Skip to main content

Full text of "Reports of the Cambridge Anthropological Expedition to Torres straits ..."

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 

d by Google 

-i-»*ir —. 



d by Google 

d by Google 

d by Google 






d by Google 



C. F. CLAY, Mahaobr 

etmtililB)): 100, PBIHCEB BIBEBT 

loniran: H. E. LEWIS, ISa, OOWEB STREET, W.C. 


firriin: A. A8HEB AND CO. 

l^jig: F. A. BBOCEHAUS 

SrfD iotk : Q. P. PUTNAM'S SONS 

Bon^aS atUi Salnitta: MACMILLAK AND CO., Lro. 

All Righti Tamed 

d by Google 








d by Google 

6 7/ 


ffambdsgt ; 


d by Google 


The present volume deals with the arts of life in Torres Straits, including those 
actions and objects which are connected with its material and aesthetic aspects. The 
social and magico-reltgious observances have been fully treated in Volumes v. and vi., 
and the objects (or artifacts as it is convenient to term them) relating to the various 
ceremonies have been described in their appropriate places. In order however to render 
more complete the survey of the material life of the Islanders, these artifacts have 
sometimes been referred to in Che present volume, more particularly when the objects 
in question are also worn or employed on other than ceremonial occasions. As I have 
mentioned elsewhere, the Islanders have such close relations with the neighbouring 
inhabitants of New Guinea (including the islands of the Fly River delta) that it has 
often been impossible to distinguish between their several arti&cts: I have not hesitated 
then to describe many objects which I know to have been imported from New Quinaa'. 
In only a few instances, such for example as the section on Houses, is reference made 
to Papuan artifacts which do not occur on the islands. 

As the essential character of this monograph is to be purely descriptive, I have 
avoided adding parallels from elsewhere, except in a very few cases where reference is 
made to what occurs in the neighbouring parts of New Guinea or Australia. Most of 
these instances will be found to illustrate more fully those actions of which the account 
from Torres Straits is imperfect, for it may be there reasonably assumed that the method 

' Tbe following note at Torning b; ProC H. N. MomUt (.VoCci by a Katumliit on the "Challenger," 1879, 
p. 361) is wonh tepeatjng : " Cape York ia a sort of emporiam ot savage weapons tad omameuta. Pearl abell- 
gathering veascU (Pearl sbellers aa the; are called) come to Bomerset vith arem whiob tbej have pioked up at 
all the isUDds in the neighbourhood, from New Qainea, and from all over the Pacific, aod they bring weapona 
Mid ornameDts from all these plaoaa with them. UoTeorer, the Uamy lalanden Ttsic the port [Somerset) in 
their eanoee, and bring bows and arrows, drams, and snoh things for barter. The water police stationed at 
Somerset deal In these anrioaitiee, bnjing them up and wiling them (o passengers in the passing ateamera, 
or to other visitors. Hence all Unde ot savage weapons have found their way into English collections, with 
tbe label ' Cape Toifc,' aud tbe Northern Australians have got credit tor having learnt the use of the bow-and- 
arrow. I believe that no Australian natives use the bow at alL...Accnr«te determination of locallt; is ol course 
essential to tbe intereet ot savage weapons." What was eharacteristio of Somerset at the time of tbe 
" CbaUengar'B " visit in September 1871 applied to Thursday Island after the seat of Oovemment had been 
transtened thither in ISTT. Even in snab remote islands as Mabuiag and Mer I have obtained wooden clubs 
made by Lofalty Islander*, tbe ooourreDce of wbiob would be dllBenit to explain. if tbe particular circumstances 
were not known. While tbe passing traveller is liable to be deceived with regard to tbe real origin ot the 
objeets which he collects, the investigator on the spot can readily distingnish between native objects and those 
which have been casually imported, directly or indirectly, through the agency of tbe white man. It should not 
be forgotten that natives frequently collect "curios," and where trade is carried on between distant peoples, one 
s finds objects which are not used by their possessors bnt are kept for some sentimental reason. 

d by Google 

of procedure of the Islanders was essentially similar to that of their neighboara. 
A discussion of the racial and cultural affinities of the Torres Straits Islanders will 
be found in Volume I. 

The bulk of the information in this volume is the result of observations made 
during my two visits to Torres Straits, but I have supplemented this with bets drawn 
from earlier writers, more especially Jukes and Macgillivray, and I believe that every 
one of their statements is recorded here ; I have, indeed, quoted from earlier authors 
certain facts which I have myself noted, as the priority of the observations belongs to 
them. I think there is no need for students to consult these earlier writers for focts 
here dealt with, though their works are well worth reading for the historical point of 
view and for general impressions, as they describe conditiona which had passed away 
even at the time of my first visit in 1888 — so rapid has been the change due directly 
and indirectly to the coming of the white man. 

Most of the illustrations of artifacts have been taken from specimens in the 
Cambridge University Museum of Archaeology and Ethnology collected by myself in 
1888-89 and 1898, others are from objects in the British Museum, many of which I 
gave to that institution in 1889, but I have not hesitated to illustrate specimens in 
other collections; except when otherwise stated, the originals of the illustrations are 
in the Cambridge Museum. Numbers which follow the name of the museum are the 
catalogue number of the specimen in question. In some cases I have referred to 
illustrations published by other authors, more particularly to those in the extremely 
valuable Album of the weapons, tools, ornaments, artieles of dress, etc. of the Natives of the 
Pacific Islands drawn and described by James Edge-Partington ; it was issued for Private 
Circulation by J. Edge-Partington and Charles Heape, and is referred to as the AUmm 
of which the First Series was published in 1890 and the Second in 1895, the Third 
Series does not concern us here. 

When possible I have indicated the painted decoration of an object by the conven- 
tional signs used in heraldiy, red by perpendicular lines, blue by horizontal, black by 
cross-hatching, and yellow by dots. 

Occasionally a number in a bracket is placed afrer the name of a native, this 
refers to the genealogical tables in Volumes V. (Western Islanders) and vi. (Eastern 

When no intimation is given to the contrary, the statements refer to the Islanders 
as a whole. If there is any doubt whether a detail is common to them all, the name 
of the island or people is specified, though in such cases it does not necessarily follow, 
unless so stated, that the object or action is confined to that island or people. Native 
names have been freely interspersed, the Western name is indicated by (W.) and the 
Extern by (E.), where no distinction is made it may be taken for granted that the 
word is used by all the Islanders. 

It is my pleasant duty to thank those who have helped me in numerous ways 
in the compilation of this volume. Mr James Edge-Partington gave me quotations fitim 
the volumes of Jukes and Macgillivray and references to a large number of specimens 
in the British Museum, these latter combined with the illustrations in his AUmm have 
saved me a good deal of time. The Directors or Curators of various museums have 

d by Google 

not only afforded me every facility in the examination of the specimens ' under their 
charge, hut they have frequently had photographs taken for me or supplied me with 
information; I hope I have acknowledged their help in all cases in the body of the 
text, but I would like to refer specially to the late Dr A. B. Meyer of Dresden, 
Dr H. O, Forbes, formerly Director of the Free Public Museum of Liverpool, Mr T, A. 
Joyce of the British Museum, and Mr J. MacXaught Campbell of the Ghisgow Art 
Gallery and Museum (Kelvingrove). I must also thank several friends who made 
photographs for me of numerous specimens, more particularly Mrs A Hingston Qui^n 
and Dr W. H. Bansall of Cambridge. Mr S. H. Bay has given me continuous help in 
regard to linguistic matters. To Mr John Bruce of Murray Island my hearty thanks are 
due for the great trouble he has taken in answering questions and in writing detailed 
information on various subjects, I believe his help has been acknowledged in every instance. 
Finally I cannot conclude without acknowledging my indebtedness to Miss L. Whitehouse 
who has given me great assistance by reading my MS. and in helping me to correct 
the proofe. 

The following is the system of spelling which has been adopted : 

a as ia "father" 

6 as German o in 

d as in "at" 

3 as aw in "saw" 

e as a in "date" 


w as 00 in "soon" 

i aa in "let" 

a as in "up" 

£ as ai in "air" 

ai as in "aisle" 

t as M in "feet" 

ou aa ow in "cow 

I as in "it" 

ffi aa ay in " may " 

as in "own" 

oi as oy in "boy" 

B as in "on" 

The consonants are sounded as 

in Englbh r 

n^ as in "sing" 

npg as in "finger" 

A. 0. HADDON. 

April 1913. 

d by Google 






IIL TEX'lILES. By A. Hinoston Quigoin 63 

IV. HOUSES. By A. Wilkin and A. C. Haddos 93 

V. DOMESTIC UTENSILS AND TOOLS. Bv A. 0. Haddon ... 120 


VII. HORTICULTURE. By A. 0. Haddon 144 



Haddon 172 


XI. SCIENCE. By A. C. Haddok, including ASTRONOMY by W. H. R. 

RivKBB and a CALENDAR by S. H. Ray 218 

Xn. MUSIC. By C. S. Mykrb 238 


XIV. SONGS. By A. C. Haddon 284 



By S. H. Bay, A. C. Haddon and J. Bruck 306 

XVIL GAMES AND TOYS. By A. C. Haddon 312 


d by Google i 

d by Google 



1. Awl for piercing the lobe of the ear and weights for distending the 


2. Ear weight, Mer 

3. „ „ 

4. Face Bcarification of a Saibai girl 

5. Girl's cheek pattern. Her 

6. Breaat Bcariftcation of Pauna of Mawata 

Kurubftt of Tutu 

Tag of Pai'ama 

Nisoroa of Mawata 

10. „ „ Magioa of Sf^piaoe 

a woman from laea, Eiwai .... 
Kaubi and Maipi of Saibai ... 

13. Girl's breast scar, Mer 

14. Scarification of Kaubi of Sui, Daadai, Mouth of the Fly River . 

15. Centipede scarification of girls, Mer 

16. Abdominal scarification of Aboka of Boigu 

17. „ „ „ „ 

18. „ „ Babaura of Daudai 

19. Youth and small girl's koitna, Mer 

20. Koitna engmved on a drum, Mer 

21. Scarifications on two wooden images of girls 

22. Koima of Bauba of Mer 

23. „ „ a Miriam man 

24. „ „ Bina of Mer 

25. „ engraved on a coco-nut vessel, Mer ...... 

26. „ drawn by Wanu of Mer 

27. „ engraved on a mask, Nagir (cf. pi. XXXT. fig. 1) . 

28. „ „ „ bamboo pipe 

29. „ „ „ dance wop . . 

30. „ of a Parama man ......... 

31. „ of Daidu of lasa, Eiwai 

32. „ of an Australian native of Gape York 

33. Big girl's shoulder koima, Mer 

34. Shoulder scarificatioQ of Siba of Dauan representing the nose of her 

deceased brother 

35. Similar scarifications on Dauan and Saibai women .... 

36. Shoulder scarification of Gemana of Saibai 

d by Google 


37. Arm scarificationa of a Saibai and a Boigu woman .... 27 

38. Leg scarificatdon of Bonel of Saibai 28 

39. „ „ a Mawata woman ....... 28 

40. Scarifications on th« legs of Miriam men 28 

41. „ „ calf of Sauaoa of Dandai 29 

42. „ „ ,, „ „ 29 

43. Bamboo comb, Mabuiag 32 

44. Feather worn in hair, Mer 34 

45. „ „ „ „ 34 

46. Ornament worn in hair, Mabuiag 34 

47. Palm-leaf frontlets, Muralug 35 

48. Fillet with Coix seeds, Mer 39 

49. Portions of forehead ornaments, Mer 36 

50. Star-shaped forehead ornament, „ 36 

51. Detail of the construction of a diri head-dress ..... 37 

52. Sketch of a faoe motif on a diri 36 

53. Designs painted on certain head-dresses 39 

34. Godegode, ear ornament, Mer 40 

55. Portion of a vxmtz necklace, Mer 41 

56. „ cowiy „ „ 41 

57. „ shell „ „ 42 

58. „ seed „ 42 

59. Gyrena shell pendant .......... 43 

60. Engraved pearl-shell pendant 43 

G1. Dibidibi pendant, Muralug 44 

62. Series of pendants, Mer . . 45 

63. Fret pattern on a mask 46 

64. 0, shell pendants worn by brides, Mer ...... 47 

63. Rsb-bones worn by brides, Mer 48 

66. Sabagorar pendant, Mer 49 

67. Necklace of waraz witli pendants, Mer 49 

68. Saiutd, artificially deformed boar's tusk, Mer 50 

69. „ „ „ „ „ „ 50 

70. Widow in mourning costume, Mer 51 

71. Mourning pendants, Mer 51 

72. SheU belts, Mer 53 

73. Strand oi ». kus took, Mer 54 

74. Belt of pearl-shells 55 

75. Armlet made of two boars' tusks, with tally notches, Mer ... 55 

76. Cone-shell arm-ring 56 

77. Ceremonial arm-guard, Mer 56 

76. Four arm-guard ornaments, Mer ........ 57 

79. Finger rings, Mer 58 

80. Siaiioam, ]eg fringe, Mer 59 

81. Method of tying on the fringe of a bieiteam 59 

82. Pearl-shell scraper used in mat-making, Mabuiag 65 


d by Google 


83. Method of finishing large plaited mat, Mabuiag 65 

81. Sampler shewing conatruction of a gararaud mat, Mabuiag 66 

85. Detail of decoration in a kuri mat, Mer . , . 67 

86. Mat made of pandanus leaves sewn together, imported from New 

Guinea 67 

87. Twist and knots at end of a belt, Muralug ..... 68 

88. Fart of a belt decorated with two rows of cM)wries, Mer . . . 6S 

89. Pattern of overlaid wefts on bands, Mer 68 

90. Bubbing of a Mawata belt <cf. pL IV. fig. 4) 69 

91. Method of construction of an armlet, Inawi, Mekeo District, Papua. 70 

92. Plaited armlet, Mer 70 

93. Detail of armlet, Mabuiag 70 

94. Armlet, Mer 70 

95. Broken nines 71 

96. Zigzag nines 71 

97. Armlet with decoration of cowry shells, Mer 71 

98. Boi, basket made of coco-nut leaf, Mabuii^ 73 

99. Balboi, woman's fish basket, Mabuiag 73 

100. Diagrams shewing the shapes of Torres Straits bflskets ... 74 

101. luside of upper edge of a basket 70 

102. Frayed leaf for the beginning of a basket ...... 76 

103. Upper edge of coco-nut leaf basket, Mer 77 

104. Soft wtdlet of banana leaf, Mabuiag . .- 77 

105. Detail of upper edge of a basket 77 

106. Base of a basket of pandanus leaf, Mer 78 

107. Upper edge of a basket 78 

108. Twilled twos .79 

109. Twilled twos and ones 79 

110. Twilled threes 80 

111. Twilled threes and ones 80 

112. Twilled oblique threes 80 

113. Twilled zigzag threes SO 

114. Horizontal and vertical twilled threes 81 

116. Detail of bands of a basket (pi. XV. fig. 4) 81 

116. „ fitched basket 81 

117. „ coiled basket, Mer 82 

116. „ buzUi epei, Mer 84 

119. Diagram to shew decoration 84 

120. Handle of kula basket, Mabuiag 85 

121. „ burua una, Mabuiag 85 

122. „ coco-nut water vessel, Mer 86 

123. Details of knotting in fringes of petticoats, Mer .... 87 

124. Front panel of tnapa petticoat, Daudai 88 

125. Worked-up rubbing of back panel, a similar petticoat ... 88 

126. Detail of tassels of teid petticoat, Mer ...... 88 

127. Bope formed of coco-nut husk, Mer 89 

d by Google 


no. PAGE 

126. Dugong rope, Mabuiag 90 

129. Worked cord 90 

130. Knots, Mabuiag 91 

131. Three huts, mUgi mud, Muralng 94 

132. Two buts, kai mud. Hunting 95 

133. Head-houee, Mabniag, drawn by Oizn 97 

134. „ „ „ Tom 98 

135. £v)od or men's bouse, Mabuiag ........ 98 

136. File^weUing, Saibai 100 

137. Sectiona of a Mimm house 102 

138. Elevations and plans of Miriam bouses 102 

139. Section of end of a Miri&m house 104 

140. Details of construction of modem bouses ...... 106 

141. Elevations and plans of two modem Miriam houses. 107 

142. „ of modem Miriam houses 108 

143. „ of a Miriam pile-dwelling 109 

144. Elevation and plan of a long-house, Eiwai 112 

146. Section of the same 113 

146. Elevations and plans of bouses at Old Mawata . 116 

147. Details of hut at Old Mawata 117 

148. Bamboo tongs, Mer 120 

149. Fire-sticks, Mer 121 

150. Water vessel, Fubub 122 

101. „ „ Cassis 122 

152. Shell saucepan 123 

153. Bamboo receptacle, Mer 123 

154. Pounding stone, Tutu .124 

155. Mallet and block for pounding, Mer 124 

156. Turtle-shell spoon, Mer 124 

157. „ scraper, „ 125 

158. „ „ „ 125 

159. Shell axe, Mer 126 

160. Shell axe blades, Mer 126 

161. Stone axe blade, Kiwai 126 

162. Rasp of ray skin, Mabuiag 127 

163. Bone cooo-nut buskers 127 

164. Pandanus brush 128 

165. Tobacco pipes (from Jukes, i. p. 165) 141 

166. Digging-stick, Mer 144 

167. Snare for catching tern, Mer 154 

168. Turtl&flhell fish-hooks 165 

169. Sinker for fishing-line, Mer 155 

170. Werir and toerea, Mer 156 

171. Drawing by Pasar of Mabuiag of men fishing 157 

172. „ Waria of Mabuiag of a man catching a turtle . 161 

173. The method of leashing a sucker-fish 164 

d by Google 





A slip knot 165 

Drawing by Warn illustrating the death of Maiak by strangulation 

when dugong fishing 167 

Drawings by U&baif^ natives to illastrat« dugong fishing 166 

Dugong harpoon, Muralug 169 

Woeipon armed with shark's teeth, Mer 173 

Bow, Mer 174 

Long and short arrows with plain heads, Mer 176 

Arrows with simple carved heads, Uer 177 

Carved arrows with beading, Mer 177 

Arrows with varied carving, „ 178 

., „ a single or double row of wooden barbs, Mer . 178 

Bird bolte, Mer 179 

Bird bolt, „ 179 

Bird arrows with more than one point, Mer 180 

Arrows with bamboo heads, Mer . 181 

Decoration on shaft of airows, Mer 182 

Man arrows . 185 

Crocodile arrows 187 

Snake arrow 189 

Ston^-headed clubs, Yam 192 

Unfianged ball-shaped stone head of a club, Mer .... 193 

Wooden club with tally marks, Mer 193 

Two old wooden clubs, Mer 194 

Old wooden club, MerT 194 

Wooden imitation of a star-shaped stone-beaded club, Mer 195 

„ clubs of New Hebrides and Loyalty Islands types 195 

Spear-thrower 197 

Bamboo beheading knife and head- carrier, Mawata .... 199 

Drawing of a warrior, by Maino of Tutu ...... 201 

„ „ feather heEui-dress in its case by Oizu of Mabui:^ . 201 

Qroin-shielda, Mer 203 

Drawing of Kwoiam by Waria of Mabuiag 203 

Amulet of tour tusks, Mer 204 

„ ten tusks, „ 204 

Spathe of coco-nut used as a bailer, Mer ...... 208 

Bow of a canoe, Mabui^ 214 

Koimai designs on Mabuiag canoes 214 

Designs engraved and painted on the bows of Mabuiag canoes . 214 

Figure-head of a Mabuiag canoe 215 

Gozed from the stern of a canoe 216 

Drawing of a canoe by Sunday of Mabuiag 217 

The Baidam constellation drawn by Naii of Mabuiag . . 220 

The Tagai constellation drawn by Mariget „ ... 221 

Tagai and Kareg in their canoe drawn by Oizu .... 221 

Drawing of the Tagai constellation by Waria 232 

d by Google 







Drawings of constellations by Oizu ....... 222 

Drawing of ITek by Gizu 223 

Drawings of coiiatellations by Qizn and Waria 223 

The moon with a halo drawn by Oizn 324 

Drawings of the sun and moon by Waria . ■ . 224 

A t*Uy, ktipe, Mer 234 

Wauri currency, Mer 236 

Stridulating instrument, kat, Mabuiag ...... 370 

" Bones," marep, Mabuiag 270 

Bamboo clapper, k«rker keber, Mer 271 

Shell rattle, Mer 271 

Battle made of ff6a nuts, Mer 272 

Pod of the Queensland bean used as h rattle 273 

Bamboo rattles, Mer 273 

Jew's harp, Mer 374 

Cracking whip, Mer 274 

Models of bull-roarers, Mer 275 


„ „ Mabuiag 376 

Slat of wood simulating a bull-roarer, Mabuiag .... 378 

Warup drum, Saibai 278 

Three buntliuru drums 279 

Drum and Jew's harp, Erub (from Jakes, I. p. 176) 280 

Four drums from the mouth of the Fly River 380 

Fly River drum 281 

Whistle, pereok persok, Mer 282 

„ anper lu, Mer ......... 383 

Pan-pipes, A. Kiwai, B. Mer 2ft2 

Flute, Mer 283 

Shell trumpet, Pulu 283 

Drawing of a death dance by Sunday of Mabuiag .... 389 

Staves held when dancing, Mer ' . 294 

Object held when dancing, Nagir 295 

Dance stave repreeenting a crocodile, Mer 295 

Man wearing a mawa mask, drawn by Sunday of Mabuiag 297 

„ dressed for the mudn kap, drawn by Waria of Mabuiag . 298 

Sketches of masks by Qiza and Sunday of Mabuiag 300 

Tracings of maaks engraved on tobacco pipes ..... 301 

Man wearing the labur mask, drawn by Joani of Mabuiag 303 

The laintr mask, drawn by Gizu 302 

Fruit used as a ball, Mer 313 

Pabn-leaf ball (South Sea type) 313 

Seed tops, Mer 314 

Peg-topa, Mer 316 

Lead spinning top, Mabuiag 317 

Toy throw-sticks, omailer, Mer 317 

d by Google 


Toy bird, Mer . 

