a". Geuis, regione paroticà et collo postico nigMs.
o'". Rectricibus exterioribus nigris, cæruleo max-ginatis . . . 7. hydrochans,
b'". Rectricibus exterioribus albis, cæruleo marginatis . . . . 8. acis
b". Genis, regione paroticà e t collo-postico satiu-atè cæruleis.
o'". Pileo ultramarino : superciliis cum nucha cyaneis : tectricibus
caudalih.,a postremis n i g i i s ................................................... 9. nmrgaretim.
b! . Pileo concolori : tectricibus caudalibus omnino albis.
o"". Dorso cyaneo m a c u l a t o ........................................................ iq nais
Ô"". Dorso c o n c o lo r i..............................................................................11. galatea.
c". Genis, regione paroticà e t collo postico viridi-cyaneis . . . . 12. riedeli.
I t is to be reg re tted th a t th e Alcedo dea o f Linnæus, th e type o f this genus, is not
recognizable ; and I have therefore le ft th e name o u t o f th e question as tending only to
confuse. The locality given by Seba and the older authors as th e h ab itat, Ternate, is now
known n o t to contain a species o f Tanysiptera. Professor Schlegel and I differ in our
estimate o f th e value o f the species recognized by me, as h e is inclined to consider them
variations o f one type, while I reg ard them as good species, seeing th a t the characters of
each are permanent, and the geographical range o f each peculiar to the species itself.
Genus 18. ClTTUE.4.
Cittura, Kaup, Fam. Eisv. p. 8 ( 1 8 4 8 ) .............................
Type.
C. cyanotis.
a. Plumis superciliaribus pallide lila c iu o - r u b e n t ib u s ........................................i . cyanotis
b. Plumis superciliaribus albo t e rm i n a t i s ...............................................................3 . sanghirensis.
T he genus Cittura has a remarkable bill, which is grooved along the culmen, as reprepresented
in th e P la te (fig. 16). W h e th e r the b eautiful C. sanghirensis is really a distinct
species, or wh e th e r i t is only th e breeding-plumage o f the ordinary C. cyanotis, remains to
be seen; b u t th a t it is n o t confined to Sangliir is certain, from th e fact o f my having
received i t from Celebes on two difierent occasions. Mr. Renesse van Duivenbode also sent
me a note to say th a t C. cyanotis varied much with th e season o f the year. Against the
p robability o f the two species being identical must be mentioned the larger bill, black
forehead, and white-tipped eyebrow, all o f them conspicuous characters in C. sanghirensis.
Genus 19. M e l id o r a .
Melidora, Less. Traité d’Orn. p. 249 ( 1 8 3 1 ) .......................
Typo.
M. macrorhyncha.
M. maci'orhyncha (Less. A’’oy. Coq. i. p. 692).
In th is genus we have th e extreme type o f the Dacelmiinw; and wo have seen th e bill, as
th e birds became more and more insectivorous, gradually go thro u g h various modifications,
becoming more and more depressed, u n til in the present genus the groove, which was appa-
GE09EAPHICAL DISTEIBUTION. xxi
ren t in the culmen o f Cittura, is here more strongly developed, while th e apex o f th e bill
is furnished with a hook, which doubtless subserves some useful purpose in procuring food,
with which we are a t present unacquainted.
GEOG RAPH ICAL DISTR IBUTION.
Genus 1. A lcedo.
Pange o f the Genus.
Over the greater p a rt o f the Old World, b u t does n o t extend into th e Australasian
Region proper—the furthest authentic range o f an Alcedo being, so far as we know, into th e
Austro-Malayan subregion to New Guinea.
Range o f the Species.
1. Alcedo ispida. Over the whole of the AA'^estem Paleearctic Region, b u t o f rare r occurrence
in the north. Foimd in North ern Africa, and extends into E gypt and Palestine. Its
range to the east is undetermined; b u t it is probably replaced th ro u g h o u t the Eastern
Palffiarctic Region from Persia and Central Asia, by th e n ex t species.
2. Alcedo bengalensis. The representative o f A . ispida, which i t replaces in the east. I t
has been found to th e westward as far as North-eastern Africa, having been m et with in
Egypt and the Sinaitic peninsula. Its eastern range extends over th e whole o f the Indian
Region and into the Eastern PaliEarctic as far as Amoorland and J a p a n ; b u t its ex ten t in
this quarter is n o t yet satisfactorily determined. I t is also found disti’ibuted over the whole
of th e Indo-Malayan and Indo-Chinese subregions, and extends into th e Anstro-AIalayan as
far as Gilolo, where, however, it can only be a rare and occasional visitant.
3. Alcedo grandis. The largest species o f Alcedo yet discovered. As yet it has been only
obtained in th e Terai, below Darjiling.
4. Alcedo moluccensis. Although belonging to th e same section o f th e genus Alcedo as
th e foregoing, its nearest ally is Alcedo bengalensis, which it replaces in the Moluccas, where
it is generally distributed, though, as far as we yet know, i t is confined to the Austro-
Malayan subregion *.
5. Alcedo quadribrachys. Confined to the Ethiopian Region, principally to th e west
coast, b u t lias been known to occur in Natal once. No t yet met with in North-eastern
Africa.
6. Alcedo asiatica. Generally distributed over th e Indo-Malayan subregion, e xtending into
the eastern p a rt o f In d ia proper, and also into Cochin China. Found likewise in Celebes.
7. Alcedo semitorquata. Confined to th e Ethiopian Region, over the whole o f which it
is distributed, b u t is more common in South Africa.
• The locality, “ Flores” (TFaZZacc), given in my account of this species should be expunged from the list of habitats,
as I afterwards discovered that the bird brought by Mr. Wallace from this iskud was not the true A. moluccensis, but
tbe bright race of A. bengalensis, named by lloichcnbach A. sondaica.