Meta-analysis reveals that pollinator functional diversity and abundance enhance crop pollination and yield.

How insects promote crop pollination remains poorly understood in terms of the contribution of functional trait differences between species. We used meta-analyses to test for correlations between community abundance, species richness and functional trait metrics with oilseed rape yield, a globally important crop. While overall abundance is consistently important in predicting yield, functional divergence between species traits also showed a positive correlation. This result supports the complementarity hypothesis that pollination function is maintained by non-overlapping trait distributions. In artificially constructed communities (mesocosms), species richness is positively correlated with yield, although this effect is not seen under field conditions. As traits of the dominant species do not predict yield above that attributed to the effect of abundance alone, we find no evidence in support of the mass ratio hypothesis. Management practices increasing not just pollinator abundance, but also functional divergence, could benefit oilseed rape agriculture.

Pulsatile secretion of luteinizing hormone during the menstrual cycle of rhesus macaques.

The secretion of LH, PRL, and cortisol was investigated in 4 sexually mature female rhesus macaques with cardiac catheters protected by tethers. Based on endocrine parameters, all 4 of the animals ovulated within 2 months from the time they were tethered, and regular menstrual cycles of 24-34 days were observed. The catheters remained patent for 6-12 months without reposition or repair. Plasma levels of 2 stress-labile hormones, PRL and cortisol, showed diurnal fluctuations comparable to those observed in untethered animals. The frequency of LH secretory episodes was determined by measuring bioactive LH in blood samples collected at 10-min intervals in the follicular phase and at 15-min intervals in the luteal phase of the menstrual cycle. In 10 trials during the follicular phase, we estimated that an average of between 14 and 15 LH pulses occurred every 12 h. The interpulse interval ranged between 20-80 min and averaged 50 min. No change in pulse frequency was observed across the follicular phase. The number of LH pulses decreased after ovulation, and by the end of the luteal phase, the interpulse interval was 4-6 h. One example during the preovulatory LH surge revealed the high frequency, high amplitude nature of LH secretion at that time. Our experience indicates that tethered animals with cardiac catheters show no hormonal indications of stress and represent the best available model for studies requiring frequent and prolonged access to the vascular system. Our data suggest that peripheral LH fluctuations in rhesus monkeys, as in other mammals, are pulsatile, and the frequency of these pulsatile episodes changes with different phases of the menstrual cycle, presumedly in response to varying stimuli to the pituitary from the brain.