Toy rat, Mer 

Groteeqae head as a toy, 

String figures, Poeition I 
Opening A 


Coco-nut paliu 

Tern . 

Bed . 

Tup . 

King fish 

Girl with big ears 

CftDoe with two masts 


Nest of the sun bird . 

Trigger-fish . 


Setting sun . 

Two men fighting 

Cuttle-fish . 


A. children, B. Oazir and Kiam 

Man passing the Main club tamer 



Ball . 

Man going backwavd^ 


Land crab . 
String trick, PUot fish 
Zigzag lines on pipes . 
Punctate lines on pipes, Mawata 
Patterns on a wooden comb, Mer 

„ turtle-shell objects . 

Pattern on a bamboo tobacco box 
Landscape engraved on a pipe . 
Sketch of Mer .... 
Scroll engraved on a pipe . 
Engravings of animals and men 
Drawing of a crayfish by Qizu of Mabuiag 

„ an isopod parasitic on sharks by Oizu 

Two sharks and a turtle engraved on a pearl-shell 
Sting-ray and two hammer-beaded sharks on a drui 
Drawing of a hammer-headed shark by Qizu . 







d by Google 





Shovel-nosed skate engraved on a pipe 
Drawing of a sbovel-noeed skate by Oizn . 
Drawings of two bonito by Gizu 
Drawing of a king Gab „ 
„ a mugarir „ 

Sucker-fish engraved on a pearl-shell . 
Drawing of a Eucker-fish by Giiu 

„ a stone fish „ 

Drawings of two stone fish by Waria 
Drawing of a load by Waria 
Turtle incised on a drum, Saibai 
Drawing of a turtle by Oizu 
Drawings of three monitor lizards by Gizu 
Two snakes engraved on a pipe 
Drawing of a snake by Oizn 
Crocodile incised on a drum 

„ and man's face engraved on a pipe 

„ engraved on a pipe 
Drawing of a crocodile by Gizu 
Sea-eagle engraved on a pipe 
Drawing of a frigate bird by Gizu 

Cassowary with footprints incised on a drum 
Drawing of a cassowary by Gizu .... 

Dog incised on a drum, Saibai 

Drawings of two dogs by Gizu ..... 
Drawing of a flying fox by Gizu .... 
Dolphin engraved on a pipe ..... 

Dugong incised on a drum 

Drawing of a dugong by Gizu 

Canoe with two mat sails engraved on a pipe . 
European boats engraved on a bamboo receptacle, Mer 

Comb with incised ray, Yam 

Decorated combs 

comb from Daudai 

l^pical ornamentation of a loarup drum . 

JDiri design on warup drums .... 

Man with a tUrt on a toarup, Saibai 

Band patterns on warup drums 

Sucker-fish incised on two warup drums . 

Decoration of three buruburu drums 

Daihau incised on a 6uru&uru .... 

Carved band of a duruiuru, Yam 

Daibau and cassowaries incised on a burvburu. 

Lizard and other designs incised on a buruburu 





d by Google 


3£i7. DecorfttioD of a gaima dram, Kiwai 




five gama drums, Dibtri 
a gama, Dibiri 
„ Fly River drum 

., pipe 
the butt-end of dugong harpooDB 
an omaiter 


a dugong barpoon, Parama 
Two decorated bamboo tobacco pipes .... 

Carved pattern at the end of a pipe, Tutu 

Dugong engraved on a pipe ...... 

Conventionalised lish engraved on a pipe, Mer. 
Pipe with two tiger sbarka and other designs . 
Patterns engraved on a pipe, Mer ..... 

Designs engraved on various pipes 

Panel decoration of a pipe ...... 

Man and snake painted on a top, Mer .... 

Frog painted on a top, Mer 

Nageg making a mat painted on a top, Mer 

Kultut painted on a top, Mer ...... 

Meidu „ „ ....... 

Wakai and Kuskus painted on a top, Mer 

Shell containing red paint, Mer ..... 

Figure of a man, tobacco charm, Mer .... 


II II Mabuiag 

madub, Mer . 
newr madub, Mer . 





Sketch map of Torres Straits 

d by Google 


W.H.B., J.B.F., C.G., A.C.H., C.S.M., H.O., G.P., A.H.Q., A.W. signify that Ae 
photographs were taken respectively by Dr W. H. Bansall of Cambridge, Mr J. Bruce 
Freshwater of the Papuan Industnea, Ltd., Mr C. Grover of Rousden, Dr A. C, Haddon, 
Dr C. S. Myers, Mr H. Oldland of the British Museum, Mr G. Penrose of the Truro 
Mnsenm, Mrs A. Hingston Quiggin, and the late Mr Anthony Wilkin. 

Plate I. Fig, 1, Deformation of an infant's head by manual pressure, Mabui^. JiVom 

a drawing by Dr Otto Finsch, Nov. 8, 1882. 
Fig. 2. DeEomied head of an infant about 10 months old, Mabuiag. From a 

drawing by Dr 0. Finsch, 1882. 
Fig. 3. Mammooa, Masig, after Harden S. Melville, Voy. of Fly, I. p. 169. 
Kg. 4. Garia, „ „ „ „ „ „ „ 

Fig. 5. XJlai, Mer (A.W.). 
Fig. 6. Gizn of Muralug wearing a wig (A.C.H. 18S8). 

Plate II, Fig. 1. Alal&n, Badu. From a photograph by Dr O. Finsch. 

Fig. 2. Gaiba, Mabuiag. „ „ „ „ 

Figs. 3, 4. Aigaga, Mabuiag. „ „ „ „ 

Fig. 5. Deau, Mer (A.W,). 

Kg. 6. Eausa, Mabuiag (A.W,), 

Flatb III. Fig. 1. Small coconut water vessels engraved with a Hp tor koima, Homiman 

Fig. 2. Breast scarification of Bonel of Saibai, after B. Bruce. 
Fig. 3, Round house at Enib with group of people, after Harden S. Melville, 

Sielehet, pi. 18. 
Fig. 4. ^enr madub, images of girls used in erotic magic. A. Mer, 346 mm. 

high, B. Masig, 315 mm, high (H.O.). 

Plate IV. Scarifications on the loins of women representing their totem animals, from photo- 
graphs and drawings by A.C.H. (1888). 
Fig. 1. Meke of Tutu (Baibemm). 
Fig. 2, Patagam of Mabuiag (Snake). 
Fig. 3. Ado of Badu (Dugong). 
Fig. 4. Wagud of Mabuiag (Dugong). 

Platb V. Figs. 1. Esaka, 2. Nawe, 3. Piwini, 4. Tibi, all of Badu (A.W.). 
Figs. 5. Gupi, 6. Monday, 7. Bagari, all of Mabuiag (A.W.). 
Figs. 8. Bablo, 9. Canoe, both of Mer (A.W.). 
Fig. 10. Maine of Tutu dressed for a war dance (A.C.H. 1888). 

d by Google 


Fig. 1. Din, Tutu (O.H.). 

Fig. 2. Dari, Mer. 

Fig. 1. Doffui, Cassowary feather head-dress, Mabuif^ (H.O.). 

Fig. 2. Dagam, head-dress of ptutnes of bird-of-parodise, Saibai (H.O.). 

Fig. 3. War head-dress of Kebisu of Tutu (H.O.). 

Fig. 4. Triangular piece of turtle-shell probably used in dressing the hair, 

Mer (G.P.). 

Fig. 5. Ka»i eauad, Mer (H.O.). 

Fig. 6. Man with umtm, Mer (A-C.H. 1889). 

Plate VIII. Figs. 1, 4. Gadodo, figa. 2, 5. Mamai. figs, 
of Mer (A.C.H. 1889). 

3, 6. Pasi, dresited for a dance, all 

Personal omanienta in the British Museum (H.O.). 

Fig. 1. Fillet decorated with Coix seeds. 

Figs. 2, 3. Frontlets of Nautilus nacre. 

Fig. 4. Coronet of dogs' teeth. 

Pig. 5. Coronet of kangaroos' teeth. 

Figs. 6 — 12, Shell nose-sticks. 

Figs. 13 — 16. Ear pendants of pearl-shell. 

Fig. IT. Ear pendant of Nautilus nacre. 

Fig. 18. Necklace of vjarax shells. 

Fig. 19. IHbidihi pendant on bead necklace. 

FigB. 20 — 22. Creacentic pearl-shell pendants. 

Fig. 23. Pearl-shell breast ornament. 

Figs. 24—26. Pearl-shell pendants. 

Figs. 1—4. Ear-weights (H.O.). 

Figs. 5—12. Comba (H.O.). 

Fig. 13. Ear- weight. 

Fig. 14. Sabagvrar. 

Figs. 15—17. Perforated pearl-shell discs i 

I by germjere h, Mer. 

Fig. 1. Turtle-shell fish-hook, Mer (H.O.). 

Figs. 2 — 10, Sabagorar, turtle-shell ornaments worn by brides, Mer (H.O.). 

Figs. 11—17. Turtle-shell bodkins, ler, Mer (H.O.). 

Fig. 18. Dogs' teeth necklace, Mer (A.H.Q.). 

Fig. 1, Sporran, Mer (W.H.B.). 

Fig. 2. The Main drum and stone-headed clubs, Mer (A.W.). 

Fig. 3. Finial of a Miriam house, Mer (A.W.). 

Fig. 4. A. B. Amulets of tusks wont when fighting, Tutu <H.O.). 

Fig. S. Ager, an edible rhizome, Mer (A.W.). 

Fig. 6. Carving in coral of a pigeon, Masig. 

Fig. 7. Model of a bvil-roarer, Muralug (H.O.). 

Fig. 8. „ „ Moa 

d by Google 


Plate XIII. Fig, 1. Mat, pattern: "crab's footprints," Mabuiag. 

Fig. 2. „ "shark's wife,'" „ 

Fig. 3. „ "bending spirit," „ 

Fig, 4, „ "like a glass bottle," „ 

Fig. 5. Baby's mat, Mer (A.H.Q.), 67 cm. x 34 cm. 

Fig. 6. „ „ „ 45 cm. x 20 cm. 

Fig. 7. Portion of a mat „ 69 cm. x 38 cm. 

Plate XIV. (A.H.Q.). Fig. I. Fringed belt from Daudai. 

Fig. 2. Armlet decorated with orchid skin, Mer. 

Fig. 3. Beaded annlet, Mabuiag. 

Fig. 4. Coarsely plaited armlet. 

Fig. 5. Narrow armlet of banana leaf, Mabuiag. 

Fig. 6. Broad armlet with two rows of cowry shells, Mer. 

Fig. 7. „ „ one row 

Fig. 8. „ „ beads. 

Figs. 9, 10. Bracers. 

Figs. 11, 12. Dance bracera, Mer. 

Plate XV. (A.H.Q.). Fig. 1. Woman's basket for yams, of jiandanus. 

Fig. 3. Man's basket for fishing line, of paiidanus, Mer. 

Fig. 3. Banket of coco-aut {lalm leaf. 

Fig. 4. „ pandanus. 

Fig. 6. „ „ Mabuiag. 

Fig. 6. „ coco-nut palm, Mer. 

Plate XVI. (A.H.Q.). Fig. I. Man's basket for fishing line, of pandanus, Mer. 

Fig. 2. Basket of pandanus, Mabuiag. 

Fig. 3. Woman's basket, Moa. 

Fig. 4. Tray of coco-nut palm, Erub. 

Fig. 5. Lined basket of pandanus, Mer. 

Fig. 6. Samoan method of beginning a basket, Badu. 

Pl^TE XVII. (A.H.Q.). Fig. 1. Woman's basket for carrying yams, Mabuiag. 

Fig. 2. Basket of Flagellaria, Mer. 

Fig. 3. Basket of coco-nut palm. 

Fig. 4. Coiled basket for a top, Mer. 

Fig. S. Basket of Xerotes. 

Fig. 6. „ „ Muralug. 

PLAfB XVIII. (W.H.B.). Fig. 1. Teid leaf and fig-tree lark, ome, petticoat, neai 

Fig. 2. Ome nesur, Mer. 

Fig. 3. Banana leaf petticoat, kaba nesur, Mer. 

Fig. 4. „ „ and fig-tree bark petticoat, zazi, Dauan. 

Plate XIX. <A.W.). Fig. 1. Huts, Old Mawata, Daudai. 

Fig. 2. Huts, Old Mawata, Daudai. 

Fig. 3. Huts, Tutu. 

Fig. 4. Old hut, Mer. 

d by Google 


XX. (A.W.). F 

Fig. 4. 



Fig. 3. 


1. File-dwellings, Saibai. 
Pile-dwelling, Saibai. 
Round hoQse, Ulag, Mer. 
Interior of old hat, „ 

Pile-dwelling and topsop banana, Mer (A.W.). 

Long house and woman with limed hair, lasa, Kiwai (cf. pi. XXXIX. 

fig. 2) (A.W.). 
Group on bwwh, Dauar (A.C.H. 1889). 

'. XXII. (A.W.). Fig. 1. Sai or fish-weire, Mer. 

Fig. 3. Method of carving a dugong, Mabuiag. 

Fig. 3. A feast, Mer. 

Fig. 4. Removing sand from an earth-oven, Mer. 

: XXIII. Fig. 1. Man with tvap on a dugong platform, Mabuiag (from a photo by 
A.C.H. 1888). 
Fig. 2. Three- and four-angled beads of dugong harpoons (W.H.B.). 
Fig. 3. Butt«nd of a dugong h^poon. 
Fig. 4. Nomoa of Mabuiag with tvap and two dugong (A.C.H. 1888). 

Preparing for a voyage in a canoe, Erub, from H. S. Melville. 
Western Island huts, Damut, from H. S. Melville. 

Plats XXIV. Fig. 1 
Fig. 2 

Plate XXV. (A.W.). 


Fig. 1. Canoe with single ontngger, Mer. 

2. Canoe, "Adi," with double outrigger, Mabuiag. 

3. Stern of canoe "Adi." 
i. Bow „ „ 

Plate XXVI. (A.C.H. 1888). Fig. 1. Midships view of Mabuiag canoe with mat sails. 
Fig. 2. Side view of Mabuiag canoe with mat sails. 

Plate XXVII. Fig. 1. Figure-head, ddgai, of a canoe, Saibai (A.H.Q.). 
Fig. 2. JFarup drum, Tutu (O.H.). 
Fig. 3. Wooden club, Mer (O.H.). 

Plate XXVIII. (A.W.). Fig. 1. Top spinning, Mer. 

Fig. 2. Modem bouses of "South Sea" type, Mer, 
Figs. 3, 4. A divinatory game, koko, Mer. 

Plate XXIX. Fig. 1. Two men dressed for a war dance, one shewing method of ai 
release, Muralug (A.C.H. 1888). 
Two men shooting with bow and arrow, Mer (A.W.), 
Stone head of club, Torres Straite. 

„ „ Daudai. 

Bamboo beheading knife, mouth of Fly River. 

Fig. : 

Fig. ; 

Fig. ■ 

Fig. ; 

Fig. ' 
Fig. ; 

Stone club used in initiation ceremonies, Mer (A.H.Q.). 

d by Google 


Figs. 1- 

Plitb XXXI. 

PigB. 1- 


Kg». 1- 

Kg.. 8- 


Kg.. 1, 

Fig. 3. 

Plate XXXIV. 

Figs. 1, 

Plate XXXV. 

Fig. 1. 

Fig. 2. 

Plate XXXVI. 

Figs. 1, 

Fig. 3. 

Fig. 4. 

Plate XXXVII. 

Fig.. 1, 

Fig. 3. 

Fig. 4. 

Fig. 5. 


Fig. 1. 

Fig. 2. 

Fig. 3. 

Fig. 4. 

Plate XXXIX. 

Fig. 1. 

Fig. 2. 

Fig. 3. 

Fig. 4. 

Plate XL. 

Pigs. 1, 

Fig. 3. 

Fig. 4. 

Fig. 5. 

Fig. 6. 

Figs. 7- 


—12. Arrows carved to represent a man. 

-13. Arrows carved to represent a crocodile. 

- 7, Degenerate crocodile arrows. 
-14. CUw arrows. 

2. Secular dtknce, Mer (C.S.M.). 
Ornaments hold when dancing, Tutu (H.O.). 

2. Side and upper views of a mask, Yam (H.O.). 

Side view of a box-mask with a fish, Moa (H.O.). 

„ „ turtle-shell fish mask, Mabuiag (H.O.). 

2. Mask from Aurid. 

Front view of mask on pi. XXXV. fig. 1 (H.O.). 

Funeral mask of turtle-shell, Mer (H.O.). 

2. Bamboo tobacco pipe. 

Stone top in its basket, Mer (A.W.). 

Masked dancer paint«d on a stone top, Mer (A.H.Q.). 

Wooden model of a shark, Mer. 

Turtle-shell fish mask. 

Engraved design on same (CO.). 
Trigger-fiah, Monacanthus (A.W.). 

Oanoe at Dirimu, Binaturi River, Daudai (J.B.F.). 
Long house and canoe at Madiri, Flj River (J.B.F.). 
Turtle-shell effigy of an insect, Mer (Q.P.). 
Sketch of same from ventral side. 

2. Upper and under sides of a wooden model of a ray (O.H.). 
Wooden model of a ray. Tutu (O.H.). 

„ „ a shovel-nosed Bkat«. 

„ „ the head of a sucker-fish, Nagir (O.H.). 

„ „ A turtle. Tutu (O.H.). 

-10. Wooden models of dugong, Mer (O.H. and W.H.B.). 

d by Google 


12. Line 7, to 'IS.' lead 'X.' 

21. LinB 8 (rom bottom, for 'in,' read 'IV.' 

25. Line 4, after 'HaMam' tid '(pi. XXXT. Qg. ]).' 

31. Une 6, for 'J.' read 'O.' 

33. Line 17. add * pi. XXIX. Sg. 1.' 

35. Line IS, for 'gagai' read ' gaigaV 

36. Line 30, for ' Bg. 2' read 'fii;. 1.' 

43. Line 5, for ' Bhinobatis ' read ' lUuDobatDH.* 
S2. Line 6 tram bottom, after 'DeDdrobiom' add 'baingan (W.).' 
61. LiM 10, delete 'at in.' 
81. Line 11, (or ■ flg. 1' read 'figa. 1, 2, *.' 

68. Line 25, after ' Hew Qoioea ' add ■ (Gg. 70 and pi. IV. flg. 4).' 
72. line 12, before 'epei' add 'aiput or.' 

72. Line 19, after 'vatti' add 'or vain li' and delete 'umiU <Tuta}.' 
77. Line 2 of eiplanation of fig. 105, tor 'ooveied' read ' nnfiniibad.' 
104. Line 1, add 'XIX. fig. 4.' 

