E t ln o p i ^ Eegions, th e species are all peculiar. No principles o f division o f territory a
proposed by Mr. Wallace and Mr. Selater a re caUed in question; and we may therefore pass
on to th e consideration o f th e characteristic forms o f the remaining re»ion
K i n . h t h ~ M l “ t r ~ ; ' ” g"«gtophical distribution of
„hsheis, I divide th e entire region in to subregions, reta ining for th e Malayan divisions
as n ea ily as possible, th e lines o f demarcation so well demonstrated by Mr. Wallace I t
must b e remembered th a t the limits o f my oivn snhregions are formed only on th e bas)s of
a u n r o T t r d P t f “ > »" m o d idunæ ot th e different localities have to be considered. iiie d lh e n th e e rtire
a. Subregio Celebensis (Celebes, Sanghir, and th e Sula Islands). Pe culiar genera- 1
C e g c o p u ^ y Monaokalogon, Z Cittura. Pe culiar species; I . PeUrgopris melanorhgnola]
g. 3- C e ym p s ix fa llm -. 4. Halcyon f o r d m i , 5. Mmiaclialcyon monachm-. 0
Tanysiptera n e d e b l -, l . a t t u r a sanghirensis-, ^ .C . cyanotis
Í. Snbregio Floresiana (Timor. Flores, lom b o ck ). Pe culiar g en e ra : 1. Caridonax
P e cu lia r species: 1. Pelargopsis floresiana-, 2. Halcyon australasice-, 3. Caridonax fulgidus
n Subregio Moluccensis (Batchian, Kaioa, Gilolo, Ternate, Morty Island, Bom-u, Ceram
Amboma, Goram, MatabeUo, KÓ). Peculiar genera, 0. Pe culiar species : 1. Alcyone afflnis ;
2. Ceyx cajeli-. 3. C. uropygialis-. 4. Halcyon lazuli-, 6. H. d io p s -, 6. I I. fu n eh r is ; 7. Tanysiptera
sobrina; 8, T. doris; 9. T. acis; 10, T. margarethæ; 11. T. nais
Snbregio Pap u an a (A m Islands, Mysol, Waigion, New Guinea, Cape-York Peninsula
Solomon Islands, Pelew Islands, Marianne Islands). Peculiar g en e ra : I. 8 y,na; 2 Melidora.
P e cu liar species: 1. Alóyeme lessoni; X Ceyx solitaria; S. Syma torotoro; i . S
f ^ v ir o s tn s ; o. Halcyon leucopygia; &. II. nigrocyanea; 1. II. cinnamomina; S. II. so rd id a -
ÿ. D ^ e lo g a u d ic lm u d i; W . P . tyro; I I . Tanysiptera s yM a ; 12. T nympha; U . T . d l iM -
14. T. hydrocharis; 15. T. galatea; 16. Melidora macrorhyncha.
e Subregio Au stra lian a (Australia generally. Van Diemen s Land, New Zealand New
Caledonia), P e cu lia r genera. 0. P ecu lia r species: 1. Alcyone diemenensis; 2. A . azurea;
i .A p n l e h r a ; A Halcyon macleayi ; i>. H. pyrrhopygia; II. m g a n s; 1. Dacelo gigas - 8
D. leachi ; 9. D. cervina ; 10. I), occidentalis.
,/. Subregio Oceamca (Polynesian Islands, eastward from th e New Hebrides group)
Pecu lia r g en e ra : I . Todh-Immphus. Peculia r species : I . Halcyon sacra; 2 .H .iu lic e i
Todirhamphus veneratus; 4. T. tu tu s ; 5. T. recurvirostris.
The following genera and species a re characteristic of th e Austra lian Eegion, b u t are not
strictly confined to one or other o f the subregions. Pe culia r genera : 1. A lc yom ; 2. Dacelo -
3, Tanysiptera. P e cu lia r species : I . Alcedo moluccensis; 2. Alcyone p u s illa ; 3. Ceux limiia-
4. Halcyon albicilla.
T h e foUow-ing species is characteristic o f the Australian Eegion, b u t yet extends its range
beyond the lim its; 1. H. sancta.
