interferon - CDC Center for Disease Control (USA) - SD standard deviation     

Morphological differentiation in Carduelis finches: Adaptive vs. constraint models

In this study we show that morphological diversification in the avian genus Carduelis (Carduelinae) has to a large extent been conservative. Using multivariate methods, we found only minor deviations from the common (ancestral) body-plan. In particular, variation in bill morphology was found to be more conservative than variation in other parts of the body. We argue that constraint models of population differentiation can successfully account for the variation in bill morphology in this genus, but are less successful in accounting for variation in other traits. This can be interpreted as a result of long-term overall stabilizing selection for a certain bill morphology which is related to the way the birds open seeds. A trait combination that is adaptive on the evolutionary time scale may thus act as a constraint on changes in bill morphology on the microevolutionary scale. We conclude that the most parsimonious explanation for low divergence in bill morphology in this genus is that all species have retained the ancestral bill morphology. This may mean that each species chooses its environment rather than being moulded by it. This argument seems to apply to bill morphology, but other traits studied in this genus appear to have evolved in a less constrained fashion. A new index of morphometric integration is introduced to describe covariance structures.     

The responses of peripheral and central mechanosensory lateral line units of weakly electric fish to moving objects

Mechanosensory lateral line afferents of weakly electric fish (Eigenmannia) responded to an object which moved parallel to the long axis of the fish with phases of increased spike activity separated by phases of below spontaneous activity. Responses increased with object speed but finally may show saturation. At increasingly greater distances the responses decayed as a power function of distance. For different object velocities the exponents (mean±SD) describing this response falloff were -0.71±0.4 (20 cm/s object velocity) and-1.9±1.25 (10 cm/s). Opposite directions of object movement may cause an inversion of the main features of the response histograms. In terms of peak spike rate or total number of spikes elicited, however, primary lateral line afferents were not directionally sensitive.Central (midbrain) lateral line units of weakly electric fish (Apteronotus) showed a jittery response if an object moved by. In midbrain mechanosensory lateral line, ampullary, and tuberous units the response to a rostral-tocaudal object movement may be different from that elicited by a caudal-to-rostral object motion. Central units of Apteronotus may receive input from two or more sensory modalities. Units may be lateral line-tuberous or lateral line-ampullary. Multimodal lateral line units were OR units, i.e., the units were reliably driven by a unimodal stimulus of either modality. The receptive fields of central units demonstrate a weak somatotopic organization of lateral line input: anterior body areas project to rostral midbrain, posterior body areas project to caudal midbrain.Abbreviation EOD electric organ discharge     

Influence of plant domestication on plant-pollinator interactions: Floral attributes and floral visitor communities in wild and cultivated squash plants

Premise

Domestication of plant species results in phenotypic modifications and changes in biotic interactions. Most studies have compared antagonistic plant-herbivore interactions of domesticated plants and their wild relatives, but little attention has been given to how domestication influences plant-pollinator interactions. Floral attributes and interactions of floral visitors were compared between sister taxa of the genus Cucurbita (Cucurbitaceae), the domesticated C. moschata, C. argyrosperma ssp. argyrosperma and its wild progenitor C. argyrosperma ssp. sororia in the place of origin.

Methods

We conducted univariate and multivariate analyses to compare floral morphological traits and analyzed floral reward (nectar and pollen) quantity and quality between flowers of wild and domesticated Cucurbita taxa. Staminate and pistillate flowers of all three taxa were video recorded, and visitation and behavior of floral visitors were registered and analyzed.

Results

Most floral morphological characteristics of flowers of domesticated taxa were larger in both staminate and pistillate flowers. Staminate and pistillate flowers presented distinct correlations between floral traits and integration indices between domesticated and wild species. Additionally, pollen quantity and protein to lipid ratio were greater in domesticated species. Cucurbit pollen specialists, Eucera spp., had the highest probability of visit for all Cucurbita taxa.

Conclusions

We provide evidence that floral traits of domesticated and wild Cucurbita species experienced different selection pressures. Domesticated Cucurbita species may have more resources invested towards floral traits, thereby increasing attractiveness to pollinators and potentially plant reproductive success. Wild ancestor plant populations should be conserved in their centers of origin to preserve plant-pollinator interactions.     

Marginal proportional hazards models for clustered interval-censored data with time-dependent covariates

The Botswana Combination Prevention Project was a cluster-randomized HIV prevention trial whose follow-up period coincided with Botswana's national adoption of a universal test and treat strategy for HIV management. Of interest is whether, and to what extent, this change in policy modified the preventative effects of the study intervention. To address such questions, we adopt a stratified proportional hazards model for clustered interval-censored data with time-dependent covariates and develop a composite expectation maximization algorithm that facilitates estimation of model parameters without placing parametric assumptions on either the baseline hazard functions or the within-cluster dependence structure. We show that the resulting estimators for the regression parameters are consistent and asymptotically normal. We also propose and provide theoretical justification for the use of the profile composite likelihood function to construct a robust sandwich estimator for the variance. We characterize the finite-sample performance and robustness of these estimators through extensive simulation studies. Finally, we conclude by applying this stratified proportional hazards model to a re-analysis of the Botswana Combination Prevention Project, with the national adoption of a universal test and treat strategy now modeled as a time-dependent covariate.