108. Line 14, for • XXIU.' read ' XXVIU.' and add ■ (see also pi. XXXni. flg^ 1, 2).' 
Line 9, for 'XXIIL' r«ad -XXXm.' 

Line IS from bottom, after 'fig. 144,' add 'pi. XXI. fig. 2, tee alao pi. XXXIX. flg. 2.' 
Line 15, for ' nipa palm ' read ' Imnkt of a palm. The trunk ia epUt, the pith rsmoTed, and 

the aurfaoe flattened ont.' 
Line 7, after 'ones.' add 'A long honee at Hadiri ie Bbewn in pi. XXXIX. flg. 2.' 
Line 17, add ' I waa informed that the aie of the oooo-nDt palm leaf broom, bSi rid (oooo-palm 
leaf bone), wai introdnoed into Mabaiag, and before then the; and the dried ioloresoence, 
maupa*, as a broom.' 
Line 1, add ' and flg. S72.' 
Line 7 from bottom, tor 'rii' read 'nit.' 
Line 0, add 'pL XXXTU. flg. 3.' 
p. 16fl. Line 2, add 'A oarrad wooden model of a turtle wbioh waa fastened to the bow of a canoe in 
Tnto to eonetrain the turtle to come and be oanght ii ahewn on pi. XL. fig. 6, and the 
carved head of a aneker-Sab from Nagir (pi. XL. flg. 0) was probabl; owd tor an analogooa 
p. 169. line 19 from bottom, for 'S67-870' read '865—388' and for '86611, 871 ' read '864 i, 869.' 
p. 171. Line 13, between -aee alao' and •Album' add 'pi. XL. flga. 7—10 and.' 
p. 217. Line 1, for '355 " read '254.' 
p. 292. Lin* at end of the third paragraph, add ' Various danoing positions are shewn in pi. TCVXlll 

figt. 1, 2.' 
p. 299. Line 21, for ' 299 1 ' read ' 800 a.' 
p. 368. Fig. 368 delete ' from Saibai.' 

















d by Google 


In this volume, which deals mainly with the daily life of the Torres Straits Isiandere, 
the various occupations are naturally dealt with separately ; but as this gives a disconnected 
view of their ordinary life, I have thought it desirable to prefece it with a more general 

The natives rise at dawn and the men and women (the latter taking their children 
with them) separately saunter off to different portions of the scrub that is left near the 
village. On these occasions perfect propriety is observed, the two sexes always remaining 

There are two main meals in the day, one in the early morning and the other at 
sunset, but in addition they eat at all times of the day ; the Miriam children especially are 
rarely about for many minutes together without finding a piece of sugar-cane to chew or 
a banana to munch. Whilst the women are occupied in their gardens they fill in part of their 
time in roasting food, and they eat or work and idle as they feel inclined. There is an old 
saying attributed to Meidu (VI. p. 14), " Miriam and Dauar men you begin food to eat small 
daylight and at night (are) finishing," in other words, owing to the abundance of food, the 
Murray Islanders eat from sunrise to sunset or even later. A Murray Islander informed 
Dr C. S, Myers, " Sun he come up, sun he go down, eat and drink all day before missionary 
come. Missionary he make him eat, breakfast sun there, dinner sun up there, and supper 
sun down there'." Coco-nut milk and water are the universal beverages. Alcoholic drinks 
are scarcely ever brought to the ishinds. These two statements refer more particularly to 
the Murray Islands, the more immediate contact with Europeans of the Western Islanders 
has considerably modified their food and drink. Certainly in the old days all the natives 
drank nothing but water and coco-nut milk. Tobacco-smoking is very general, but by no 
means universal, among the adults, male and female; children do not smoke. 

The natives go to sleep at all hours of the night, but rise at daybreak. A Miriam 
man said to Dr Myers, "We go sleep midnight. We get up along sun. We go sleep 
sometime two, sometime three hours, sun up high. Suppose we tired, we sleep longer'." 

Perhaps bathing is less common since the enforced wearing of European attira The 
women wear a hideous long loose-fitting gown ; the men's dress varies from a mere loincloth 
to a complete European suit of clothes. The men rarely, and the women never, dream of 
changing their clothes before or after bathing'. 

The Western Islanders are essentially a settled people, but in certain islands more or 
less nomadic habits prevailed until recently. The Muralug people had their head-quarters at 
' St BarthoUmew't Boipital ReporU, hit. p. 82. 

H. Vol. IV. 1 

d by Google 


Port Lihou, but wandered about the isUnd in communitiea in quest of food ; this they did to 
a certain extent as late as 1888, and probably later still. As they can scarcely be said to 
have cultivated the soil, they were more dependent upon wild fruit for food than the other 
islanders. The Island of Yam being the garden of Tutu, there would necessarily be a 
good deal of going backwards and forwards between the two islands, especially at the 
planting season and at harvest time. There was a good deal of "flitting" of the popu- 
lation of the smaller islands in the central region of Torres Straits ; for example, Macgillivray 
found most of the Nt^r people temporarily settled at Warabar, and often the whole of 
the population, or nearly so, of such islands as Aurid, Masig, etc., would be located on 
some outlying sand-bank for weeks together, mainly for the purpose of catching turtle and 
feeding on them and their eggs. The unexhausted reefe would supply any amount of shell- 
fish and other food. 

The migratory existence of the Central Islanders is well illustrated in an account in the 
Ifaut. Mag. (vi. p. 659) of the natives who massacred the crew' of the Oharlea Eaion in 
1834. About sixty natives resided on Pullan during the fishing season, feeding chiefly on 
" turtle and small Ssh, which they caught with hook and line ; and shell-fish, which abound 
on the reefe. The island also produces a small finiit ' like a plum with a stone in it,' 
probably a species of eugenia The fish is broiled over the ashes of the fire, or boiled in the 
basin of a large volute.... The island is covered with low trees and underwood, and the soil 
is sandy. In the centre is a spring.... After remaining here two months the Indians 
separated. One party. ..after half a day's sail reached another islet to the northward, where 
they remained a day and a night on a sandy beach; and the next morning. . .reached 
another island similar to Pullan..., where they remained a fortnight. They then proceeded 
to the northward, calling on their way at different islands, and remaining as long as they 
supplied food, until they reached one [probably near Aurid] where they remained a month ; 
and then they went on a visit to Damley's Island, which they called Aroob, where, for the 
first time, Ireland says he met with kind treatment. After a fortnight they again embarked, 
and returned by the way they came to an island called Sir-reb [Sirreb or Marsden Island 
lies three milesN.W.ofMassied (Masig)]. ..where their voyage ended, and they remained until 
purchased by Duppar, the Murray Islander," who " learning that there were two white boys 
in captivity at Aureed embarked in a canoe with his wife, Pamoy ; and went for the 
express purpose of obtaining them. ..the price of their ransom was a bunch of bananas for 
each. They returned by way of Damley's Island, where they stopped a few days, and then 
reached Murray's Island, where they remained ever since most kindly treated.... When at 
Aureed the Indians had named Ireland, Wak, and little D'Oyly they called Uass." The 
latitudes of Pullan and Emb are about two degrees (120 nautical miles) apart. We thus have 
evidence that the Central Islanders voyaged to the islands and sand-banks within the Oreat 
Barrier Reef to a distance of over a hundred miles. 

Wyatt Gill (pp. 200, 201) says of the Badu people in 1872, that they, "like the 
aborigines of Australia, build no houses, and have no fixed place of aboda The cause 
of this bird-like mode of existence seems to be that the Batu [sic] people never 

■ Some of the crew were mnrd«r«d on Bojdin, ODe ot the HannibAl Isluids near the Aaitrativi ooMt and 
due west ot Raine leland eatraooe, where the Dativee had endentl; ([one to flah. The reat ot the orew were 
moTdered on Pallftii, a neighboormg island. 

d by Google 


cultivate anything, living on fruits and roots growing apontaneouBly ; so, like tramps, they are 
compelled to be continually on the move." I suspect this usually vety trustworthy observer has 
oonibsed the Badu with the Moa people, for, according to the folk-tale (v, p, 36), Yawar of 
Badu was a very successful gardener, and we know that the natives of Badu and Mabuiag 
have long been intimately connected ; the latter certainly cultivated the ground, and thanks 
to Dr Landtman I am enabled to give an account of horticulture in Badu. The Moa people, 
on the other hand, had close relations with the inhabitants of Muralug. A few years before 
my visit to Dauan in 1888 the natives of Boigu had migrated to and settled in the former 
island on account of the ravages caused by the periodical descent of the Tugeri pirates 
on their island ; and I saw in Saibai a numl»er of Dauan natives who were living there for the 
same reason. As the Tugeri men only came down during the north-west monsoon, some of 
the Boigu people paid occasional visits to their old home to make their gardens or to bring 
away the produce. 

The Eastern Islanders were permanently fixed to the soil, but in the Murray Islands 
considerable bodies of people from Mer visited Waier to attend the Waiet zogo {vi. 
p. 278), or when there was a taboo on Dauar the natives camped at Eeauk on Mer 
(VI. pp. 170, 172). 

At certain seasons of the year a considerable amount of time is spent by the Miriam 
in preparing their gardens, and nearly every day the women had to go there to procure 
their daily supplies of food. The men do all the heavy work in horticulture ; they cut down 
the bush and clear the ground when new gardens are required, or clear the overgrowth 
in those about to be replanted, and make the fences. On the whole they may be said to 
take their fair share of the work. 

A little fishing is indulged in by both sexes when they feel inclined for a change of 
diet; but at certain periods fishing becomes more of a general occupation. At low tide 
men, women and children may be seen searching the reef for shell-fish and fish which 
have become imprisoned in rock -pools, but as a rule this simple collecting is done more by the 
women and children. Although serious fishing is more particularly men's work the women 
also take a part, but definite fishing expeditions and the quest of dugong and turtle are 
confined to the men. Practically the fishing of the women is limited to that which they can 
undertake on the fidnging reef of their home Island. 

The building of a new house is a noteworthy event. It takes some time gradually to collect 
all the materials for its construction. In former days, however, it must have been much less 
uduous work to build the simple huts characteristic of most of the Western Islanders than 
to erect the round houses of the Eastern Islanders ; the modem South Sea type of house is 
a relatively elaborate affair. Feasts are given to those who assist. An Eastern round house 
had to be renewed about every six years, and the Western huts were probably less durable. 
The bouses are little used in the daytime except in wet weather. The people often sit under 
a high framework which is roofed with palm leaves. 

The digging and clearing of water-holes, or wells, was not an easy task, with the 
imperfect tools at their command, and it required the help of friends, who were repaid by 
a feast. 

The men sometimes went considerable distances to hunt dugong or turtle, and to 
seek for turtles' and terns' eggs on sand-banks, or to hunt on distant reefs at low spring 


d by Google 


tides for shells to be employed in domestic use or for the making of ornaments. Many objects, 
especially the elaborate masks, were made of turtle-shell (tortoise-shell) for which the 
hawksbill turtle had to be caught. It appears from J. Lane Stokes (Diacoveries in Avstraiia, 
u. 1846, p. 257) that the Miriam or other Torres Straits Islanders on these expeditions went 
far down the Great Barrier Reef, as he met with them at Restoration Island, near Cape 
Weymouth, 165 nautical miles S.E. of Mer. The whole area of Torres Straits from the 
Great Barrier Reef on the east to the deeper water in the west, and from the mouths of 
the Fly river to Boigu on the north to the northern point of Cape York on the south, was 
more or less known to the islanders; probably the Western and Eastern Islanders kept 
mainly to their respective halves of this area, but may have overlapped to some extent in the 
central islands and ree&. There is no doubt that practically every man had a veiy extensive 
and at the same time sufficiently precise knowledge of a lai^ area, being acquainted not 
only with the special products of various islands, but with the position of sand-banks and 
reefe that are exposed only at low tides, and with the seasons for collecting the marine 
fauna of which they were in need. In this and in their gardening operations they were 
assisted by their knowledge of the movements of the stars, many of which they had grouped 
into named constellations. For these voyages they must have had a considerable amount of 
weather-lore and a knowledge of tides and currents. 

The trading voyages, too, gave them a wider outlook and brought them into contact 
not only with other islanders, but with Papuans on the one band and, more rarely, with 
Australians on the other. One result of trading voyages, of friendly visits, and probably 
sometimes of war (V. p. 234) was occasional intermarriage ; for example, Melville, who was 
on the " Fly," mentions meeting a New Guinea woman in Erub (see explanution of pi. III. 
fig. 3). It is recorded in the Miriam genealogies (vr. pp. 67 — 91) that two men married Fly 
river women, and thirty-nine marriages have taken place between Murray Islanders and the 
natives of Erub and Ugar. Dr Rivers states that, " Erub women have a great reputation 
among the Murray Islanders as hard workers, and the comparatively small number of Miriam 
women outside their island suggests that they have not an equal reputation with respect to 
this first requisite in a wife" (vi. p. 120). Only four marriages are recorded between the 
Miriam and Central Islanders, There is no record of any intermarriage between the Miriam 
and Western Islanders, except of late years since the breaking down of the old division 
between the two groups. In recent years two Miriam men have married Australian 
women. The most typical of the Western Islanders, the Gumulaig (natives of Mabuiag 
and Badu), in fonner days married, with few exceptions, among themselves (v. p. 233). 
There is evidence that intermarriages between the Saibailaig (Boigu, Dauan, and Saibai) 
and the inhabitants of Daudai were not infrequent, and probably the same occurred among 
the Kauralaig (Prince of Wales Islanders, etc.) and the Australians of Cape York. On the 
whole one must admit, however, that intermarriage could not have had any appreciable 
effect in the interchange of culture. 

In both the Western and the Extern folk-tales there are several instances of a cultural 
drift from New Guinea to the islands and among the islands mainly fivm west to east. 
Certain tales tell of the spread of improved methods of horticulture or fishing, while others 
record the introduction of new cults. This subject is discussed more fully in the first volume ; 
it is mentioned here to prove that the cultural life of the people, from their own shewing, has 

d by Google 


not beeD Btatiooaiy trom time immemorial, but that outaide iBfluences have come in from 
time to time in the different islands, and the natives have adopted new methods and ideas 
(vi. p. 2), Nor should the influence of foreigners be ignored, for during the past forty years 
this has been constantly increasing. The object of the expedition, however, has been mainly 
to record purely native conditions rather than to describe the present modified conditiona 
The main effect of this influence has been to eliminate individuality ; the distinctive houses, 
canoes, implements, ornaments, customs and ceremonies have either passed away or are 
rapidly becoming obsolete. In 1886 Dr Otto Finsch wrote, " Auf den Inseln der Torres- 
Strasse ist nickts mehr zu haben, da der rege Verkehr der Ferlfischereien alle Eigenthilm- 
lichkeiten ausgelSscht hat. Noch vor wenigen Jahren verfertigten die Eingebomen sebr 
originelle und kunstreiche Masken aus Schildpatt, jetzt machen sie dieselben aus Blech 
von weggeworfenen Conservebiichsen I " It must be bome in mind that he was only a 
short time in Torres Straits in 1882 and visited but few islands. He was essentially on 
a collecting expedition and had no time to study the social and magico-religious life of the 
people. Had bis statement been strictly true the present volumes could not have been 

A great deal of the time of the Miriam was taken up in preparing and performing 
the many ceremonies connected with their zogo, several of which are described in Vol. vi, 
pp. 192 — 280. The Bomai-Malu cult (vi. pp. 281 — 313) one way and another absorbed a 
considerable amount of time and energy. The elaborate death ceremonies (VI. pp. 126 — 162) 
not only took up a great deal of time, for the natives were very punctilious in canying 
them out, but gave rise to increased activity in the gardens, as large quantities of food had 
to be provided for the accompanying feasts. Indeed had not the very old and the very 
young dead been exempted from the full funeral rites, the living would have been perpetually 
occupied with funeral celebrations. Our knowledge of the Western Islanders is less complete 
than that of the Miriam, but from the data we have published and other information which 
has been too fragmentary to publish it is evident that in their case also religious and other 
ceremonies must have taken up a good deal of time. 

There were many social events of the Miriam in which feasting and the exchange of food 
were prominent features ; this necessitated the canying of large quantities of food-stuff fiom 
their gardens to the places where the feasting occurred ; some of these social ceremonies took 
from two to three weeks from first to last. 

The making and decoration of weapons, implements, ornaments, and dance and ceremonial 
paraphernalia necessitated much labour, especially when we remember the primitive character 
of the tools at their command. 

When all these circumstances are taken into account it will be evident that in the olden 
days these savages were by no means lazy. Time, energy, thought, ingenuity were employed 
not only in material existence but for social customs and ceremonies, and one is apt to over- 
look or underrate the very important part which the latter play in the existence of savages. 
Pride, fear of ridicule, the religious sentiment, and the stimulus of competition which was 
keenly felt all served to keep them up to the mark. 

The Miriam and probably the other Eastern Islanders of Enib and Uga must have spent 
happy lives. They were not liable to attack from enemies, there was an abundance of food 
and they could do as little routine work and indulge in as much amusement as they chose. The 

d by Google 


fishing operations were also largely of the nature of a recreation, as were the veiy numerous 

Life was certainly not so easy for the Western Islanders, The islands are less fertile and 
the inhabitants had to depend to a larger extent than the Eastern Islanders on the spon- 
taneous produce of the soil (which was not of much account) and on fishing. Fighting was 
also more frequent. 

The women appear to have had a good deal to say on most questions and were by no 
means downtrodden or ill-used. Macgillivray (Vol. il. p. 9) states that the Eastern Islanders 
"always appeared to me to treat their females with much consideration and kindness." 
I should say that this was characteristic of the islanders as a whole, but an exception must be 
made in the case of the inhabitants of the Prince of Wales Group, for according to Macgillivray 
(whom I have quoted in Vol, v. p, 229) the Muralug often beat their women and infiicted 
savage acts of cruelty on them. 

d by Google 


In this section I deal with the treatment that is applied to the form and sur&ces of the 
head, body and limbs, and also with the implements used in artificial deformation of the nose 
and ears. Objects which are attached to the person are deaciibed in the next section. 

Artificiai. psfobhation of the Head 7 

Nose akd Ear pikrginq 9 

Other forks of deforuation 12 

Permanent decoration of the Skin (Scarification and Tattooing) 13 

Tbmporabx decoration of the Skin (Paintino) 89 

Tbkathbnx of the Hair 30 

Toilet 32 

Artificial defobhatiox of the Head'. 

In Mabuiag, and doubtless in other western islands, a bead had to be low in the forehead, 
atad paru, flat at the back and not too well developed above, in order to be considered 

To obtain this ideal of beauty the mothers of Mabuiag resorted to two devices. 

In the first instance a handsome man was called in before the birth of the child, to sit 
behind the mother. Such a custom presupposes a veiy strong belief in the efficacy of pre- 
natal influences. 

In tbe second place there was the practice of artificial deformation, paru luaian, of the 
infant's head by skilful manipulation. The mother placed her left hand on the occipital 
protuberance, kote, (in general but slightly marked among the islanders), her right being kept 
free to smooth down the forehead, paru, bregma or region of tbe frontal fontanelle, si, or 
vertex, guai. The head was also firmly stroked from the outer margin of the orbits back- 
wards along its lateral surface, and from the same point forwards and downwards along tbe 
side of the face, following roughly the direction of the jaws. The process was continued 
whenever the mother felt so inclined, either by night or by day, until the portion of the 
skull about the bregma ceased to bend under the pressure. My fi^end, Dr O. Finsch, has kindly 
permitted me to reproduce drawings (pi. I. figs. 1, 2) made by bim in Mabuiag in November, 
1882, which illustrate the method and result of this manual pressure, 

> I Am indebted to Ui Wilkia and to Dr Seligiiuaii for lome of the obwrTAtionB in this aection. 

d by Google 


It is said at Mabu&ig that the people of Boigu, Dauan, Saibai, Mabuiag, Badu, Moa, 
Waiben, Muralug and of Mowata and Tureture have a similar custom of head deformation, 
but that the Masingara bushmen do not practise it, though it is thought to be done at 
Parama, Kiwai, etc 

The inhabitants of the mainland of Australia, with whom the people of Mabuiag occa- 
sionally came in contact, were despised on account of their bulging foreheads, high crowns 
and prominent occipital protuberances, whence they have been called half contemptuously, 
half jestingly, koiaar hwikukya, i.e. too many heads. In Vol, v. p. 81, it is pointed out that 
Kwoiam, the great hero of Mabuiag, had a head of this kind ; he in common with the 
Australians is also said to have bad a long narrow head, saked kwik, as contrasted with 
the atad hunk, or flat, broad bead, resembling the plastron (ventral shield), ata, of a 

Dr C. S, Myers has published in the St Bartholomew's Hospital Reports, Vol. xxxv, p. 95, 
the following information : "As to post-natal deformity, a native [of Mer] informed me, 
' When piccaninny bom, him head too long, too wide, too round. Woman she lay hand 
on sides of head or on front and back. She press sometime one hour, sometime more. That 
old-time fashion, that no longer '." 

