Avian eggs exhibit substantial intra‐ and interspecific variation in shape, size and colour. Considerable efforts have been made to better understand the evolutionary drivers behind such variation, often using museum egg collections. Usually it is assumed that museum collections accurately represent the variation seen in natural populations, but this may not be the case if there is collection bias. Collection bias may lead to the over‐representation of certain egg traits in collections, due to the aesthetic (or other) preferences of collectors. The aim of this study is to begin to look for the occurrence of potential collection bias in museum egg collections by comparing three shape indices (pointedness/asymmetry, elongation and polar asymmetry) and egg volume between subsets of eggs in museum collections with those of recently sampled eggs in the field for three different bird species: common guillemot Uria aalge, razorbill Alca torda and northern fulmar Fulmarus glacialis. We found no evidence of collection bias in our sampled razorbill and northern fulmar museum collection eggs, but some evidence for a bias in sampled museum collection eggs of common guillemots. Since the guillemot's egg differs from most bird eggs in being pyriform, we suggest that collection bias by historic egg collectors may be more prevalent in species with extreme egg traits. Researchers using museum egg collections to examine questions relating to egg shape should be aware of collection bias risks and consider how to minimise the effect of these possible biases on accumulated datasets.

中文翻译：

---

禽蛋收藏：博物馆收藏受形状和大小影响

禽卵在形状，大小和颜色上表现出种内和种间差异。人们经常使用博物馆的鸡蛋收藏品，为更好地理解这种变异背后的进化驱动力做出了巨大的努力。通常假设博物馆的藏书准确地代表了自然种群中的变化，但是如果存在藏书偏差，情况可能并非如此。由于收藏家的审美（或其他）喜好，收藏偏差可能导致某些蛋性状在收藏中的过度表现。这项研究的目的是通过比较三个形状指标（指向性/不对称性，Uria aalge，razorbill Alca torda和北部海ca Fulmarus glacialis。我们在我们的剃刀犀鸟和北富马博物馆的收藏鸡蛋中没有发现收藏偏见的证据，但是在普通海雀科动物的博物馆收藏鸡蛋中却发现了一些收藏偏见的证据。由于海雀科的卵与大多数鸟卵的不同之处在于呈梨形，因此我们建议，具有极高卵性的物种中，有历史的卵收集者的收集偏见可能更为普遍。使用博物馆鸡蛋收藏来研究与鸡蛋形状有关的问题的研究人员应意识到收藏偏差的风险，并考虑如何最大程度地减少这些可能的偏差对累积数据集的影响。