I t will be seen th a t th e geographical distribution o f Kingfishers is well illustrated by the
lig h t o f th e scheme o f geographical distribution as propounded by Dr. Selater; and the
various divisions of the Indo- and Austro-Malayan subregions, as proposed by Mr. Wallace,
are also corroborated by the distribution o f this family. Celebes is a geographical puzzle,
and is remarkable for h aving three genera confined w ithin its limits. O f these, Monachalcyon
and Cittura are certainly allied to th e Austro-Malayan forms Melidora and Tanysiptera, and
indicate the affinity of the Alcedinidoe, while, if it is tru e th a t a Tanysiptera is found there,
this will be another point o f connexion with Austro-Malayana ; but, on th e other hand,
the presence o f a Pelargopsis, albeit of a different group from the Indo-Malayan members of
th e genus, and o f Halcyon coromanda, though n o t absolutely identical with the typical
species, forms a link of connexion with the Indo-Malayan subregion. B u t th en there is
Ceycopsis, a truly Celebesian form, and a very remarkable one too ; for it is to be noticed
th a t th e two sections o f the genus Ceyx, which I have called th e rufous-backed section and
th e blue-backed section, have distinct ranges, the one being Indo-Malayan, and th e other
Austro-Malayan. Yet these two groups ai-e only sections o f th e genus Ceyx ; th e habits are
th e same, and th e form th e same, only three toes being present. The island o f Celebes is
th e point where th e two sections m ig h t be expected to jo in th eir range ; b u t tü l a year or
two ago no Ceycean form was known from this locality, and no true three-toed Ceyx has yet
been found there. Its place is tak en by the recently discovered Ceycopsis fa lla x , which is
closely allied to Ceyx, b u t has a tiny inner toe, and, more curious still, unites the characters
of th e two groups o f the genus, which converge from opposite sides upon its flanks, as one
may say ; for i t is red in general plumage, b u t has a b rig h t blue back. The study o f the
Kingfishers, therefore, does n o t throw much lig h t upon th e geographical affinities of
Celebes. The balance o f relationship is in favour o f A ustro-Malayana ; b u t still an Indian
element is clearly present, while Ceycopsis bears th e nearest relationship to Ceyx, on th e one
hand, and on th e other to the Ethiopian genus Ispidina as represented by Ispidina madagascariensis
o f Madagascar.
Tlie Polynesian genus Todirhamphus is ev-idently a modification of a certain group o f the
genus Halcyon; for it is impossible not to recognize th e close relationship o f Todirhamphus
tu tu s with Halcyon sacra, and T. recurvirostris with H. sancta. In plumage they are almost
identical ; b u t the flat bills o f the Todirhamphi indicate a more iusectivorous diet, superinduced,
no doubt, by theh- isolation, and consequent modification in the moans of p rocuring
food.
As regards the islands of Flores, Lombock, and Timor, the geographical distribution of
Kingfishers confirms the correctness o f “ W allace’s line,” as drawn by th a t gentleman foi-
th e division of th e Indo- and Austro-Malayan subregions ; for tb e affinities of th e Kin g fishers
ai-e decidedly Anstro-Malayan,' Halcycm australaàoe is closely allied to II. cinnamomina;
and the genus Caridonax, which is peculiar to these islands, is certainly a modification
o f Tanysiptera ; for take away the tails o f the two birds, and Caridmutx might be mistaken
for a stout Tanysiptera. On the other hand th e presence o f a Pelargopsis indicates Indo-
Malayan affinities ; b u t this is to be accounted for by th e narrow space which separates the
two subregions, and there can be no doubt th a t the Pelargopsis, which is a bird of strong
flight, like all piscivorous Kingfishers, could easily cross such a feeble barrier.