The sources of data and method 
These criteria were applied to the 345 species on the Yemen list, and Table 
1 was prepared using population and range data for Yemen and other 
areas of the world from a combination of: Brooks et al (1987), Cramp et 
al (1977 - 1988), Hall & Moreau (1970), Hollom et al (1988) and Snow 
(1978). It must, however, be clearly stated that because so few population 
data exist and because of the provisional nature of so many breed ing and 
wintering ranges, Table 1 must be regarded as provisional. Nevertheless, 
I believe it to be a fair assessment of current bird conservation priorities 
in the Yemen even though very little is known about the southern part of 
the Republic. 
Discussion 
It follows that the species that occur in Yemen that do not appear on Table 
1 are notof such high priority in conservation terms as the 101 species that 
do. Examination of the various 'conservation' categories into which these 
101 species fall in given in Table 2. Of these, the greatest priority must be 
given to the endemics, the species which breed and winter in 
internationally important numbers and those which breed and winter in 
regionally important numbers, especially if their distribution is localised . 
Because the avifauna of Yemen has been so poorly studied, especially 
that of southern Yemen, and because there have been no repeat censuses 
(even sample censuses) of any of its bird populations, it has not been 
possible to identify any species which is declining in numbers or whose 
range is contracting. 
Nearly 40 (c. 70%) of the 56 regionally important breeders are African 
species whose most northerly population occurs in Southern Arabia. 
Similarly, over 60% of the rare breeders are species that are common in 
Africa. 