It has been questioned whether temporary and discontinuous manual pressure of an 
infant's head could permanently affect its shape. The evidence fiv)m Torres Straits is as 
follows. Various travellers, myself among the number, have seen motbera so engaged. No 
mechanical method of pressure was adopted, either by means of bands or boards. Some of 
the skulls afford indication of artificial deformation ; this is mentioned in vol. I,, where it is 
also pointed out that there has been an immigration of a brachycephalic people into the 
western islands. It would seem that there was a desire on the part of the Western Islanders 
to exaggerate the normal low brachycephaUsm of a part of the population, and also to minimise 
the dolichocephaly of the remaining portion. 

The following is all the previous information that I have been able to gather on the 

" A peculiar form of head, which both the Kowrarega [Kauralaig = Prince of Wales 
Group] and Gddang [Cape York] blacks consider as the beau ideal of beauty, is produced by 
artificial compression during in&ncy. Pressure is made by the mother with her hands — as 

1 have seen practised on more than one occasion at Cape York — one being applied to the 
forehead and the other to the occiput, both of which are thereby flattened, while the skull is 
rendered proportionally broader and longer than it would naturally have been " (Macgillivray, 
II. p. 12). 

In a paper entitled " Cranial deformation of new-bom children at the Island of Mabiak, 
and other islands of Torres Straits, and of women of the S.E. Peninsula of New Guinea^ " 
(Proc. Linn. Soc, New SouUi, Wales, vi., 1882, p. 627) Baron N. de Miklouho-Maclay writes : 

" In April, 1880, visiting the islands of Torres Straits, I had the opportunity of seeing, 
at Mabiak, an interesting operation performed on the heads of new-bom children. During 

* The latter part ot this sliort paper reten to s tnuiETerse depreMion, 9. little behind the sntnia ooronaliB, 
in Oie eknllB ot the women, vhiah ia due to the prautioe of oftnyiDg 'hiea^rj buideiu in large bags, the handles 
of which are saspended from the orowD of the head. D'Albertii noticed something limilar amongst the women 

d by Google 


the first weeks after the birth of the child the mothere are accustomed to spend many hours 
of the day compressing the heads of their in&nts in a certain direction, with the object of 
giving them a quite conical shape. I have seen it performed daily and on many children, and 
have convinced myself that the deformation, which is perceivable in the adults, is the result 
of this manual deformation only. This observation was specially interesting to me, remem- 
bering having read, many years before, the opinion of the celebrated biologist and anthro- 
pologist K. E, de Baer, member of the Imperial Academy of Sciences of St Petersburg, who 
would not believe that a manual pressure could have such an effect on the skull (vide 
K. E. de Baer, ' Ueber Papuas und Alfiiren,' Mimoirea de I'Acad. Imp. dee Sciences de 
St Petersburg, 6 s^rie, t. viii., 1859, p. 331). K. E, de Baer expresses the opinion, analysing 
the information given by J. Macgillivray [see above], that the observations of Macgillivray, 
who has seen the same above-mentioned manual deformation performed on children at Cape 
York, are not exact enough. Remembering this contradiction, I was careful to decide the 
contested point, and now, after careful examination, measurements and inquiries, I believe 
the question may be regarded as settled, and that the information given by Macgillivray about 
the head deformation at Cape York was not too hasty, and was correct. As far as I know, it 
will be the only well authenticated example of cranial deformation by means of Tnanual 

Dr A. B. Meyer, in his monograph entitled " Ueber kilnstlich deformirte Schadel von 
B<5meo und Mindanao im kijnigl. anthrop. Mus. zu Dresden nebst Bemerkungen Uber die 
Verbreitung der Sitte der kiinstlichen Schiideldeformirung " (QrattUaiionsackriJi an Rudolf 
Virckow, Leipzig, 1881), refers to cranial deformation being common in New Caledonia, and 
Malekula ; it is also very frequent in northern New Guinea (Geelvink Bay, Waigeli, Rawak 
and Boni), and deformed skulls have come from other islands in this part of the world. 
Reference is also made to this custom in Crania ethnica (1877, p. 207) where de Quatrefages 
and Hamy describe a deformed skull of a woman from Toud (Tutu) which is figured in 
figs. 220, 221 ; the same authors give a wood-cut (fig. 222) of a cast of a Tutu native's 
bead in profile in which the antero-posterior fiattening is well shewn. This deformation 
appears to occur to a variable extent in the skulls {mm that island examined by these French 

Nose anu Ear piercixg. 

A man with a small nose, magi mawa, is not regarded in Mabuiag as being handsome, 
but a prominent nose, km mawa maiiu, is greatly admired. Dr Seligmann was informed that 
if the child's nose has been flattened during birth, the midwife gently presses it into shape 
with her teeth. 

The artificial deformation of the nose is confined to two kinds of perforations, 

(i) By far the most general, and at one time probably universal, is the piercing of the 
nasal septum <pl. I. fig. 4 ; pi. U. figs. 1, 2 ; pi. V. fig. 9), but this custom is now dying out. It 
was done merely for decorative purposes, in order that a nose-stick, gub, occasionally called 
jftgfiifc (W.), kirkub (E.) might be inserted (pi. VIII.). 

I was informed in Mabuiag that the piercing was done when the child could crawl. 
Dr Seligmann was told that it was when the child first smiled. Macgillivray (ii. p. 12) says 

H. 7ol. IT. • 

d by Google 


that in Muralug the nose waa pierced when the infant was about a fortnight old. We were told 
in Mabuiag that the child was placed lying &ce uppermost on the back of one of his wadwatn 
(v. p. 147), and various people held down his head, arms and legs. Any man (but not a 
woman) might make the hole, and for this a turtle-shell bodkin was generally used. First 
a stem of unu grass, bok, or of kawipa, was inserted in the orifice, and later a grass with 
a thicker stem. In order to prevent interference on the part of the infant with the proper 
healing of the hole, its hands were tied with a slack soft cord to the legs just above the 

In Mer the nose is pierced a few days or weeks after birth ; it has nothing to do with 

(ii) For information respecting the second kind of nose perforation in Mer, I am 
indebted to Mr J. Bruce. Two small holes, pit n«6, may be bored in the tip of the nose 
of youths at about the age of fifteen. They have no significance, but are merely used as 
a means of decoration. The holes are bored right through into the nostrils with a very fine 
pointed piece of hard wood, and pieces of the midrib of the coco-nut palm lea^ be lid, are 
inserted in the holes until the wounds are healed. I was informed in Mabuiag that formerly 
the men of the Tabu clan had two small holes bored in the tip of the nose which were 
evidently intended to represent the nostrils of the snake. If the Miriam in their old totemic 
days followed a similar custom, assuming them also to have had a snake clan, it is only to be 
expected that on the lapse of totemisin the custom would have no significance and its origin 
would be forgotten. 

The bodkins or awls, ter or luper (E.), used for piercing the nasal septum of in&nts, are 
made of turtle-shell. They vary considerably in length, the average sizes are from about 
24—28 cm. (91~11 ins.). Figs. 11 to 17 of pi. XI. sufficiently illustrate the variation in form 
of these implements ; they may be plain or decorated with simple patterns of incised lines or 
of fine wavy lines. 

The favourite method of treating the ears is to produce a fleshy pendant', muli (W.), 
laip aak (E.), at its lower end (pi. I. figs. 1, 6; pi. II. figs. 2. 4, 6; pL HL fig. 2; pi. V. fig. 9), 
and to puncture the margin of the helix, but these perforations nearly always become torn, 
so that the rim of the helix becomes notched (pi. I. fig. 5 ; pi. II, figs. 1, 4. 6). Among the 
younger people the deformation of the ear is decreasingly practised. The lobe is rarely cut 
without previous elongation (as in pi. V. figs. 6, 8), but a small perforation may be made 
for an earring. The margin is not now perforated. 

In Mabuiag the lobes of the ears are pierced at the same time and in the same manner 
as the septum of the nose. One informant said that the implement used was a needle-like 
wing bone of a flying fox', but doubtless the ter was more generally employed or even 
a sharp-pointed piece of wood. An avbau leaf was greased with turtle or dugong fat, 
and the roUed-up leaf inserted in the greased orifice. I^ater a needle of uraka wood 
(Hibiscus tiliaceus) was inserted, as it has a saponaceous Juice. 

When the hole, nahattaizinffa or kaura terta, was healed a pendent weight of vhar wood 

I '• The lower lobe of the e«r is slit, ftnd faangB very low, some being three inebea long," Mer, Nam. Mag. 
•n. p. 788. 

■ SapM kiaau ia the name tot the implement, but the name of the bone before it ii removed &om the bftt 
(Pleropos) i« lapur fat. 

d by Google 



(MimuBops browniana) wae suspended in the lobe in order to distend it The process of 
distension must have extended over a number of years as the weights employed for this 
purpose are of considerable size and the process was probably not completed till puberty. 
Sometimes two weights were used in one ear. When the lobe was fully distended, a ligature 
of strong fibre, muti, was wound round the part of the loop nearest the &ce, and fresh 
ligatures were Bubsequently tightly tied. I was informed that "the fibre eats" {mvU 
kaikai), the end of the lobe, which swells up and festera Eventually there was only a 
small strand of flesh left which was cut with a bamboo knife. The free cut end of the lobe 
was then rolled up in leaves which were made fast with a piece of fibre till the wound was 

The small holes, kaura tira, round the margin of the ear were all made at the same time 
for one ear, and when that one was well the other was similarly perforated. I was infonned 
that the marginal holes were pierced later than the lobe. Dr Seligmann was told that 
when the child had become tolerant of the gub (p. 9) a series of holes was made round the 
periphery of the helix, and that commonly the lobule was not bored till the child was eight 
or ten years old. 

In Mer, when the child was quite young (one informant said six days old), the lobe of 
the ear was pierced by means of a thin sharp-pointed awl, tol (fig. 1 a), of hard wood, or with 
a ter (p. 10). Then grasses of increasing calibre q 

were inserted as the hole, Imp ne&, became 
larger, till it was of sufficient size to receive a 
blunt, narrow cone of wood, laip tut, which 
was the name also given to the dumb-bell- 
shaped weights used for Che distension of the 
lobe (fig. 1). 

We have definite information that the 
Uiriam sever the end of the distended lobe 
where it joins the face as a sign of mourning, 
and probably this was also the case for the 
Western Islanders. The operation and the 
reasons for it are described in Vol, vi. p. 154. 
Mr Bruce says : " The ear piercing is a separate 
ceremony performed on them at an earlier age, 
but not in connection with death." 

The ear-weights, vbar (W.), laip tut (E.), 
vary so much in form that only a few charac- 
teristic examples can be figured (figs. 1 — 3 : pi. 
X. figs. 1 — 4, 13). Other illustrations will be 
found in the Album, I. pi. 334, Nos. 6, 7 ; 335, 
Na 12. 

All the good specimens are made of a hard heavy wood ; I believe that of the Mimusops 
browniana tree, vbar (W.), wagai (E.), was the best for this purpose, but I have occasionally 
heard them called vbar in Mer. They consist essentially of two lobes connected by a narrow 
bar ; the latter may be straight or bent, and the lobes vary -extremely in size, weight, form and 


Flo. 1. ImplementsfordUleDdingMirB. A, 7oI, wooden 
awl for pieroing the lobe of the ear, Her, 83 mm. 
long. B, Laip tvt, cooe of light wood, probably 
HibUons, toi dilating the hole, Mei, 6fi mm. long. 

C, Vbar, ear-weight, Maboiag, 80 om. wide, on 
the pair weishe 7^ oz., and the other 6} 

D, Laip tut, ear-weight, Mer. 67 mm. wide, weight 
a little leu than 1 oz. E, 65—70 mm. wide, weigh! 
nearly I oz. 

d by Google 


workmanship. To take two extreme types, each lobe of the ear-weight shewn in fig. 1 c, 
is a thick flattened triangle ; while that of fig. 3 is plum-8h^>e(I and very neatly worked. 
Some are quite plain, others, as figs. lE. 2, are slightly carved with simple designs; occasionally 
a face is carved on the upper sur&ce of the lobes. The weight may be as low as half an 
ounce, but from 1 to 2 oz. is the average weight; one specimen in the Gla^ow Museum 
weighs 4^ oz., another 9^ oz., and one in the Liverpool Museum ( 11 oz. ; the 
last has a height of 11 cm. and a width of 175 mm. (pi. IX. fig. 13). 

Fill. S. Nntl; earred ew-»eight, protMbly 
I bom Uer; width KMnmi., weigbt 3Jos. 

Fro. 3. Eu-weight, la^ tut. Her. 10x11 am.; 
wdght 8^01. (A. 0. H. Col.). 

Perfectly similar ear- weights are employed at the mouth of the Fly river ; in a MS. 
catalogue, Chalmers calls them ouou (? oriou), and says that they are worn by boys and girls 
and mode by their fathers. Some of these are figured in the Album, II. pi. 190, figs. I — 4^ 

Other forms of deformation. 

With the exception of the treatment of the head and of the nose and ears described 
above, no other deformation of the head was practised ; front teeth were not knocked out 
as is so commonly done in North Queensland. No alteration was made in the contour 
of trunk or limbs, nor were they treated in any way, except by scarification. 

I did not personally hear of a case of circumcision nor of any other mutilation of the 
sexual organs of either sex. Mr J. Bruce has, however, given me the following information 
concerning the Miriam, and possil:>ly it may apply to other Islanders. 

" The males practise circumcision in some cases, when they arrive at the age of puberty, 
the foreskin being cut with a piece of glass, but the custom is by no means universal. 
Women when quarrelling with men sometimes taunt them by saying that their penis 
is dirty under the foreskin, putpe eb. This is regarded as a great insult by the men, 
and all those present hang down their heads in shame when one of their number is 
accused thus by a woman. One case of this sort has been brought into court, and on 
another occasion the parties were advised to apologise and be friends ; but no doubt 
formerly the woman who taunted a man in that manner would suffer all the pains and 
penalties that could be brought to bear upon her. Possibly the fear of this disgrace may 
induce young men to circumcise in order to keep clean, but I have never heard that 
reason given." 

d by Google 



Being a dark-skinoed people, scarification was employed for the permaDeot omamen- 
tation of the skin, and so far as oar evidence goes tattooing was formerly unknown. At 
the present time a number of the younger men have casual tattoo markings, but this 
practice has been copied from Polynesians and others, numbers of whom have frequented 
Torres Straits for the last fifty years. 

Scars were made in many cases only deep enough to produce a cicatrix, but when 
raised cicatrices were desired other means were employed. According to MacgilUvray 
<Vol. ti. p. 13) the latter were formed in Muralug by cutting the skin «-ith a piece of 
glass, then a chewed leaf of a certain plant was introduced into the wound to prevent 
the edges frvm uniting, and a daub of wet clay was placed over all and kept there until 
the necessary effect had been produced, Dr Seligmann was informed in Mabuiag that 
the patterns of the deeper scars were traced in red ochre, then, when possible, ligatures 
were applied above and below the marking, and the pattern was cut with a sharp 
fragment of quartz or sbelL The juice of ffruat (Sesuvium portulacastrum) was instilled 
into the wound to prevent healing by first intention and to produce suppuration, the 
amount of which ensured a more or leas raised scar'. According to Dr Seligmann, small, 
-scarcely noticeable, circular actas resembling those left; by old sores were coounon on the 
fiexor surface of the forearms of the Saibai people ; they were produced by a moza made 
■of a small splinter of the midrib of a coco-nut palm leaflet, which was stuck into the 
skin. Scars somewhat similar in appearance, but less durable, were effected in Mabuiag 
by rapidly rotating between the thumb and forefinger a length of the midrib of a coco-nut 
leaflet, one end of which was pressed firmly against the skin. 

Mr J. Bruce informed me that in Mer the design chosen was drawn upon the skin 
in red paint (tnaier), and was then incised by means of a small shell (kaip, or tu). Those 
who were expert in the process were called koivvi totiar le. Women generally operated 
-on the girls, and men on the youths. Whilst the incisions were being made the young 
people were held down by their elders, as they often took fiight and tried to run away ; 
sometimes they binted under the ordeal. The cutting of several of the koima designs 
was especially painful, particularly of those on the shoulder and breast. The cuts healed 
slowly, and until the healing was complete the patients were exempt from labour and 
wor« their arms in a sling. In cutting the koitna shewn in fig. S3 a piece of skin was 
excised so that a large wale might result. I have heard that large scars were cut with 
a bamboa 

In the following account of the practically extinct custom of scarification of the Torres 
Straits Islanders I have not hesitated to give illustrations of some of the cicatrices that 
occur in Daudai (the most convenient term for the neighbouring coast of New Guinea), 
in the adjacent islands, and in Kiwai island at the mouth of the Fly river. The reason 
for including these districts is that their inhabitants have an intimate relationship 
with the Torres Straits Islanders, and when the former are thoroughly investigated we 
may expect to have light thrown upon practices which are no longer observed among 
the hitter. 

> MurfuUne (dmmg the Camiibalt, etc., 1888, p. 130) aaya that the tribM of the P»piiaii Gall make theee M»r« 
bj " oatting and inHrUng into the woniid powdered Bhell, which givM it when healed a awollen, rib-Uke an>earaiioe." 

d by Google 


Some of the designs I collected tnyself, but others are taken from a long series of 
tracings which Ur R. Bruce kindly seoC me in 1894, 

Cicatrices were made for various purposes, bat they must aot be confounded with 
the scars that were incidental to scarifications made for theropeuCic purposes, which 
were usually irregular scratches or short simple cuts incised on the forehead, back or 
elsewhere. The objects for which definite cicatrices were made may be classed as (1) social, 
(2) commemorative, (3) moumiiig, (4) magical, (5) therapeutic, and (6) decorative. 

1. Social. Among the totemic Western Islanders some of the cicatrices represented 
the totem of the individual, either by a realistic figure or by a conventional design, as 
for example the kibu minar of the women (p. 21), and other marks on men and women 
to which reference will be made in the appropriate places (c£ also Vol. V. pp. 158, 159, 
163 — 169). Owing to the absence of totemism among the Miriam (vi. p 254) these designs 
do not occur in Mer. 

2. ConunemoratlTe. The baga minar (p. 15), simu minar (p. 16), and the kilm 
minar were said to be cut when the girls reached puberty and consequently they would 
indicate nubihty, as did thigh marks (fig. 38). The piH tonar (p. 26) was a record of 
the length of the nose of a deceased brother; it may also have been a sign of mourning. 

3. Houmlng. As mentioned in Vol. vi. p. 154, the young Miriam adults had 
cicatrices cut as a token of mourning for a parent or near relative. The blood flowing 
fix>m the koima wounds cut on the back, and, according to some, from those cut on the 
breast, was allowed to drop over the corpse. The first koima design (figs. 19, 40 d) was 
of a simpler character than those which were made on subsequent occasions of mourning ; 
it was cut on the calf, forearm, or loins. A tew years later, in the event of another death, 
this lesser koima {kebi makerem a kM neurra koima, youth and small girl's koima) was 
succeeded by other larger koima, au koijna, which differed in form according to the sex 
of the mourner. Figs. 13, 15, 33 represent those cut on young women ; the loin and 
cheek designs were cut only at the death of a rehitive. 

4. magical. The only one direct statement concerning the " magical " significance 
of scarification is given on p 28, but doubtless there were other instances. Possibly 
the cutting of centipedes on the legs of Daudai women (fig. 42) and similar scars on 
Miriam women (fig. 15) were charms to prevent their being bitten by these creatures; 
see also description of fig. 6. 

5. Therapeutic. Besides casual scratchings, cuts were fi^uently made in a determinate 
manner to relieve pain and as a counter-irritant in various cases of sickness. Instances of 
this are noted on pp. 20, 21. 

