Abstract
Changes in morphology have been postulated as one of the responses of animals to global warming, with increasing ambient temperatures leading to decreasing body size. However, the results of previous studies are inconsistent. Problems related to the analyses of trends in body size may be related to the short-term nature of data sets, to the selection of surrogates for body size, to the appropriate models for data analyses, and to the interpretation as morphology may change in response to ecological drivers other than climate and irrespective of size. Using generalized additive models, we analysed trends in three morphological traits of 4529 specimens of eleven bird species collected between 1889 and 2010 in southern Germany and adjacent areas. Changes and trends in morphology over time were not consistent when all species and traits were considered. Six of the eleven species displayed a significant association of tarsus length with time but the direction of the association varied. Wing length decreased in the majority of species but there were few significant trends in wing pointedness. Few of the traits were significantly associated with mean ambient temperatures. We argue that although there are significant changes in morphology over time there is no consistent trend for decreasing body size and therefore no support for the hypothesis of decreasing body size because of climate change. Non-consistent trends of change in surrogates for size within species indicate that fluctuations are influenced by factors other than temperature, and that not all surrogates may represent size appropriately. Future analyses should carefully select measures of body size and consider alternative hypotheses for change.