Birds of Celebes; Coraciidae. 
311 
Salvador! (23) considers it indubitable, as we do too, that no species of the 
o’enus (JovocicLS occurs in hiew Guinea and that the indication of Guoy & Gai- 
mard is, therefore, wrong. The species seems to be strictly confined to Celebes. 
In North Celebes this is a common bird. It follows the hoe of the field- 
labourer, says Platen (28), examining the freshly broken ground for booty, 
always pursued and attacked by the greedy Crow, C. enca. Meyer (22) remarks 
that it “usually flies singly; but after feeding, several play together. T'hey 
frequently sit on dead twigs and look out for grasshoppers and other insects; 
then suddenly rushing upon their prey they return to their perch. Cry tschirrr . 
In South Celebes M^allace found it a rare bird; Platen and Weber, too, 
obtained each but a single specimen. Wallace (16) observes that “it has a 
* most discordant voice, and generally goes in pairs, flying from tree to tree, 
and exhibiting while at rest that all-in-a-heap appearance and jerking motion 
of the head and tail which are so characteristic of the great Fissirostral group 
to which it belongs. From this habit alone the kingfishers, bee-eaters, rollers, 
trogons, and South American pufi-hirds, might be grouped together by a person 
who had observed them in a state of nature, but who had never had an oppor- 
tunity of examining their form and structure in detail”. 
Pater in the same work '16) Mr. AVallace remarks that “the Celebes 
Roller is an interesting example of one species of a genus being cut off from 
the rest. There are species of Coracias in Europe, Asia, and Africa, but none 
in the Malay Peninsula, Sumatra, Java, or Borneo. The present species seems 
therefore quite out of place; and what is still more curious is the fact, that it 
is not at all like any of the Asiatic species, but seems more to resemble those 
of Africa”. These views appear also to be those of no less an ornithologist 
than Dr. Sharpe who places (31) C. temmincki between C. oUvaceiceps Sharpe 
of South Africa and C. cyamgaster Cuv. of Senegambia. Air. Dresser (XXXIII) 
does not enter into the question of its affinities, but places it last in the genus 
next to C. cyamgaster. In our opinion C. temmincki has nothing to do with the 
latter species, which was jilaced with other African forms in a different genus 
by Bonaparte on account of its Swallow-tail, and it is very questionable 
whether C . oUvaceiceps can be regarded as a nearer ally than C.affinis McClell. 
of India. In a later work (24) Mr. Wallace cites C. temmincki as a Hima- 
layan form in Celebes, a de.signation which has at least an equal right to ac- 
ceptance with his earlier view of the African affinities of the species. 
In that the range of the genus is bounded to the east by the Moluccan 
Straits, Coracias conforms to the general rule found among Asiatic forms in 
Celebes. Its absence in Borneo, Java, Sumatra and Malacca is an enigma. It 
may be suggested that the genus was once represented in these countries, and 
has' become extinct there; or it may be thought possible that an ancestral form, 
a migrant like C. garrulus, accidentally straggled to Celebes, and settled in 
Africa, varying in the same direction in both places; or half a dozen other 