6. DecoraUve. The koimai or koima of the men appear to have been purely ornamental, 
and the breast koima (fig. 13) and sometimes the large shoulder koima (fig. 33) might be 
cut on Miriam young women merely for purposes of decoration, and not only as a sign 
of mourning. But for whatever object the scars were made there is no doubt that they 
were considered as ornamental and that the people took great pride in them. I do 
not know whether it is really significant, but it is worth noting, that the word koimai (W.) 
may be derived from koi, great, mai, mourning or grief; if so it may have originated as 
a token of mourning, whatever its later development may hare been. 

It is stated on p. 16, that in the western islands and New Guinea the breast scarification 
was made to prevent the breast from becoming too pendulous and in order that the women 

d by Google 


might look " flash." I have heard that thia is done by the North Queensland aborigines 
for the same purpose; the young Miriam women however have theirs cut as a sign of 
mourning or merely as a means of decoration. Mr J. Bnice definitely asserts that the 
Miriam women did not have bars cut across the breasts to assist in keeping them up; 
they rather gloried in their extension, as it made the design on them more effective by 
displaying it to greater advantage. 

In former days the Western women were frequently ornamented with two main 
kinds of cicatrices, the vtat and the mt'nar. At Saibai scars corresponding to the 
Mabuiag ueal are called wvz. Ua or wux (Saibai) is also the name for quartz, and the 
scars received their name from the fragments of quartz with which they were cut; in 
Saibai the same name, according to Dr Seligmann, was applied to the irregular scars 
produced by medicinal scarification. The ordinary laal were small simple cuts, whereas 
the word minar signifies a pattern or mark. 

Scarifications were called uaer and koima in Mer, here the word us means a thin, 
sharp shell used for carving ; perhaps the shell In the east replaced the quartz of the west. 

The two/ consist of a series of small, generally parallel scars often covering a considerable 
area of the face and breast. The commonest forms are a series of small, closely set, linear 
scars, about one centimetre long and roughly parallel with each other except where they 
follow the curve of a bone or the outline of some prominent feature. They are produced 
after puberty and may be added to indefinitely, and they are often arranged symmetrically 
on the two sides of the body. Barely a double row of ttaal may be arranged convergently 
\/, the pattern then consists of a number of such marks in series. The names given 
to the various lines of usal are derived from the part of the body which they decorate. 

8caxlflcatlon« on the fiice. 

The following fecial scarifications occurred among the Western women. The para 
tisal, forehead usal, starts above the root of the nose, crosses the 
forehead above and roughly parallel with the eyebrow, and descends 
behind the outer margin of the orbit to terminate at a point 
behind the most prominent part of the malar bone. The bof/a 
minar, or mausa (or masa) uso/' extends firom the angle of the 
mouth, up the cheek and round the cheek-bone. The baga minar 
of Kausa of Mabuiag (pi. 11. fig. 6) consists of a series of short, 
paired, vertical lines. At Mabuiag a modification of the baga minar, 
called bidil (flies), was met with. It consists of a number of inverted 
V-shaped marks arranged in series. These commonly run from the 
moat prominent part of the cheek-bone vertically downwEtrds as iar as, 

or a little lower than, the level of the anele of the mouth. According ^"*- *■ Saa\&<Mioa on 
■ , .. T» Ti ■ 1 ■ r. -v ■ /o /; I"" f^** "f s Baibai 

to information given me by Mr K. Bruce, a girl in baioai (ng. 4) giri ; the smm ue 8 

is decorated with two leaf-tike markings, danamon, on the left cheek em. mnd 46 em. long; 

Q^Jy_ ftfter B. Brooe. 

■ Baga ii "chMk"; I do not know the meuiiDg ol nuaua, perhtpi the term hu been introduoed bom New 

d by Google 



The koitna on the face, neurra bag war or neurra tote chib (girl's cheek pattern or 
girl's 8C€ir, dtA, that resembles the feathers of a certain bird tote, perhaps a snipe), was- 
cut as a sign of mourning only on young Miriam women, or on 
newly married women if it had not been cut previously to their 
marriage, which, however, was rarely the case. The design 
shewn in fig. 6 was generally cut on both sides of the foce. 
The upper curve followed the line of the cheek-bone about 
25 mm. below the orbit, starting about 13 mm. from the nose. 
One end of the lower curve was close to the nostril. The size 
of the pattern was approximately 3 cm. square. There was a 
star, wer, in the centre. One simple face koima was a cut 
which started about 13 mm. from the nose and proceeded back- 
wards along the cheek-bone, curving round it and running 
forwEuiis again to terminate about 13 mm. from £he mouth. Another variety, extended 
from the centre of the ear to the most prominent point of the cheek-bone, when it curved 
downwards and backwards past the comer of che mouth, ending at the angle of the 
jaw ; or, according to another account, it passed forwards and downwards past the mouth 
to the chin. It is stated in the Nautical Magazine (vi. 1837, p. 763) that both sexes 
in Mer " have a figure resembling a banana tree, or a cocoa-nut tree, on each side of 
the head." 

Fid. 6. Neirra tolt dup, e 
Buse, after J. Brao«, 

Bcorlflcatioiu on the Body. 

The susu usal of the Western women consists of one, two, or more parallel lines of 
vertically arranged usal stretching horizontally or ob- 
liquely aross each mamma (fig. 11). The mitav, usal 
consists of two lines of convergent scars, which, beginning 
above at the fold where the mamma joins the chest wall, 
in a line corresponding pretty nearly with the outer edge 
of the rectus muscle, run down in this line to curve 
inwards and meet in the middle line at a point about 
two inches below the umbilicus. A single or double 
row of scars, zugu uaal, may stretch from high up on 
the anterior axillary fold across the swell of the deltoid 
muscle on to the back of the arm. Doubtless other vari- 
ations occurred, as for example the markings on Fauna, 
wife of Maino of Tutu (fig, 6), herself originally a native 
of Mawata, New Guinea ; those usal forming an oblong patch on the breast were called 
sedau or aedor, and those forming a loop on chest and arm were called eheri, centipede; 
perhaps these represent the track of a centipede. 

The A'Shaped breast scarifications of the women (sutu minar, breast mark (W.), nana 
user, breast scarification or nana dvb koima, breast scar koima (£■)), were probably frequent 
everywhere, as they still are on the neighbouring coast of New Guinea and the adjoining 
islands. I was informed in the western islands that these scarifications were made when 
the subjects were girls for the purpose of holding up the breasts so that they should not- 

Fta. 6. 

d by Google 


become pendulous'. Mr R. Bruce informs me that in Saibai, Daudai and the Fly river 
district all the women who are scarified in this way have short breasts and not long 
pendulouti breasts like other women. They told him that they were scarified when they 
were young so as to look " flash " (smart) and also in order that when they bad suckled 
children their breasts would not hang down. 

A good example of a sttau minar is seen in pi. III. fig. 2, which is drawn from 
a tracing made by Mr R. Bruce of the scarification, susu guda-mulam, on Bonel of Saibai. 
Mr Bruce adds that the flesh of this scarification is raised up in a round, smooth wale 
as in the case of the Australian blacks. 

When' I was at Saibai I enquired about this particular scar, and was informed that 
it was a kabu minar, chest mark, that guda-mulam (mouth open) referred to the space 
between the upper limbs of the marking, and that the design represented a dXri, or 
headdress of white feathers (pi. VL). Its extreme breadth measures 
11-5 cm. This scarification is very similar to that indicated by Melville 
in the standing girl of pi. III. fig. 3 ; the sitting woman to left in 
the same group appears to have breast scarifications. The scarification 
of Kurubat of Tutu (fig. 7) was similarly described as a dJfri. I am 
unable to say in either case whether these markings were intended 
to represent a dJfri, or whether they were so called because the 
traditional scar bore some resemblance to the headdress. 

The two wooden images of girls described in Vol. vi. p. 222 are* 
incised with breast scars (pi. III. fig. 4 and fig. 20 a, d). One (A) was obtained at Mer, 
and the other (b) at Erub, though it was stated to have come from Maeig: they were 
employed in erotic magic. On plate XVII. of his Sketches, Melville figures two women 
with a similar design, one was a native of Erub, the other came from Daudai (pi. III. fig. 3). 

Somewhat similar is the cheat scarification on Tag (fig. 8), the widow of Matau of 
Mer (5), herself a native of Parama, an island off Daudai That of Nisoma of Mawata 
(fig. 9), as might be expected, is more like the diri design of the western islands. The 
chest scarification (fig. 10) of Magina of Saguane, Kiwai island, is not very unlike that 
of a woman (fig. 11) whom I sketched at lasa, in the same island; it is called amo iwi, 
breast scar. 

Of a somewhat different character are the breast scarifications of two Saibai women, 
Kaubi and Maipi (fig. 12), tracings of which were sent to me by Mr R. Bruce. B^ubi 
however originally came from Sui, at the mouth of the Fly river. Mr Bruce calls 
these scars sedatia or aedau, which is the general name for a mark. 

The large koitna marks on the Miriam girls' breasts (ati neurra nana dnb koima, big 
girl's breast scar koima; or nano user, breast scarification) were of various designs. One 
of them (fig. 13) was usually cut at the age of sixteen or eighteen years. It consisted of 
two upper scrolls incised upon the sternum at the level of the upper margin of the 

' Dr Beligmum found that unoiig the Olati, an AaBtralian tribe on the eut coast of the Cape York peuinBiiIa, 
a lets elaborate aear wai made with this object, and at Old MawaU, in Daud^, we uw a nninbei of women 
with loan which were better oalenlated to attain thia end than thoM in tue in Torrei StraiU. Two broftd 
doatrieial bands itarting above the right and left breasts, oroeaed eaoh other, and ran obliquely downwards and 
OQtwarda on the left and right breaatl reapeotiralf. Fig. 10 iUnatratee a nearl; nmilar airangement. 

B. Toi. IV. a 

d by Google 


Fro. Q. Sokrifloation of NiaonM of Ifamte; obUqaa 
Langth ot ebest mm, lS-0 em. ; aie*Matio aoM, t bj 
3'fiani.; length of arm loai, 16om.; altwK. Brnea. 

Fio.lO. SoariSoationofUBgiDaot Sagnane; obliqoe length 
of oheat lou, IS-S om. ; total length of bieaat mm, 14 em. ; 

d by Google 


Fio. 13. Bnwt iMjifioktioDB of Kanbi {k), aDd Hupi (B), Pia. IS. Au a«urra mko dub koima; 

both of Baibki; ) lut. nie; ftfter B. Bmoe. ftfUr J. BniOA. 

Fio. 1*. Souiaofttioii of Eaabi of Sai, DKodai; breut mu, 11 by 7-3 am.; 
um aoar, IS-S bj 6 am.; after B. Bmoe. 

d by Google 



breasts. From these scrolls a fringed pattern diverged to cover the inner side of the 
two breasts. This fringe was called nana dubara pern, the fingers of the breast scar. 
The inner markings were solely decorative, no teir. As the breasts grew larger and 
pendulous, the koima increased correspondingly in size (a nano dub ekaselt). 

Attention should be called to the very broad cicatrices at the junction of the mammae 
in fig. 9, or somewhat higher up in fig. 18. This kind of scar is rather common in Daudai, 
where it, as well as the yy-shaped chest scar, is called ddra; this is probably the same 
word as dSri, or it may be the da of Boigu and Saibai, or the dUra of Muralug, both of 
which mean breast or mamma. 

Another variety of breast scar, aedau, is seen in Eaubi of Sui', now living in Saibai 
(fig. 14), and in Magina of Saguane (fig. 10). 

The three rows of small scars on the left breast of a Elwai woman (fig. 11) were 
cut on au occasion when she was sick ; they therefore have a therapeutic significance. 
I do not know whether those of Fauna (fig. 6) belong to the same categoty. 

Another Miriam breast koima represented the backbone and side-bones of the ubar 
fish, a kind of sole. ,a, 

A centipede, iai, was a bvourite design, cut either on each breast, nano 
iti, or on the upper arm near the shoulder, tugar in. It was cut only on 
women. The centipede was represented crawling upwards (fig. 16). It was 
about 12o mm. in length. The islanders distinguished A, the " teeth," tereg ; 
B, the " curling hair," pis mu« ; c, the " head," iw«n ; D, the "feet" or "hands," 
teter or tag; E, the "extensile body," buber gem; and p, the "tail teeth," upi 
tereg, as they suppose the centipede to bite with its tail. No reason could be 
discovered for the adoption of this design other than that it was effective 
and easy to draw. 

These three Miriam designs were signs of mourning, though the first 
might also be made for the sole purpose of decoration. 

There is no evidence that abdominal scarifications were at all frequent 
in Torres Straits, although they are common in the neighbouring districts 
of New Guinea. The abdomen of Abaka of Boigu (fig. 16) is ornamented 
with scars {fig. 17) which I was informed represent the scutes of a crocodile's 
tail ; they were called pata minar, cut mark. Dr Seligmann was told that 
they are called kodalau tar, crocodile's claws. Mr Bruce was told that the inni. mis; 
scars represent " the leaves of a water-lily that grows in the fi^sh-water lakes Attar J. 
of Boigu." B™»- 

Transverse parallel cuts on each side of the umbilicus occur on the love charm from 
Masig (pL III. fig. 4 B), and a raised star surrounds the umbihcus of a stone effigy of a 
woman on Dauar (vi. pi. V. fig. 3). A bopuro iwi, navel scar, is shewn in fig. 11. 

The extreme case of abdominal scarification known to me from this region is that 
of Babaura of Fomogora, Daudai (fig. 18), for which I am indebted to Mr B. Bruce. 

So &r as I am aware, the upper part of the back was only occasionally scarified 
and then only with the small usal scara Dr Sehgmann noted in Saibai Icotami vnu. 

Pio. IS. 

■ Tboia ippnr, tiom Mr B. Bmce'a nol/.a, to haTe beSD t 

ibi liTing ID Skibfti. 

d by Google 


Bcapula scars, consisting of two bilaterally aymmetrical series of cicatrices which, starting 
CD each side &oin low down on the posterior azillaiy fold, skirt or cross the inferior 


Fia. 16. Ab>k» of Boigv ; kttar B. Bnuw. 

angle of the scapula, whence they run up the vertebral border of that bone to end opposite 

its spine or superior angle. I noticed that over the 

upper border of each scapula of the Kiwai woman 

shewn in fig. 11 there were three horizontal rows 

of small vertical scars (resembling those on her left 

breast and cut for the same purpose); there were 

also two horizontal rows below the right scapula. 

Probably all these usal scars are therapeutic. 

The jti6u tninar, loin mark, was perhaps the 
most important scarification of the western women. 
During my expedition in 1888 — 89 I saw only four 
iifru minar, and as they were on elderly women they 
were not very distinct ; repeated enquiries failed to 
elicit other examples. Owing to the present custom 
of wearing calico gowns the marks are not ordinarily 
visible, but in former days they would readily be seen 
above the waistband of the petticoat. Patagam (5 a) 
and Wagud (5) both belong to the Toiu, Dangal 
(snake, dugong) clan of Mabuiag, but the former has 
the Taim augud (totem) on her back (pi. IV. fig. 2), 
while the latter has the Dangal (pi. III. tig. 4). Wagud 
married a Tutu man and was living in that island 
in 1888. Ado of Badu has a Dangal cut on her 

Fia. 16. SoariflMtion ol BalwiiTa of Porno- 
gon, Dftudai ; Dppermoit ohett bou, 9 em. 
by IT mm. ; iftsr B. Brnoe. 

d by Google 



back (pL IV. fig. 3). There waa considerable difficulty in determining the eigmficance of 
the augud cut on MSke of Tutu (pi. IV. Sg. 1). It was said to represent the serrated 
spine of a sting-ray, the tail of a crocodile, or the mark of a centipede, but Maino, 
the chief of Tutu, said that she had Baiheaam, which belonged to Sigai (or Kurai), as 
her chief augud and Warn (turtle) as her little augud ; the cicatrice would thus represent 
the haihemm, a cresceutic object decorated with cassowary feathers, which had a magical 
significance (v. p. 374, c€ pp. 371, 373). Thus all the kibu minor known to me were 
representations of totems, and probably in former days it was a frequent practice to 
have totemic designs scarified on the loin in this way. I was told at Mabuiag that 
this was done by the women of most of the clans (v. pp. 163 — 16$). 

A corresponding scarification in Mer is the kip aor koima, back koiina, or kip user, 
back scarification, which was cut on a girl's back above the level of the belt as a sign 
of mourning (fig. 19). The two scrolls are called pis mus (a term for anything that 
curls over), the fringe round the design is called koimara pem or koimara war, marks 
of the koima ; Mr J. Bruce says that they are spoken of as " the fingers of the koima." 


Fio. 19. YoQtb and hdikII giTl'a 
ioima ; about i a 
J. BnlW. 

i. 8iM ; altsr 

Fta. 30. BabbiDg of part ot th« 
mouth o( a gama dnim, Mer ; 
i Dst. Bize. 

Fia. 21. SearifloatioDB of the wooden 
imagea ot giilB (pi. III. Bg. 4). 
A, B, Breait and loin marks on A; 
C, D, Shoulder and breaat marki of 

A. coco-nut water vessel in the Homiman Museum, Forest Hill, London, has engraved 
on it the same design (pi. III. fig. 1), doubtless it came &om Mer; and it occurs also 
on a drum of the Kiwai gama type, with three jaws, made by Baton of Las, Mer (fig. 20). 

The wooden figure of a girl from Mer has two somewhat similar marks cut on 
the loins (fig. 21 B). 

Boaiifloatlons on the Amu. 

The shoulder was the part of the body that was most frequently decorated by 
scarification, especially in the case of men. This cicatrice has been noted by most 
voyagers to Torres Straits. Dumont-D'Urville (Voy. au Pole sud, IX. 1846, p. 236) says: 
"They [the natives of Tutu] execute a tattooing in relief, which marks their shoulders 

d by Google 


with Reahy pads or swellings [bourrelete chamus] arranged like the fringe of an epaulette." 
Macgillivray says: "The Torrea Strait Islanders are distinguished by a lai^ complicated 
oval scar, only slightly raised, and of neat construction. This, which I have been told 
has Bome connection with a turtle, occupies the right shoulder, and is occasionally repeated 
on the left" (Macgillivray, II. 13). It is probable that a young man was not allowed 
to bear a cicatrice until he had killed his first turtle or dugoi^. Jukes gives figures 
of three men, in which it is very indistinctly seen; these are "Mammoos" of Masig 
(I. p. 159) [which I have reproduced in pi. I. fig. 3], "Old Duppa" of Mer (ii, p. 236) and 
"Manoo" of Erub (ii. p. 237). These are abo reproduced in Sketches in Auatralia and 
the adjacent Islands {pis. XVL and XX.), by Harden S. Melville, the artist who waa 
on board the Fly; the former I have copied in pi. XXIV. fig. 2. Gill (Life in the 
Southern Isles) also gives a sketch (p. 241), and states that a "symmetrical scar is made 
on the shoulder of all males in Mauat [Mawata] and in the Straits." Dr Gill's sketch 
is veiy similar to fig. 31 A. In a small book by W. R Brockett, entitled Narrative of 
a Voyage from Sydney to Torres Straits, five sketches of " marks cut on the natives' 
shoulders" are given oa his pi. II. All the illustrations in the pamphlet are very 
rough, BO that too much stress must not be laid on their accuracy, but as the work 
is rare I reproduced the figures in my original paper {Joam. Ant/i. Inst. xix. 1890, 
p. 366, pi. VII. figs. 1—6). 

The koimai (W.), koima (E.), as this scar is called, was cut either on one or on 
both shoulders. I gathered in 1888 that among the Western Islanders its presence, 
either single or double, or its absence, had no special significance, and I was informed 
that if a man had a fine shoulder and wanted to look "fiash" he would have it cut 
on one or both shoulders. Some said it was cut on "big men." In Mer it was more 
particularly called kimiarra tugar koima (man's shoulder koima), and was described as 
"kab koima teir" (dance koima decoration), "note bvd kak" (not mourning at all). It 
was cut on young men, generally on one shoulder only as few could bear the pain 
of having both done. The young women were greatly captivated by this device, and 
considered the man who had it as "au tetor le" (a "proper fiash man"). On festive 
occasions they painted it red or white (d«6e teirem, well decorated). This koima was 
not confined to the Beizam le (vi. p. 169), although it was characteristically their 
decoration. Young women did not wear this device, as they had their own (p. 25). 
In the NavticcU Magazine (vi, 1837, p. 763) we are told that the Miriam "do not 
scarify the body so much as the New Hollanders do, but the men generally have a 
scarred figure representing a shell on each shoulder, and the women are marked with 
the same figure on the breasts." 

