2 GEOGRAPHICAL DISTRIBUTION 



certain day is not enough. The characters relied upon for the determi- 

 nation of tlie species should^ in all cases^ be stated. The condition of the 

 plumage and the appearance of the feet should be noted, and any pecu- 

 liarity of habit which might possibly throw any light upon the chances of 

 the bird having escaped from confinement ought to be recorded. Every 

 scrap of evidence to prove that the example was actually procured in this 

 country in an apparently wild state, and was examined in the flesh by 

 absolutely trustworthy witnesses, is of the greatest interest and scientific 

 value. 



In consequence of the absence, in so many cases, of this all-important 

 evidence too many records have to be accepted or rejected on the most 

 unsatisfactory ground — the supposed probability or improbability of the 

 occurrence of the species recorded. 



It seems to be generally admitted that birds which are only seen and 

 not procured, however competent the observer, must be entirely excluded. 

 The maxim of the " Old Bushman " that ivhat is hit is history, but what 

 is missed is mystenj, must be rigidly enforced. Of birds that have only 

 once been recorded as British the balance of probabilities must be weighed 

 as best it may. The fact that the Siberian Thrush [Geocichla sihirica) 

 has occurred at ten or more places in Europe increases the probability 

 that the English record is a correct one. The circumstance that several 

 species of Gallinules {Porphyrio) are kept in aviaries and on ornamental 

 waters very much weakens the evidence in favour of any of them having 

 visited the country in a wild state. Many examples of the Virginian 

 Colin [Ortyx virginianus) and of the Andalucian Hemipode {Turnix 

 sylvatica) have been turned out in the attempt to naturalize them in this 

 country, and it is only reasonable to suppose that records of the capture 

 of either of these species probably refer to imported birds or their 

 descendants. 



Some species are rejected on the ground of the inherent improbability 

 of their ever having voluntarily paid a visit to our Islands, such as the 

 Gold-vented Bulbul {Pycnonotus capensis), a species supposed to be abso- 

 lutely confined to the extreme south of Africa. There can be no doubt 

 that migratory birds are much more likely to visit us accidentally than 

 those which are non-migratory ; and it must also be admitted that most 

 accidental visitors from America or Asia which have been caught in 

 Europe are those which breed in the Arctic Regions, and have, conse- 

 quently, to deviate ^ss from their usual course in straying to our shores. 



Supposed Accidental Visitors. 

 In a score of cases I have admitted the existence of some evidence in 

 favour of a species having paid a voluntary visit to the British Islands by 

 recording it as a supposed accidental visitor. It is impossible to say how 



