INDONESIAN ANNOTATIONS ON THE VOCABULARY. 113 
decision, yet it is by no means impossible that asu or ahu may repre- 
sent a development stage of the primordial onomatopoetic au. 
We find the word for a dog in these languages to have developed 
out of the sound of the dog’s bark; in other words, these peoples in 
the childhood of common culture, employ the same term as the begin- 
ning speakers in childhood of an advanced culture stage; children here 
and adults there, alike, call the dog bowwow. We have observed that 
in Polynesian culture the dog has a name derived from some other 
source; none the less the bowwow persists; even if not employed to 
name the barking animal it designates his bark; thus Samoa du, 
Maori ao, Hawaii and Tahiti aoa, all signifying to bark. 
17. bal (man-bal) pigeon. 
REFERENCES: Melanesische Wanderstrasse, 138. Deutsch-Neuguinea, 201. 
INDONESIAN. 
45. balod Visayan. 
MICRONESIAN. 
46. paluch Ponape. 48. bolochdl Palau. 
47. baluk  Satawal. 
POLYNESIAN. 
49. lupe Samoa, Niué, Futuna, Uvea, | 52. ruve Viti. 
Sikaiana, Hawaii. 53. rape Marquesas. 
50. lube Tonga, Vanikoro. 54. na-lopa Aneityum. 
51. rupe Maori, Tahiti, Rarotonga, | 55. upe Marquesas. 
Nuguria, Mota. 56. rube Gilberts. 
Attention has already been directed upon the fact that the habitat 
of this balus form is narrowly restricted to the Bismarck Archipelago 
and to a certain stretch of the northern coast of New Guinea. The 
sole occurrence of the type of which we have any record in the Indo- 
nesian area is confined to the Visayan of the southern Philippines, a 
region immediately proximate to the New Guinea and Bismarck 
Archipelago area of occurrence. Without adventuring upon the little- 
studied problem of the relation of Micronesia to the other subdivisions 
of Pacific ethnography, we observe that in the Palaus and Carolines 
we find a somewhat clear indication of intercourse with the southern 
Philippines. I have appended the series of affiliates of the general 
Polynesian word for pigeon and have therewith included from Mela- 
nesia Vanikoro, Mota, and Aneityum, from Micronesia the Gilbert 
Islands. It is necessary to use caution in establishing affiliation in 
these groups of languages whose structural laws we are but beginning 
to explore; yet we know that metathesis is of great frequency in the 
speech of the Pacific. With this note of caution it is suggested that 
lupe preserves the early stem, which may have been lupes, for the 
modern disposition of the Polynesian to the open form would account 