I have quite &iled to discover the meaning of the design itself I was informed 
in 1888 that it represents the coils of the intestines of the hirmiu fish, but this is 
so improbable that it may be ignored; that the design looks like intestines is quite 
another matter. As similar designs occur on the men of the neighbouring coasts of 
New Guinea the explanation will veiy likely be found there. 

Even in 1888 not a single man of the Western Islanders, so far as I could learu, 
had a koimai, although I made repeated enquiries about it on every possible occasicm, 
and in Mer only three old men had a koima. 

d by Google 



The general character of the shoulder scars can hest he judged from the accompanying 
illuBtratioQS. Fig. 22 ia from a sketch made hy me in 1888 of the left shoulder of 
Bauba (4 f), and fig. 23 is the koima on the right shoulder of another Miriam man. 

Fig. 24 is the ill-made koima on the right shoulder of Bina (13 A). Fig. 26 ia from 
an engraved design on Baubs's coco-nut water vessel. Wanu of Mer made the drawing 

Jha. 36. Sttb koima tHr; about ^ uat. a: 

: dmwn b7 Wana of Mer. 

Flo. 37. Bobbing of th« bMk ol a m 
from Nagii. Brit. Mas. i nU. six 

Fio. 38. BobbiDg of • koima od i tobaooo 
pipe. Biit. Mob. Nat. size. 

shewn in fig. 26; he said that the pis tnva (p. 22) in this instance represent fishes' 
tails, V; the cross in the centre is a star, koimara wer. 

d by Google 


I hare veiy little evidence as to the form of the kotTnai prevalent among the 
Western Islanders. Fig. 27 is a rubbing from the engraved and painted back of a 
square mask that I obtained in Nagir in 1888 and gave to 
the British Museum. Fig. 28 is a copy of a similar design 
which occurs on a bamboo tobacco pipe in the same museum, 
which is said to have come from Cape York ; there is do 
doubt that the pipe was made and decorated tn Torres Straits. 
A koimai design, fig. 29, forms part of the decoration of a 
carved dance toap. 

This completes all the information that I have been able 
to gather concerning this peculiar scarification in Torres Straits ; 
but as it also occors in the neighbouring part of New Guinea 
I have thought it advisable to give, for comparative purposes, 
some of the designs I have collected there. 

Mr R Bruce sent me a tracing (fig. 30) of a scarification, 
called gas-awari, which occurred on the right aim of a man on 
Parama (Bampton Island) ; the design measures 19 by 13'5 cm. 
Fig. 31 is from sketches which I made of the shoulder scarifica- 
tions of Daidu of lasa. Eiwai ; the marks are called tigiri parara, 
shoulder scarification. 

It is also interesting to observe that the &ehion has spread 
to North Queensland, as fig. 32 represents the shoulder marks Fia.S9. BabUng of the oured 
of Aire, a native of Somerset, Cape York, and a member of the '^^"^''' ' ^"^ '""■ 
Eokiamg (?) tribe. The design baa now become formless. 

I was informed in Mabuiag that the men of the Dangal (dugong) and Kaigas 
(shovel-nosed skate) clans had a representation of their totem on the right shoulder, 
but I have no corroborative evidence tq adduce. 

A B Fto, S3. BboDlder iMiifioktioiu of 

Fra.Bl. SboQldH «»rifi«.tioni of m AiistMlUi. mUw rt Cape York. 

Dkidn of Isik, EiwM. 
Fu. BO. Bo«rifia«tioD on the iboDlder A ri^t, B. left ahonUn. 

of KDuniii Ptnm*; ftfter B. Brooe. 

When between seventeen and twenty years old, the young Miriam women were cut 
with the lai^ shoulder koima (att newra tagar koima, big girl's shoulder koima), which 
extended down the arm, measuring about 14 cm. by 9 cm. (fig. 33). If they could bear 

d by Google 



the pain, it was cut on both arms. Id the cutting of this koima a piece of akin waa 
excised, ao that a large wale might result. Though usually made aa a aign of mourning, 
it might be cut for the aole purpose of decoration. The 
four scrolls of the design are called pis mua ; the fringe, 
koimara pem ; and the croae, near koimara wer (cf, p. 22), 

The wooden figure of a girl (pi, m. fig. 4 B and fig. 
21 o) has a peculiar koima engraved ou each shoulder. 

Another type of koima ia that borne by Tag (fig. 8), 
but as she came fi^>m Parama (Bampton Island) it is 
probably not a design that was current in Mer. The 
scarification tu parara or tu paara (arm scar) of a Eiwai 
woman (fig. 11) represents her oi nurumara (coco-nut 
palm totem). An analogoua decoration of Eaubi of Sui 
(fig. 14) represents a ngata, an edible shell-fish that lives 
in crevicea of rocks, but I was informed that she had 
Umai, dog, as her chief totem ; ahe is married to Qabia 
of Saibai (v. p. 159). The New Guinea woman shewn in 
pL HL fig. 3 has a scarification on the right upper arm 
which may represent a lizard or crocodile. 

The centipede, as already stated (p. 20), was often 
cut on the upper arm of Miriam women. 

It was the custom in Saibai and Dauan, and possibly elsewhere, for a sister to 
have a piti tonar, nose sign, cut on the left shoulder on the death of a brother, and 
occasionally one would be cut on the right shoulder to commemorate the death of a 
aister. The length of the scar was made to equal the length of the nose of the deceased. 

ig girl's ahoulder kotma; 
IB ; after J. Brooe. 


Fia. K. ODllioM of louB on the left 
ahoolden o( three women of Danan and 
Saibai, leprsBcmting the noseaotdeoeaaed 
brothari; -) aaX. size; otter B. BniM. 

Fia. SB. Sear oi 

aiae ; after B. Bruoe. 

Fig. 34 repreaenta this scarification on the shoulder of Siba of Dauan; her brother waa 
a very tall man and had a very fine long noae, eo that the scarification meaaures 12 cm. 
in length. 

The three piU tonar of fig. 36 represent: A, the same scar as fig, 34; b, that ou 

d by Google 


s DauaD woman, it is 9 cm. long ; and c, that on Dawa of Saibai, wife of Aki, it is 
55 cm. long. In the two latter figures the transverse curved bar represents a shell 

I of Gemana, a woman 

the right forearm of Maipi of Saibai 


I do not know the significance of the shoulder scar ( 
of the Snake clan of Saibai ; it is 85 cm. long. 

Mr R. Bruce sent me sketches of designs i 
and of a Boigu woman (fig. 37), He said 
that they were called sedau, but this is merely 
the Dandai (New Guinea) name for a mark. 
I was informed at Saibai that they were 
augaiamiin or avgud taman, that is totemic 
(avffud or augad, totem). Probably in both 
cases the design was intended to represent 
a couple of zigzags, each composed of two 
parallel lines, and these would seem to indi- 
cate two snakes; as a matter of fact they 
were recognised as such when I shewed the 
drawings to some Saibcu people, and the 
women were said to belong to the Tahu 
avgud. The arm scarification of Niaoma 
(fig. 9) is a clan mark ; according to Mr R. 
Bruce her sister has a similar one. 

Dr Seligmann noted in Saibai round 
scars on the forearm (p. 13) which were 
commonly so arranged as to occupy the 
angular points of an imaginary diamond- 
shaped area, the long axis of which started 
immediately above the wrist in the mid line 
of the limb and ran parallel with the axis 
of the forearm. 

Young Miriam people of both sexes ^a. 87. SaanfiOBtioi) on the right loreann : <A) of » 
sometimes had the design shewn in fig. 19 Saib«,«id(B)o(»Boigowoman;{A)m 

cut on the outer side of the forearm as a 
sign of mourning \ it was then called tag merbd kovma. 

SoibM, uid(] 
D kngth, (B) 8-S; ktter I. Bmoe. 

10 am. 

Soarifioatloni on the Le^. 

Scarifications on the legs do not appear to have been very firequent among the 
Torres Straits women, as I am aware of only one or two instances, but they seem to 
be more Sequent in New Guinea. Bonel of Saibai (fig. 38) has a zigzag line of scars 
down the outside of the right thigh, which represent pelicans, owat, flying or floating 
in a sinuous line; the design was termed awaiau, ita labai (of pelican these cuts). 
These marks are cut by the father on the right thigh of the girl as a mark of puberty, 
and serve as a sign that she is ready to be married. 


d by Google 



Fig. 39 is a similar acarificatioD, parara, on a fine, strapping woman (about 30 yeaiB 
of age) of Mawata, New Quinea; it is stated b; Mr B. Bruce to be a clan mark; 
it probably indicates that she beloi^ to the coco- 
nut totem (cf. fig. 11 and p. 26). 

The women of the Cassowaiy clan of Mabuiag 
are said to have had a V-ahaped scar on each 
calf, which represented the foot, or footprint, of 
that bird. 

The young Miriam adults of both aexee some- 
times had the device shewn in fig. 19 cut on the 
calf as a sign of mourning, teter merbd Jcoima.. 

I noticed in 1889 the following scarifications 
on Miriam men shewn in fig. 40: A on the right 
calf of Marizet of Dauar, B on the right calf of 
Ulai (merbd uwr), c on the left leg of Bina, and 
D on his left buttock (kip user) which he said had 
been cut on the death of a relative. Dual (28), a 
Dauar man, had a representation of a hammer- 
headed shark, called iruapap tarim (described as 
"the greatest of the hsmmer-headed sharks"), cut 
oo the fi^mt of the thigh as a charm against sick- 
ness. He was the only man then alive who had 
one of these. 

I was told at Mabui^ that men would often cut a long feather-Uke mark on the 
calf of each leg, madu utal, for the purpose of drawing the attention of women to 
their fine legs and their activity in dancing. The same custom obtained at Yam. 

W. B8. SoMtSoctioiu OD the right thigh of 
Boiul at Skibu ; direct length of tbaM Mart, 
20 om.; after B. Brace. 

Fio. S9. BoarificatioD <m the right thigh of a 
Maw»ta woman; the total length of the 
•oar is BO-5om.; after B. Brace. 

Pio. 10. BaariScationa on the legs of Uiriam a 

The men of the Tabu clan of Mabuiag were said to have a coiled snake cut on 
the calf of each leg. 

The scarification of the calf of the leg is also practised in Daudai. Two examples 

d by Google 


tnofit Boffice, both of which were cut on the calves of womeD's legs. Fig. 41 waa 
said to be a Torres Straits pigeon (Carpophaga), ^mai. Fig. 42 is a centipede, eheri 
or gera, and more closely resembles the original animal. 

Fio. 41. Searifloation on the oalf of Skouu, of Fio. 13. 8«uifloa(ion on the a«U of Skwiui (?)i 

Itobobo, Dandai ; length 163 mm. ; aftsr B. Bnia«. length 133 mm. ; ftftec B. Brnoa. 

Tehpobart d£cobation of thi: Skin. 

The face and the body were partially or wholly ornamented with p^ment on various 
occasions. The ceremonial painting, which probably was always definite and traditional 
in character, is dealt with in the descriptions of the several ceremonies. Yam and Tutu 
girb had their whole body blackened at puberty (v. 202), and this waa also done to 
lads during the initiation ceremonies in Tutu, N^ir and Mabuiag (v. 209, 212, 213); 
&ce-painting occurred at Saibai (v. 216) and Mer (Vl. 292), and other parts of the 
body were also painted red in Her. In Mabuiag men who represented ghosts in the 
death-dances (v. 253, 254) painted themselves with charcoal. Moumeis who were not 
related to the deceased were marked with charcoal on the shoulder in Mer (vi. 146) 
during one part of the funeral ceremonies, and the relatives were similarly marked on 
another occasion (vi. 157); near relatives were smeared with a greyish earth (vi. 163). 
Certain zogo le painted themselves before officiating at the zogo (VI. 203, 219, 268, 274). 
The painting of the participants in the Bomai-Malu ceremonies is fully described 
elsewhere (vi, 289 — 294). A round spot of red paint on the chest was said to be the 
distinguishing mark of a male member of the Crocodile clan in Mabuiag (T. 165). 

Warriors on the war-path were generally painted red all over (v. 299), and in a 
war-dance in Mer the performers were decorated in the same way (Tl. 276), but in 
war-dances witnessed by me in the western islands (v. 302, 304) the painting was 
nsually of a more varied character. 

In the secular dances, or on other occasions when a man wanted to make himself 
"flash," the painting on body and face was unconventional, and that of the latter was 

d by Google 


gsDenilly limited only by the resources and ingenuity of the individual. I have seen 
young men and girls with the two sides of the face painted in different colours, and 
symmetry in lines or dots was rarely adhered to. The people were always very proud 
of their decoration and behaved in a serious manner, though the effect was generally 
ludicrous to the European. 


The hair was, I believe, never allowed to grow indefinitely, nor was it left entirely 
undressed. Formerly it was worn long by the men (pis. I, fig. 3, XXIV. fig. 2) and short 
by the women (pi. I. fig. 1, pi. II. figs. 1^-4). I do not know that the latter ever 
wore their hair long, but in the story of Qelaro it is stated that his mother rolled 
her hair into long ringlets, iaiai (v. p. 38). When the hair was cut short, tal bup, 
it was occasionally combed; probably every woman had her comb, which she kept in 
her basket. If the hair is cut very short it has, owing to its frizzly nature, the 
appearance of growing in little tufta, which somewhat resemble the so-called "pepper- 
corn " hair of the Bushmen of South Afiica. 

The young men now frequently cut their hair in various styles (pi. V. figs. 1 — 9). 
I noticed in Mabuiag in 1888 that one favourite method was to cut it in such a way 
that it sloped backwards from the forehead so as to cany on, as it were, the slant 
of the forehead, while it was trimmed behind to form a &irly straight vertical line. 
This fashion appears to have had some connection with the antero-posterior cranial 
deformation formerly practised, ae it emphasised the local conception of a well-shaped 
head. Macgillivray (ll. p. IS) says that at Cape York and Muralug the hair is almost 
always kept short ; the Muralug natives are however affected by their Australian neighbours. 
"Sometimes the head is shaved, leaving a transverse crest, a practice seldom seen among 
the men, but not uncommon among women and children from Damley Island down to 
Cape York" (l.c. p. 13). Jukes saw two Masig women who "had their frizzled hair 
closely cropped all over, except a ridge about half an inch high, running from one ear 
to the other, over the crown of the head" (I. p. 166). Through the kindness of my 
friend Dr Otto Finsch, I am able to shew the photograph of a Habuiag girl (pi. IL 
figs. 3, 4) whose hair is dressed in this style. Cutting the hair is a sign of mourning 
amoi^ all the islanders (v. p. 249). In Mer (vr. p. 153) the relatives and friends 
helped one another to shave their hair, and the men removed all hair from their &ces. 
The men left a transverse ridge of hair across the head in front of the plane of the 
two ears. This ridge, about 5 cm. (2 in.) broad was called kaisu or mta dart &<om its 
resemblance to the headdress, dart, worn at dances. The man in the centre of fig. 1, 
pi. XXVIIL has his hair cut in a similar manner. The female relatives and friends left 
a similar ridge running from ear to ear (i.e. a little further back than that of the men), 
or, if they preferred it, a small tuft of hair at the vertex; this was called kuk from 
its foncied resemblance to a shell (Nerita, or other shells). The women now frequently 
part their hair in the middle (pi. II. fig. 6). 

A characteristic mode of dressing the hair, now quite obsolete, was to twist it up 
into long pipe-like ringlets by rolling it between the hands and saturating it with mud; 

d by Google 


this was termed iaJai (W.), ad (K), (pi. I. fig. 3 ; pi. XXIV. fig. 2). Frequently in Mabuiag 
a small triangular slab, kuruai, was placed f^faiost the nape of the neck on which the 
hair was rolled in mud ; it was made of turtle-shell, or of iiraka, naiwa, mipa, or kusvb 
wood. I suspect that the triangular decorated turtle-shell object (pi. VII. fig. 4) in the 
Truro Museum is a kurvai; it is described aa a "neck ornament from Murray's Island." 
It was collected by Lieut. J. B. Eempthome during 1840 — 41, and measures 192 by 
170 mm. In a Miriam folk-tale (vi. 31) a youth was laid (ace downwards on the 
ground and mud, bud, rubbed into his hair; after a few hours the mud was taken off 
and the hair rolled into rope-like ringlets with oil and red earth. In Mer at the end 
of mourning the hair of men, which by this time had grown long again, was freed 
from ashes and trimmed; the men then worked it into ringlets, ed, employing at the 
same time charcoal made of burnt coco-nut shells, ketf, mixed with coco-nut oil (vi. 160). 
I have also a note that in Mer hair is blackened with coco-nut oil and coco-nut charcoal, 
u meaur keg (u, coco-nut ; mes, coco-nut husk). Jukes (I. 159) saw some Masig (" Masseed") 
islanders whose "hair was dressed into long, narrow, pipe-like curls, smeared with red- 
ochre and grease" (pL I. fig. 3) and in Erub the pipe-like ringlets were sometimes left 
in their natural black colour (l. 171). 

The hair was and is still often reddened by the use of lime; I never heard that 
the lime was employed for altering the colour of the hair, but only to get rid of lice. 
On certain occasions, as at the end of a girl's puberty ceremony in Yam And Tutu 
(r. 202), the hair was ruddled with red-ochre, parama. 

There is a fiur amount of hair on the &ce, but it could not be described as 
luxuriant, and it is certainly trimmed. The &ce is now frequently shaved, a moustache 
being left. 

Brockett says of the Miriam: "For shaving, they sharpen a piece of bamboo, bend 
it nearly double, and then draw it down the Bice" (1836, p. 27). Glass bottles were 
eagerly sought for from passing sailors, the splinters of which were employed as razors. 

Wigs, adiui-ial (W.), were occasionally worn in all the islands: they were made of 
human hair from live men, which was always dressed in long ringlets. I do not know 
how wigs werB made before the introduction of calico. Those which I collected were 
made on a foundation of foreign material In one which I collected in Muralug in 
1888, the hair is twisted into cords or ringlets 16 to 24 cm. in length, each cord is 
doubled on itself and is sewn at the bend on to a foundation of some European textile ; 
the inside is lined. At the vertex is a strip of red flannel which forms a tassel. The 
forehead band, kuiad uru, of the wig is decorated with finnly sewn beadwork in dark 
bine, red and white (pL I. fig. 6). 

We collected in Mer some hair fringes which were worn as a kind of wig. Each 
lock of hair is treated in the same way as a bunch of fibre in the petticoat shewn 
in fig. 123 a, but with a single weft. In one specimen, with reddened hair, in addition 
to the cord which passes through the loop of the lock, a second cord passes between 
the long and the short ends below the lashing. The fringes are decorated along the 
top border by a row of overlaid twining of a bUck and a yellow strip which conceals the 
method of construction. The longer strands range fiwm about 16 — 20 cm. in length. 

Jukes says that the wigs were "not to be distinguished from the natural hair, till 

d by Google 


cloeely examined" (l. I7l); thia entirely coincides with my experience in 1888. He says 
(p. 169), "Oaria had a black wig dressed like their hair, but his beard and whiskers 
were nearly all grey" (pi. I. fig. 4). D'Albertis states that in Ernb (1875) "The men 
also wear wig8, which the women make for them. These wigs consist of a number of 
curls attached to a kind of cap; the length of the curls varies from four to six inches. 
The cap is kept on the head by a strip of bark which passes under the chin" {New 
Guinea, I. p. 239). 


The natives frequently bathe in the sea, but never in fresh water; indeed the 
supply of fresh water is extremely scanty, and there are do pools iu the great majority 
of the islands where bathing would be possible. The skin was 
sometimes anointed with coco-nut oil, which was expressed &t>m 
the scraped kernel. 

Most of the slight amount of attention paid to the toilet 
was spent on the hair, and probably every adult person possessed 
a comb which was used for combing the hair and not as an 
article of adornment. 

