We have made an effort to independently assess all described subspecies that appear in modern ornithological literature within the range covered by the book. This has forced us to spend much time in museum collections, the only place where series of closely related but differently named subspecies can be compared side by side.Between the two of us we have spent many, many months in museums in order to write this handbook. One important object of this laborious work has been to decide whether a subspecies is sufficiently distinct to be recognised and treated as separate from other subspecies of the same species. We have, by and large, tried to apply the so-called 75% rule (e.g. Amadon 1949), meaning that at least three-quarters of a sample of individuals selected at random (in some cases of one sex or age) ofa subspecies must differ diagnosably from other described subspecies within the examined species. The specimens selected for assessment should as far as possible be from similar age categories and seasons to allow for the most comparable plumages to be fairly assessed. The main limiting factor for this part of our work hasbeen the scarcity of comparable material of some taxa in collections.
We have long felt the need to focus on distinct subspecies and to dismiss the many subtle or even questionable subspecies that were perhaps based on too short series or on skewed or inadequate material, or which in reality represented only minor tendencies within a population, far from the minimum 75%. Natural variation within any one population has often been underestimated when new subspecies have been described. In our opinion, it is better to concentrate on clear differences, and to account for slight clinal tendencies or deviations from the most typical under a single subspecies heading, rather than to create a confusing mosaic of subtle or dubious but formally named subspecies. Most of the final assessment of subspecies taxonomy was done by LS.
Compared to other handbooks and checklists, we accept about 15% fewer subspecies. Those deemed to be too similar to a neighbouring race to be upheld have been lumped with that race and treated as a synonym.
Much geographical variation within continentally distributed species is smooth and gradual, this variation usually being termed clinal. There are taxonomists who tend to see this as a problem and who prefer to treat the entire such range as one subspecies. It is true that clines create circumscription difficulties. However, we prefer to judge every case by its own merits; if the ends of a cline are very different,a minimum of two named subspecies seems reasonable, one at each end and with a diffuse centre of intergrading characters. If there are hints of stepwise change along the cline, these suggest subspecific borders and lead to recognising additional subspecies along the cline.
We want to stress that the definition of a subspecies is based on morphology.If size, structure, colours or patterns differ with sufficient consistency (in this work,as stated, by at least 75% of the examined specimens of at least one sex or age)within a geographically defined part of the range of a species, then it is a valid and distinct subspecies irrespective of whether an examination of its mitochondrial DNA or another genetic marker reveals no clear genetic difference from neighbouring subspecies. Geneticists are sometimes so impressed by their new tool that they forget these prerequisites and distinctions. If we are to re-define subspecies by their genetic properties, all existing subspecies need to be discarded and all work on geographical variation redone in laboratories. But in such a case, scientists need first to agree on exactly which genetic definition all variation is to be judged against. Since this is unlikely ever to happen, we had better stick to the morphology-based subspecies taxonomy we already have!
It is best to add that our approach to subspecies taxonomy is quantitative rather than qualitative. If a described taxon does not in our view reach the required level of distinctness, we place it in synonymy under a senior name. But that does not mean that we claim to know better than others regarding the recognition of a valid taxon,only that it is not in our view distinct enough according to our requirements. If others prefer to describe every tiny variation under a formal name it is up to them.
Subspecies deemed by us to be borderline cases as to whether they are warranted or not, and subspecies only differing very subtly from other valid subspecies, have been marked with an open circle in front of their names. For subspecies that are either for some reason questionable, or for which we failed to find sufficient or indeed any material, a question mark has been placed in front of their names. Subspecies known or suspected to be now extinct carry a dagger mark in front of the name.
At the end of each subspecies entry a selection of synonyms has been listed alphabetically within brackets; these are subspecies names that appear in the literature or on some specimen labels in museums but which are either junior synonyms or are invalid for one reason or another. They are shown as a service to the reader and to make it clear that they have not simply been overlooked by us. We have as a rule only included the more recent and commonly seen synonyms, but a few older or more rarely used ones have also been included if considered helpful to the reader. Anyone seeking a more complete list is referred to Sharpe (1874–99), Hartert(1903–38), Peters et al. (1931–87) or Vaurie (1959, 1965).
