The UV reflectance of the crown plumage of female blue tits significantly affected male investment in feeding nestlings. Males made less frequent feeding trips when paired with UV-reduced females, as predicted by Burley . The original DAH predicts that individuals face a trade-off between current and future reproduction and therefore adjust their parental effort in accordance with their mate’s aesthetic traits, whenever these traits are linked to the mate’s quality . This is based to the assumption that heritable quality (“good genes”) will increase offspring value and eventually result in more grand offspring and, consequently, higher fitness. Consequently males should invest more in offspring provisioning when paired with ornamented females if male provide parental care and female quality strongly affects offspring survival.
UV coloration is thought to be an indicator of individual quality and to be a sexually selected trait (but see ). The UV reflectance of structural feathers is determined by a combination of pigments and nanometre-scale structures [39, 40]. Recent studies suggested that fast moulting and nutritional stress can affect the colours of structural feathers . However, a recent experimental study found no evidence for a relationship between nutritional and body condition and UV coloration in blue tits . Recent studies have revealed that the UV reflectance of structural feathers is sensitive to wear and might easily be affected by the accumulation of dirt and parasites [42–45]. Individuals need to invest time in feather maintenance to keep their feathers in good condition. Plumage maintenance is a costly and time-consuming process, which forces individuals into a trade-off between plumage maintenance and time they need for other activities (e.g. foraging) [46, 47]. Hence, it could be that only individuals in good condition can afford these costs and therefore high UV reflectance .
Our results and recent studies suggest an association between female blue tit UV coloration and condition as well as female breeding performance [14, 34–37]. Therefore, we consider UV reflectance to be a potential indicator trait that might be used by males to assess female quality. Even though a decrease in UV reflectance during the breeding season is a normal process  the sudden reduction caused by our treatment might indicate a drop in the condition and parental quality of the current mate. Given that food provisioning for the offspring is costly for the feeding parent , our data suggest that males differentially allocate feeding investment to enhance their chances of future reproduction. Females in bad condition might not be able to provide sufficient parental care, which in turn affects nestling body mass and growth. A study on ring-necked pheasants (Phasianus colchicus) revealed that selected adult sexual ornamentation can reflect nutritional condition during early phases of growth . Furthermore, recent studies on blue tits indicate that the UV coloration of nestlings is affected by condition [51, 52]. Juvenile males in better condition appear to develop more colourful tail feathers  that are not moulted during the post-juvenile moult  and therefore might enhance attractiveness in the first breeding year.
According to the basic idea that individuals choose mates on the basis of ornaments that reflect quality, males paired with females in poor condition might face the problem of producing offspring with low reproductive success. In other words, because of reduced female quality the reproductive value of the offspring will be reduced, also from the male’s perspective. The difference in male feeding investment in response to the female treatment is therefore consistent with the prediction that DA is strongly related to reproductive life-history. Our results on paternal care investment complement other studies, which demonstrated the female’s differential allocation of parental care in blue tits in response to the manipulation of male UV coloration of the crown [28, 29]. Both studies indicate that male UV coloration is under selection pressure, created by the female allocation of parental care. Based on our results and other studies [32, 33], which likewise showed assortative mating in response to UV reflectance, we suggest that male differential allocation may affect selection on female UV coloration.
Whereas the original concept of the DAH also suggests that less attractive partners should increase their parental investment [18, 19], we found no effects of the treatment on female parental care. Johnsen et al.  demonstrated that males invested more into parental care when their UV reflectance was reduced. One possible explanation for the lack of compensation by females could be that they were already making the maximum possible investment . The weather during spring 2009 and especially 2010 was cold and rainy. Food resources for the birds were probably limited and female food provisioning was restricted by the availability of resources. Although males provided significantly less food to nestlings from UV-reduced females than to females from the control group, and females did not compensate for this lack of feeding effort, we found no effect of treatment on body mass in either group. One reason could be that the experiment was conducted only a few days before fledging, a time during which passerine nestlings often lose weight . Furthermore, daily data on nestling development were not collected after female treatment, although this could have been informative.
The experimental approach of Burley  revealed a correlation between female attractiveness and male feeding investment in zebra finches, whereas no relationship was found between the attractiveness of the female and the nest defence behaviour of their mates. A recent experiment on rock sparrows showed a reduction in male nest defence intensity in response to a reduction in female attractiveness, but not in feeding investment [24, 25]. On the contrary, here we found no differences in the nest defence behaviour of male blue tits facing either control or UV-reduced females. Our results may reflect the lack of treatment effect during the nest defence experiment. The reduction in UV reflectance was only small (13.36%), with the aim of avoiding a confounding “strange-mate” effect. The treatment is known to diminish UV reflectance for at least 5 days in wild birds . Whereas the feeding investment observation was conducted the day after manipulation of the female crown plumage, nest defence behaviour was observed 3–4 days after treatment. During this period the UV-reducing chemical could have been removed by plumage maintenance. Further research is required to test whether male differential allocation only takes place during certain stages of parental investment and to test in which stages of the breeding cycle male allocation occurs.
In conclusion, our results indicate that male blue tits make parental care decisions in accordance with their mate’s quality. To our knowledge this is one of very few experimental studies demonstrating male differential allocation in relation to female attractiveness.