Combs, sak, ial aak, iai pat^ (W.), kerem aeker^ (E.), are 
usually made of bamboo or wood, I know of but one made of 
turtle-shell (pi. X. fig. 6), which I collected at Mabuiag; it measures 
about 76 mm. each way. The handle is either quadrangular or of 
a domed or conical contour (pi X. figs. 5^12), the curved tops being 
characteristic of those made of bambdo. The teeth are strong and 
usually long, and vary from six to eleven in number. Out of sixteen 
combe in various collections, all but two have some pattern or other 
carved or engraved on the handle. The nature of the decoration 
of various forms of combe is dealt with in the section on decorative 
art. The general appearance of the combs is shewn in fig. 43 and 
pi. X. figs. 6- — 12. Combe vary in length from about 133 mm. 
(5i in.) to 216 mm. (SJ in.). I was informed in Mabuiag in 1888 r^. «. B«mb«. comb, 
that the combs with a square handle were a new fashion intro- HkbnUg; length is em. 
duced by South Sea men; they were called sal sak, but I am '*• ^' ^ Coll.). 
not certain that this was the case. The comb, ial »ak, formerly in use, was said to 
have had a very slender T-shaped handle. 

In the vocabulaiy occurs the word pai (W.), for a fan. I do not know whether 
a regular &n was made before the coming of the South Sea men. In the &nning 
game, totuam, described later, the Miriam used the leaves of the abal tree, Pandanue. 

■ ial, hftir; pat, k ehatplj pointod lUok; paU, to go into or entei; lak, oomb. 
* lierem, h«»d; lektr, uiTthiiig long, thin Hid ibarp, » comb. 

d by Google 


OBNAHEirrs worn on the person may be classed under two_ headings: — (1) those 
which are worn in every-day life or on festive occasions, and (2) those which have 
a ceremonial significance. With the exception of a few objects worn by the senior 
members of the Bomai-Ualn fraternity of Her (vi. p. 295) no objecta were worn, eo 
&r as I am aware, as symbols of rank or status, but certain ornaments such as a 
particular ear-pendant (p. 40) were worn by "big men," probably because they were 
objects of value or beauty which would naturally tend to come into the possession of 
important men. 

For every ceremonial occasion a special form of decoration, eamiak (W.), peror (K), 
was employed. In some cases, even if masks were worn, the costume was not elaborate, 
the performers wearing a petticoat, baldrics, armlets and leglets made of young coco-nut 
leaves, tu (v. pL XVIII. fig. 2); in these instances the mask was the distinguishing 
feature. In the majority of cases, however, the painting of the body and the particuW 
costume or decoration were distinctive and were not employed for other occasions. Theee 
are dealt with in Vols. v. and VI. when describing the ceremonies themselves, and need 
not be here repeated. 

Head Obnahents. 

Brightly coloured flowers, of which the scarlet hibiscus is the fovourite, are frequently 
worn in the hair; occasionally brightly coloured leaves might also be thus utilised, 
but I never saw combs used for onuuroental purposes, these being always kept in 
baskets; an Erub man, however, is drawn by Melville (Jukes, ii. p^ 237) with a comb 
in his hair. The decorative treatment of the hair has ab'eady been described (p. 30). 
The wigs alluded to on p. 31 were purely ornamental in character. 

I believe that feathers were worn in the hair only on ceremonial occasions, indeed 
the single record I have of them is in connection with the Bomai-Malu cult (vi. pp. 288 — 294). 
The essential feathers are the black-tipped white wing-feathera of the Torres Straits 
pigeon. The daumer lub, as it is called, may be a single feather (fig. 44), the quill 
of which is stuck on the sharpened point of the denuded shaft of another feather, so 
as to increase its flexibility ; or may be composed of several feathers (fig. 45). The more 
complete daumer lub consist of a black-tipped feather inserted on a split ruddled quill, 
the object being to keep the feather in a constant state of quivering. The shafts of 
two white feathers are split in a zigz^ m&(mer, and one half of each ia bent round 
to form a ring. These are tied to the base of the supporting quill, to which are also 

H. Vol. IV. 6 

d by Google 


laahed two wooden spikes to serve as a comb. The total length of such specimens is 
from 35 to 40 cm. 

Fia. H. Damttr Ivb. Via. 4C. Dawmer lub. Fio. 46. TitnUitni. 

A. head ornament (fig. 46) formerly worn in the hair by the young men of Mabuiag 
was called tituititui {titui = star). It consists of a diamond-shaped object made of stripe 
of Pandanus leaf supported on a short horizontal and a long vertical stick, the npper 
end of the latter is decorated with a tuft of cassowary feathers and a bunch of bisi 
(shredded sago-palm leaO dyed russet. The centre of the diamond is black and there 
is a marginal and intermediate black band. The sticks are reddened and vaiy in 
length, in the four specimens made for me, from 21 to 36 cm. The diamonds average 

d by Google 


12 cm. in diameter. An ornament of this kind, the free ends of which weie Tariously 
decorated with feathers, was occasionally held in the hand when dancing in Muralug, 
and probably elsewhere. Similar objects are made in New Quinea, at Mawata, and by 
the Motu of Port Moresby and Huhi, where they are said to be stuck in houses for 
play (Head Hunters, p. 225 and fig. 26). "We look at hill, and make him all the 
same," said a Mawata man to me; but the only hill in that part of New Qoinea is 
Mabudavum, which is a long way off. One is figured in the Albvm (ll. pi. 199, No. 2) 
where it is described ae a "charm, pipifoa, used in adorning a new house, made of banana 
fibre and coloured red and buff, Port Moresby." 

Prontlets, kuik um (W.), mat lager (E.), of various kinds were occasionally worn. 
Coronets made of several rows of dogs' teeth (pi. IX. fig. 4) were called by the same 
name as the dogs' teeth necklaces (p. 41), all of which were I believe imported from 
New Quinea. I obtained a coronet, pik uru, at Mabuiag (pi. IX. fig. 6) which was 
made fix>m kangaroos' teeth, which also was necessarily imported. 

Frontlets, kviokaia (W.), made of the yellow sprouting leaf of the coco-nut palm, ia 
(W.), «*, « kupi, ura kupi (R) were commonly worn; these were tied in various ways 
(fig. 47 ; pi. v. fig. 10), usually a single or doable 
ring was made (gagai dan, "eye of the king-fish" 
(W.), irkep, •• eye ball" (E.), c£ Vol. v. p. 249, VoL VL 
pp. 16, 132), sometimes with long ends projecting 
beyond the head. A similar frontlet, kauwidoi (W.), 
was made of the leaves of the Pandanus pednnculatus, 
kaiiaa. Ao- Vt- Ttarae palm-leif frootleta, 

The kataclrUTV. or Jtiwod-wl (W.), is a fillet «'^°«- 

decorated with Coix seeds, hu; it was bound round the 

head of the Mabuiag kemge, or young men during the period of initiation (v. p. 213) to 
keep the long ringlets of hair, to/, off the forehead ; an elaborate one from Tutu is shewn 
in pL IX. fig. 1, where it may have been used on 
similar occasions, but doubtless was not confined to 
them. The band measures 266 x 60 mm. ; it is com- 
posed of a narrow encircling plaited band and a central 
broader plaited band on which the oblique seeds are 
sewn, the two rows of vertical seeds bridge the interval 
between these rows, the empty spaces between the 
vertical seeds give a light appearance to the band ; the 
central tassel consists of Coix seeds. A similar fillet 
was obtained at Mer (fig. 48), it consists of a simple 
plaited band about 37 by 2 cm., on the foce of which 
strips of yellow orchid skin are applied and a double 
row of Coix seeds; there is also a central tassel of 
seeds and violet wool. 

I obtained in 1888 at Tutu a fillet made of cuscus fiir, 
feari((Fhalangermaculatus),whichcamefromNewGuinea ^''- "" "Hot deo^t»d with Coii wed., 
{AU>um, I. pL 339, No. 12). It is in the British Museum. 


d by Google 


A row of oblong, oval, circular or diamond-Bhaped pieces of nautilus nacre, kaura (W.), 
piau (R), strung on two strings, was worn across the foiehead on festive occasions 
(fig. 49; pis. VIIL fig. 2, IX. figs. 2, 3). In Mer it was called piau mat lager, or 
idaid (nautilus) mai (forehead) lager (rope or cord). I obtained one at Muralug in 
1888 made of large irregular oval pieces of nacre, it was called kaura dan, nautilus' 
eyes; the name at Mabuiag is kawra kwik uru. 

The aaurisauri piau of Mer is a star-shaped ornament (fig. 50) 68 x 35 mm., made 
of nautilus nacre. It is perforated with five star-like holes and two small simple boles 
by which it was attached to a band which passed over the forehead. Sauriaauri is 
the name of a common blue starfish (Linckia bevigata) which frequently has four arms 
1 of the customary five. 

Fia.t9. Piau mat logn-, Mer, ftboat I Dftt. Bise. The Fie. GO. 5aiiri(auri piau. Mar. 

length of the nftoreons portion is 16'S om. in A, uid 
S6 om. in B, eaah piew being IB min. long in A., uid 
10 mm. in B. 

The most common head-dress (pis. VIL fig. 1, VIII. figs. 2, 5) is that made of 
black cassowary feathers, sam; it is called dagui {dagori, dagoi) or aamera (W.), and 
mm or dagui (E.). In some specimens the feathers are long and curve backwards over 
the head and usually at the same time away from the middle line; rarely they curve 
over in front. In others they are short, in which case they have a tendency to stand 
erect, and are frequently brownish in colour; rarely they are cut to form a straight 
upper edge. A headdress of short feathers is called in Mer ker/ear sam {kerger aam) 
or young sam and also wer sam. Sometimes a tuft of the plumes of the paradise bird 
(Faradisea ra^ana) is inserted in the centre (pi. VII. fig. 2), and occasionally small 
white feathers are fastened to the ends of many of the cassowary feathers with beeswax. 
The head-dress is made io the following manner. About 8 quills (each of which benrs 
two feathers) are bound together by fibre for about 25 mm. from the end. A large number 
of these bundles are fastened together ; two longitudinal strings are placed close together near 
the upper and two others near the lower end of the wrapped-up portions of the bnodles of 
feathers at the back of the head^lress; a string b twisted round each bundle so as to include 
the longitudinal string and passes in front over the next bundle aa well, round which it is 
twisted, and so on. Thus looked at from behind each bundle is encircled by a string which 
encloees the longitudinal string, and in front each loop of the string passee over two bundles' ; 
' Thii a oolled "wntpped veaving" b? 0. T. Mbbod {"Abotig. Am. Basketry," Btp. U.S. Nat. Mui. for 
1903 {1901), p. 331), bnt the emploTment of the longitadinsl etring cftoeeB it to approximate to wrapped twined 
weaTing, p. 23S. 

d by Google 


as the front strings are arranged obliquel;, the two upper and the two lower Btrings form 
an npper and lower herring-bone deaign. The longitudinal strings extend far beyond the 
forehead-band thus formed, and are pUit«d together, with the weaving Btrings, to form a cord 
for tying the head-dress on the bead. Sometimes each pair of weaving strings is a single 
sbring doubled on itself in which case the longitudinal strings are passed through them at 
the taid where the doubling occurs. 

The bound up portion of the feathers usually appears as a broad forehead-band 
which is generally painted white, while the string-work above and below is reddened. 
Rarely the band is decorated with Coix seeds; in pi. VII. fig. 1 the seeds are threaded 
on red wool which projects beyond each convergent pair of seeds In a vertical tassel. 
The band may vary from 14 to 40 cm. (5^ — 15J in.) and the length of the feathers may 
reach 43 cm. (17 in.). 

Usually a supplemental forehead-band is placed in front of the original one, this 
is composed of several strips of ratan or some flexible strand over and under which 
a string is tightly laced. The "stroke" (see section on Basketry) is a plain check, but 
the technique is essentially wickerwork. This forms a very firm and durable band, 
which, in our specimens, is variously painted red and white, a string cloison being 
attached at the junction of the colours. 

Melville drew an Erub man wearing a head-dress composed of the black-tipped white 
feathers of the Torres Straits pigeon (pL III. fig. 3). This is the only record we have 
of a very effective ornament. 

Very rarely one finds a head-dress of simple construction entirely composed of the 
plumes of the common biid-of-paradise (Paradisea raggiana) ; it is called by the same 
name as the bird itself, ddgam (W.), degem (K). The specimen figured on pi. VII. fig. 2 
was obtained at Saibai, and was said to have come originally fitim the Tugeri. 

A waipat (W.) is a head-dress consisting of a single plume. 

The finest head-dress is that called dhi or dri (W.), dari (E.) (pis. VI., VIIL, 
figs. 1, 3). It consists of a fl'^^aped framework, bume (W.), either open or covered, 
from which projects a tan-shaped arrangement of the large feathers of the white variety 
of the reef-heron (Demiegretta sacra), karbai (W.), sir (E.). 

In those with an open-work frame (pL VI.) the fl-shaped frame consists of several 
bars of split cane or ratan, which are fastened together 
with soil string by wickerwork technique (fig. 51). A peri- 
pheral one is separated frnm the outer bar of the conjoint 
group by a gap which is crossed by a slip of cane sewn to 
these two bars in such a way as to form a zigzag: it also 
is covered with string (fig. 51). Two strips of cane served 
round with string in the usual manner to form a single bar 
are festened ceatripetally to the lower portion of the frame : 
about its middle they span the open space in an M-shaped 
manner'. A vertical band consisting of several bars passes 
down the median line fr^m the apex of the fl ^° tbe centre 
of the bent bar. Usually a transverse band composed of 

' In the Bpeoimen shewn in pi. VI. Sg. 1 there ore tn 

d by Google 


two or three bare passes from one side of the frame to the other, and is fastened to 
the lower end of the vertical band and the apex of the M-shaped bar. The spacea 
between the vertical band and the frame above the M-ehaped bar are crossed by crescentie 
bars served with string. The framework is variously painted in red, white and blue. 
The quills of the white feathers, which form the characteristic feature of the ornament, 
are fastened between a split cane or two strips of ratan in tfae manner shewn in fig. 51. 
This structure is bstened at intervals to the outermost bar of the frame. A strand 
of black cassowary feathers, closely bound to a strip of cane, is &8tened to the front 
of the support that carries the white feathere, thus concealing their mode of insertion, 
but frequently it also conceals the space crossed by the zigzag. The feathers at the fr^e 
ends of this straud project to form a kind of tassel. The white feathers forming the 
fan-shaped arrangement are cut in the characteristic way shewn in the illustrations. 
A V-shaped gap is always left in the centre; occasionally the free vanes of the two 
feathere bounding this gap are notched as in a specimen from Tutu which I gave to the 
British Museum (pL VL tig. 1). A long white feather projecte horizontally beyond each 
base. A tuft of bird-of-paradise plumes or a single long feather of some sea-bird is 
generally inserted in the central gap. The apex of the framework may be decorated 
in front with tufts of feathere, a seed, or other object. 

It is unnecessary to go into minor variations, but one point is worth mentioning. 
There is considerable variation in the disposition of the symmetrically disposed bent 
bars, and I was at a loss to underetand their significance till 
I saw a specimen which belonged to Mr Frank Summers, then 
resident in Torres Straits (fig. 52). It is obvious that the median 
bar represents a nose, and the curved line above the eyes eyebrows, 
while that below is probably intended for a nose-stick, below this is 
a mouth. The eye discs are made of pearl-shell with a pupil of 
black beeswax. There is, of course, no evidence that this specimen 
retains the original motif, it may be the last of a series, but I am 
inclined to regard the simple examples as degenerations from a face ^**- '*'■ ste'<* "t Uw 
design. This conclusion at which I arrived in 1889 is strengthened hnmll'faoe'deiiim" ffor 
by the foct that we obtained at Mer a dri (pi. VL fig. 2> having whole head-dre™, we 
two large eyes and a characteristic wooden nose. A red wadai bean Joum. Anth. Intt. xn. 
is tied above the nose. It measures 50 x 37 cm. P'- '""' ^8' ') 

Specimens average in length horn about 48 to 54 cm. and in breadth firom 37 to 43 cm. 

In the other type of dSri the framework is covered (fig, 53; pi. VIII, figs. 3, 6; 
Album, I. pi. 334, No. 1 from Mer, n. pi. 191, na 5 from Debiri, delta of Fly river). 
The forehead portion consists of a stiff triangular, semi-ovoid or rarely diamond-shaped 
structure which is composed of horizontal strips or bare of ratan or similar material 
bound in the same manner as the forehead-band of the cassowary head-dresses (p. 37) 
or the framework of the other kind of dSri (fig. 51). The fiw ends of this wickerwork 
on the convei'ging sides are kept rigid by two stiflf bent strips of ratan, which are 
served with string in the usual manner. In those specimens in which the lower margin 
is not straight (i.e. formed by a continuous strip) the free ends of the strips are 
strengthened by an additional continuous bent strip. The central vertical line of the 

d by Google 


strootare is strengthened by two 8ti£F alips, which are uaually a folded atrip of nitan, 

the bent end of which forms a loop projecting beyond the lower margin. A ab<ong 

loop is worked into the lateral aogleg for the 

attachment of the cord by which the head-dress 

is festened on the head. The convergent sides 

are finally furnished with an open-work zigzag 

and fim of white feathers. 

The body of the frame is painted white and 
there is usually a central design in red and black, 
margined by a string cloison, which is variable 
in form (fig. 53), the significance of which has 
not been elucidated. The circumference is usually 

I believe all these head-dresses are imported 
from New Guinea, and I am under the impression 
that they are definitely war accoutrements in the 
districts where they are made. 


Fia. SB. Doiigui punted ou oert4iii bcad-drauea. 
A— D, Cunb. Has.; E, Oiford; F, Q, QUagow. 

Nose and Ear Obnahents. 

The septum of the nose is pierced (p. 9) in all the old and in most of the 
middle-aged people, but, like the distension of the lobe of the ear, this custom has now 
fiiilen into desuetude. Into the orifice, piti tarte (W.), au pit neb (E.), was inserted 
a long or short shell rod or plug, yu6, guha, or gigub (W.), kirktd>^ (E.) (pi, VIII.). 
The long forma probably were worn only on festive occasions, but the short may 
have been in more general use ; none are now worn. The alse nasi, lips or cheeks, 
were never pierced, but two small holes are sometimes bored in the tip of the nose 
of some Miriam men (p. 10). Mr J. Bruce informs me, that "when the men wish 
to be particularly bscinating in their appearance, they insert into each hole a shoot 
of young grass or piece of the young shoot of the coco-nut palm leaf. The shoots are 
held in position by pressure only, and being a pretty pea-green in colour look very 
efiective against the black ground. This is also used as a dance decoration, together 
with the kirkub." 

The nose-sticks vary greatly in size and form (pi. IX. figs. 6 — 12). Occasionally 
they are colunmar with square ends, more often fusiform with blunt or sharpened points, 
and either straight or curved. 

They &11 into several groups: 

(1) Short, thick and more or less cylindrical, but usually oval in section, the ends 
are generally squared. They may be so short as to be little more than a plug for 
the hole in the septum nasi. The length varies irom about 2 to 5 cm. They appear 
to be generally made of the shell of the giant clam (Tridacna gigas), maiwa (W.), miakor 
(&), but some are made from the columella of a large Conus millepunctatus, wavri. 

I The preBiM in (W.) And Xir(B.) [probabl; a modifioation of gir (E,)], "boar's tusk," snggett tlut (ormeil;, 
pertutpt betore th« uktivM had migrated (o the Islaadi, the nose oroament was aotnallj a boai'a task. 

d by Google 


(2) One short variety is shaped like a narrow, sharply-pointed spindle, 47 — 61 mm. 
in length, and is generally made of as, Cassis comuta. Our specimens came from Mer. 

(3) Long, thick, straight or but slightly curved, and more or less circular in section. 
The Cambridge specimens came from Mer and were said to be made of as shell; one 
which was broken at one end measures 225 mm. in length. They are made from the 
shells of Tridacna, Conus, Cassis, Megalatrachus aruanus, 6m (W.), ttud>er (E.), and Melo 
diadema, alup (W.), eter (K). They vary in length. One made from a nudwa shell is 
flattened and much curved, and measures 212 mm. in length. Those made from cone 
shells are usually the shortest. 

(4) A more curved variety of the long nose-sticks is oval or flattened in section. 
These seem to be uenally made from the outer whorl of the maber or eter shells. One 
specimen of maher shell from Mer measures 27 cm. along the curve and 188 mm. from 
point to point, and the longest (broken) specimen of ezer shell is over 20 cm. in length. 

In the NavHcal Magazine (Tl. 1837, p. 753) it is recorded that in Mer, " the septum 
nariwm is perforated in which, at times, they wear a circular hook of tortoise shell"; 
I think this must be a mistake, as no mention is made of the shell nose-stick. 

Ear ornaments and pendants were of frequent occurrence, but I have never seen 
OF heard of actual ear-ringd that could be directly fastened in the ear. The piercing 
of the lobe of the ear is described on p. 10, the distended lobe, taatU (W.), being 
reduced to a ring of flesh, but most frequently it was severed on the side nearer the 
face, thus forming a pendulous, fleshy cord, muti (W.), iaip eak (EI.), which was a 
veiy remarkable feature in the appearance of these natives. The outer margin of the 
ear, {e6 (R), was usually pierced in a consecutive series of small holes and I believe 
these were sometimes continued down the pendulous lobe (pL I. fig. 4). 

Small pieces of grass, small flowers, seeds, uta htrsai (W.), bits of worsted, or other 
objects are even now occasionally inserted in these marginal orifices. The fully decorated 
ear of an ancient native must have presented a striking appearance. 
This was known as mvti (W.), teb (E.), names which also apply to 
ear pendante of seeds, Inu leb (E.), and other ear ornaments. Tt is 
this which is alluded to in the folk-talc narrating the parthenogenetic 
birth of Eusa Eap (T. p. 24). The seeds worn in this case, and 
indeed practically universally, were those of the kus, Coix lachiymie, 
"Job's tears." 

Definite ear ornaments made of shell were occasionally tied on 
to the end of the pendulous lobe, Iaip; characteristic examples are 
shewn in pis. IX. figs. 13 — 16 ; VIII. figs. 2, 5 ; those made of pearl- 
shell are called gagi mai (W.), mai 2e6 (or Iaip) (E.), they vary from 
about 60 to 75 mm. across. I suspect that the ear pendants were su^ested by small 
examples of the crescentic breast ornament. They are provided with a central shank, 
which is perforated above, and usually with a pair of spurs below. I was informed 
that those, idaid Iaip or Iaip piau, made of the nacre (piau) of the nautilus {idaid) 
were worn by the "big men." One beautifijl little pendant, 19 mm. wide (pi. IX. fig. 17) 
is made from the septum of a nautilus. A ring, godegode (E.) (fig. 54), made &t>m 
the top of a cone shell (wauri) was worn in Mer; it was tied on to the lobe by 

d by Google 


means of a fibre attached to the projecting knob. They vaiy from 82 to 53 mm. in 

Leaves or flowers were sometimes carried between the ear and the bead, indeed 
this space often serves as a pocket. 


Necklaces, tabo kaabkaab (E.) "neck spheres" (perhaps it should be goffob, a ring 
of rope), were commonly worn and were generally made of plaited stripe of ratan, 
bamboo rind, or similar material, or of twisted or braided coco-nut fibre (pi. I. fig. 4). 
Seeds and shells were strung for necklaces or sewn on plaited bands. In later times 
stripe of European stuffs, turkey-red twill and the like, were employed. Very frequently 
the necklace was employed merely to cany a pendant. 

A necklace of the canine teeth of dogs, wnai-danffol (W.), geserig or auauvri (E.), 
was a much-prized object, probably due to the &ct that a dog could supply only four 
teeth so that a long necklace would represent a lai^ number of dogs. Each tooth is 
peribrated near Its base and it is sewn through this faole between two braids (pi. XI. 
fig. 18). It was worn only by women and girls. 

Fio. SS, PoTtioD of ft wanu DecUam, Her. 

Fid. S6. Fortlou at ft 00*17 dboUmw, Uei ; sbont ) Dftt. nie. 

Small olive shells, urax or toarat (W.), waraz (E), were of considerable value 
especially when made up into a necklace. A good one was equal to the highest unit 
of value (a canoe, etc, p. 44). When fr«sh this olive shell is of a grey colour, but 
I was informed that "cook him and he come white." I obtained one fine specimen 
in 1889 at Mer which is about 91 cm. (36 in.) long, the shells of which are fostened 
to a cord of plaited coco-nut fibre; it is now in the British Museum (pi. IX. fig. 18). 
The longest specimen obtuned in Mer in 1898 is 282 cm. (9 ft. 3 in.), it is composed of 
buff and a few grey oHve shells and two or three Deatalium shells, all stnmg lengthways 
on fine twine (fig. 55). A short one is shewn in fig. 67. 

H. Vol. IT. 6 

d by Google 


One rooghlj made necklace from Mer (fig. 56) is compoeed of small white cowries 
with a Dolium shell in the centre. Another one from Mer, 73 cm. long (fig. 57), is 
made of Cantharidus torresi, interspersed among which are solitary specimena of Liotia 
varicosa, Euchelus altratus and Trochus fenestratua, it bears a tassel of coloured wools 
and a pendant. 

The only seeds that were employed for necklaces were the hard grey fruits of the 
grass, Coix lachryms (Job's tears), kus or kuaa (W.), ktttu (K), which were perforated 
through their long axis and were either used entire or halved transversely; usually they 
were threaded in a single series but in one old specimen (fig. 58) they are in threes. 
I collected one specimen of halved seeds, about 68 cm. (27 in.) in length, at Muralug, 
but it was said to have been made in Moa; it is now in the British Museum. In the 
western islands these necklaces are called Inual and a tassel of the same seeds is called kus. 

Fia. 68. NeaU«oe of teedg ; i nat. i 

Fio. S7. Shell ueoUaoe with pendut; Um. 
The iapar (W.) is a temporary necklace made of the scented leaves of the watnadai 
plant and worn by women ; it was also worn by ipikamarlxti during the death-dances 
in Mabuiag (v, 254). A necklace was made for amusement of the scarlet Bowers of the 
Indian coral tree (Eliythrina indica), piner (W.), naur (E.). 

Various objects were employed as chest pendants*. A simple one is an old Cyrena 
shell, 85 mm. long, which was fastened to a strip of red calico (fig. 59). Other shells, 
such as id, are worn. In the British Museum is a trimmed pearl-shell from Nagir' 

' la the TooabulaiT kaubkaub (E.) u giveit u (he tenn for a pendant Rud (or a European bead, primarily 
it Bignillea a ball or sphere; kaubkaub neb ia a ring, nft=a hole. 

■ In dewribing his visit to Sue Island (Waraber) Uaogillivra? ujt: " Tbeee blaeke belonged to the Kalkalega 
or EnlaUaig tribe. ..tbeii head-qnartera are at Mount Erneit [Nagir].... The onlj omamenti warn irere the 
large ronnd pearl-shell on the breait" (ii. p. 40). Melville haa drawn an Emb man wearing one (pi. QL fig, S). 
IVAlbertiB (i. p. 305) flgnrea a " mother-or- pearl breaBlplate— Fly Biver, Moatta, and Hall Sonnd," judging from 
the anipending band, whieh ia fastened to two perforations in (he upper third of the ihall, the Bpeoimeti 
figured oame from one of the two former looalitiee. 

d by Google 



(pi IX. 'fig. 23) which was need for this poipoee. Similar pearl-sheUfl were worn by 
some of the Western waniore when on the war path (v. p. 311), and, according to a MS. 
note of Mr Wilkin'e, "many were inacribed by sharks' teeth with the owner's augud" 
(totem). A shell (fig. 60) in the Oxford Museum ia inscribed with a kaigaa, the shovel- 
nosed skate (Bbinobatis), which is a common Western totem (v. pp. 164 — 5, 164). 

Fro. 09. CjitcoA shell pendant. 

Fia. 60. Engnved pewl-iheU 
pendftot ; J luit. VEe. 

Creecentic chest ornaments made of pearl-shell are in common use throughout 
British New Quinea; by all the Torres Straits Islanders they are termed mat, which 
is the same name by which the pearl-shell itself (Mele^^na margaritifera) is called. 
Frequently the Western Islanders call the ornament danga max {danga mari, in Muralug 
and Moa) or "pearl-shell tooth," from which term it may be suspected that the shell 
was originally carved to represent a boar's tusk, or perhaps two tusks fastened together 
by their bases. 

The crescent is generally broad, kemerkemer max (E.), "filled up mat," but may 
be quite narrow (pi. IX. figs. 20 — 22). I was informed in Mer that the kemerkemer mat 
was worn when fighting and that a small one, pek mat, was worn in the dance, hA. 
Usually the contour is regular, but there may be a slight projection in the centre of 
the concavity and conveiity (pi. IX. fig. 22) or two divergent spurs on the latter (pi. I. 
fig. 6). Sometimes they are quite plain, but usually the margins are incised with closely 
set fine lines, the concavity in one or two specimens is marked by a row of small 
pits, or these nuy occur all round the object. Very ruely there is a central design. 
There are generally two, rarely three, holes bored near the concavity at a variable 
distance apart, by means of which the ornament is suspended, the smaller specimens 
may have but one hole. The largest one I have seen measures 173 mm. (6} in.) in 
direct length (pi. I. fig. 6). They usually vary firom 140 to 158 mm. (6i to 6i in.). 

The most characteristic chest pendant is the dUndibi^ (pi. Vlll. £ga. 1, 3; fig. 61), 
a circuhtr polished disc made fix>m the flat end of the la^ cone shell, wauri (Conns 
litteratns var. millepunctatus). The whole of the flattened spire of the shell is chipped 
off from the outer whorl as a thick disc. The upper and under surfaces are then ground 
down to form a thin disc, the upper surface of which is generally slightly convex uid 

> Huij of the Wsateni Iduiden ealled it dilndih or dHadib. 


d by Google 



the flattened under sur&ce retains the spiral of the whorls. The ornament ia subsequently 

polished. In some specimenB the grinding is continued until both surfaces are planes, 

bat these are not so handsome as those of which the upper sur&ce is slightly convex. 

A well-made dibidibi is a beautiful object. The margin is 

usually lefl entire but occasionally it is nicked as in pi. IX. 

fig. 19; in this specimen there was evidently an imperfection 

on the upper surface and this has been rectified by inserting 

a small disc of pearl-shell. Occasionally one or two small 

dibidibi are &8tened to a necklace, as in fig. 67. The discs 

vary in size, usually running &om 60 to 80 mm. in diameter, 

a very fine specimen may have a diameter of 06 mm. (3} in.). 

The dibidibi, even more than most ornaments, except the 
vxiim or wauri (p. 56), served also as a kind of currency. 
They varied much in size and finish and had a corresponding 
value, thus no table of equable exchange can be drawn up. 
I gathered that ten or twelve d^ndibi of fair size would be 
equal in value to a large shell armlet, waiwi (W.), wauri (E.), 
to a canoe, to a dugong harpoon, or to a wife. Three or four 
dibidibi would constitute an annual instalment for a canoe 
(flee section on Currency and v. p. 296). 

When I was in Mer in 1880, I obtained a pendant chest 
ornament which was usually made &om pearl-shell, max weapu, 
but sometimes from the black shell of a Pinna, matib weapu. 
sexes. After a great deal of trouble I found that weapu was the name of the larva 
of the ant-lioo (Hyrmelion)'; this then gave me the clue as to the meaning of the 
ornament. One or two of those that I collected were of better workmanship than the 
others; they were made of pearl-shell, and were similar to that drawn in fig. 62b; 
others were roughly cut and carelessly finished off. It is permissible to re^ud the 
former as being more or less original designs, and the latter as unintelligent copies of 
them, which is the usual fate of copies. Now, the former may veiy well be conventionalized 
representations of the larva of an ant-lion, as may be seen on comparing A and B, Sg. 62. 
The long jaws, the three pairs of legs, and oval body are indicated; the latter, too, 
has transverse lines suggestive of segments. In some of the pendants the body and 
appendices are completely margined with short lines, which are evidently meant for the 
hairs of the insect. I have sketched four examples of degeneration which require no 
further comment except the last, which is doubled, and in this respect I believe is 
unique. Another specimen, 64 mm. long, is shewn in fig. 57. It is made of pearl-shell 
and has a lateral spur on each side of the orifice for suspension. 

The history of this ornament is not less interesting, as it offers a good lesson of 
the caution that is necessary in dealing with such subjects now that facility of tnmsport 
' Thii Uttk ii Toy bui; uid poweuM %a immeiiM pur of jawB (flg. 63 1). Tb«7 are oommon od the 
und-beaoh of Uer where thej make forroir-lilie pit.faUi, at the bottom of whioh the; bury themselTef, with 
but the tips of their jaws projeutiDg above the land. Unwai; ante tnmble into the pits and are promptly 
■eiied in the jaws, and laoked Azj b; the voraoioaB lam. Wt-apu raeaas und-mother; there are other ezamplea 
of tbii nae of (he word opu, tbtu a be« it itan-apit, wax or honer.mother ; a shrimp ie mtg-apu, tide-mother. 

Fia. 61. Dibidibi bom Uar^og, 
8Smm. in greatest diameter. 
Th« neoklaoe ooosists of a 
simple plait with appUqni lig- 
laga of yellow orohid skin 
(1888, A. C. H. Ool.). 

These were worn by both 

d by Google 



has led to intercourse between peoples who a few years previously had never heard of one 
another. Owing to the interest which I took in these objects I elicited the information 
that they were first made by a native of Tanna, one of the New Hebrides, who waa 
shipwrecked on Mer several years before my first visit to that island, and that the 
Hiriam had copied his original design. 

Fia. 63. Beries of {Mnduita in the BritiBb Uaaeam, oolleoted by the ftathor in Uer. Id all cmw 

it TOm with the perfontcd end npiMrmoat. Above (A) is a ibetoh of the Urra of an uit-lion, aboat natanl 
lite. The pendants aie all of peatl-ihcU, eioept E, irhioh ie made of the black ehell of the Pinna ; two-thiids 
of the natml Bise. 

Other pendants of animal forms were commonly worn, those in the western islands 
were totem badges (T. pp. 164 — 169, pi. XI. figs. 4 — 7), while thoee in the eastern 
islands probably had the same significance originally, though toteniism has for some time 
been obsolete in these islands (vi. p. 256, pi. XXIV. figs. 4, 6, 7). 

I cannot say whether the pearl-shell pendant in the form of a tree-frog, goai 
(pi. IX. fig. 24), or the turtle-shell fish, totoam (Platax vespertilio. Sea-bat), 74 mm. long, 
shewn in the Album, l. pi. 335, No. 8, had any special significance. I collected them 
in Mer in 1889 and they are now in the British Museum. 

Sometimes part of the totem animal itself was worn, as in the esse of the crocodile 
scutes which we obtained at Saibai (v. p. 166, fig. 14). 

Among various pendants that were worn may be noted two pearl-shell objects 
(pL IX. figs. 26, 26) from Mer, and in the same island a gerer tnoder (Pandanus mat), 
a small square of plaited gerar, was sometimes fostened to a necklace. 

The red beans of the Mucuna, wadai (W.), wada (E.), were frequently attached to 
necklaces or used in conjunction with pendants; a ceremonial pendant of these beans 
is described in Vol. ti. p. 292, Tassels of whole or more usually halved Coiz seeds, 
ku8 or kusa (W.), kusu (E.), were frequently employed. 

A racquet-shaped chest ornament (179 mm. x 97 mm.) was obtained in Mer which 
is made of kerosene tin, thickly coated with beeswax, on which are crowded as many 

d by Google 


crabs'-eyee eeeds (Abroa precatorias) as can be placed on both sides. The broad part 
is perforated in an irregular manner, the main idea being a central bar from which 
croes bars proceed to the rim. It was suspended round the neck by a piece of calico, 
and a small white cowry ia placed above the pendant. 

The most beautiful of the personal ornaments are the remailmble perforated pearl- 
shell discs which were the distinctive badge of the Qeregere le of the Malu fraternity 
of Mer (vi. p. 295). I know of only three specimens. One (pi. X. fig. 16) presented 
by Mr R. Bruce to the Kelvingrove Museum, Glasgow, is about 120 mm. in diamet«r, 
another fph X. fig. 17), obtained long ago by the London Missionary Society and acquired 
by the British Museum, is about 116 mm. in diameter, and the third (pi. X. fig. 15), 
collected by ourselves in Mer and now in the Cambridge Museum, is 110 to 117 mm. 
in diameter. Although these objects diflfer fi^m each other, it is evident that the designs 
are essentially similar. In fig. 16 there is a central star (but in figs. 15 and 17 there 
is a perforated disc, &om which radiate a series of tridents), this is succeeded by a 
ring of tridents, which is followed by a ring of circular perforations; in figs. 15 and 17 
there is another ring of tridents and a mai^inal one of circular perforations. Con- 
centric bars or rings delimit each series of patterns. Figs. 16 and 17 have a symmetry of 
eight, while fig, 15 has a symmetry of six. 

In fig. 16 the 8 rays of the central star are continued as 8 tridents, intermediate 
to which are 8 Y-shaped bars, making 16 elements in the ring; the next ring contains 
eight groups of 3 tridents, the bases of which practically spring from the forks of the 
inner tridents, each set of 3 being separated by a single bar, m^dng a total of 32 
elements in this ring; there are 48 perforations in the marginal ring, each of which 
corresponds to a space between the forks of the outer ring of tridents. 

Fig. 17 contains 8 tridents in the first row, and 16 in the second; the next ring 
contains 32 perforations, and is followed by ?30 tridents (the number should be 32, 
to be symmetrical); there appear to have been 46 marginal perforations. 

In fig. 15 there appear to have been 6 tridents in the first row and 12 in the 
second ; the next ring contained 24 perforations, followed by 24 tridents ; the marginal 
perforations were probably 48 in number. 

I have given this analysis of the arrangement of the elements as it is of some 
interest to determine how accurately a somewhat complicated design can be cajried out 
by natives in a low stage of calture. 

I do not know of ornaments simihir to these fix)m any other locality, and they 
do not appear to me to be very charac- 
teristic of the decorative art of the Torres 
Straite Islanders as there are no other 
examples of the technique of nacre fretwork. 
Fretwork in turtle-shell is characteristic of I 

masks from this reirion, but the trident 

, ■, , Flo. 6S. Fret uttem on m mtak ; Vienna Uos. 

motive, BO mr as I am aware, occurs only 

in a turtle-shell fret pattern (fig. 63) on the side of a human fece on a crocodile-mask 

in the Vienna Museum (No. 53387). This mask is also provided with a band and is 

surmounted by a made-up bird. On the whole I am inclined to look upon these objects 

d by Google 


as importations, but I cannot at present suggest from which of the neighbourii^ districts 
of New Guinea they might have come, for they certainly did not come from Australia. 
According to tradition (vi. p. 282) the Oeregere le are said to have come from Tuger; 
the geregvre is a small migratory bird (ti. p. 287). 

The following pendants formed part of a bride's decoration in Mer (cf. Vol. vi. p. 114), 
the number and variety depending upon the wealth of her parents. They were worn 
for one or two months before the wedding feast. The older married women also wore 
many of these objects on special occaaons, but never during widowhood. 

Fio. M. abell pendaintB worn b; brides. No. 1 is llSmm. loag, 3 ia STm 
The tabagarar al no. a ia 100 nun. long. Nob. 1 (95 mm.) and S {107 mm 
othera we in the Cambridge Hnsenm. 

Triangular pieces from the outer whorl of the wawri shell are called o, waun o, 
or kaakau (hanging or suspended o) in Uer (fig. 64). They are worn only by married 
women, no single girl or woman may wear them. Most are triangular in form, and 
are suspended by means of a single orifice in the base, very occasionally the hole is 
at the apex. They vary greatly in finish, some being excellently made, while others 
are very rude ; some very rough ones are more or less quadrangular in outline. Occasionally 

d by Google 



they are slightly decorated with simple patterns. Two, or at moBt three, could be made 
from a large wauri shell; a cone shell from which one has been made is shewn in 
the section on Carrency. There is great variation in siae, hut they naay be considered 
as aven^;ing about 8 — 9 cm. in length and about 6 — 6 cm. in breadth. 

The fish-bone pendants (fig. 65), naged lid or tar Ud (naged bones or fish bones), worn 
by brides and married women (vi. pp. sii, 114), are in some 
cases the remarkably modified intemeurals and interhiemals 
of Platax arthriticuB. Those shewn in fig. 65 a are said to 
be the bones of the ab, a large blue fish. 

The most interesting and varied pendant was that known 
as gagi (W.), mbagorar (E.). Of late years well-made ex- 
amples have been very raxe, the few specimens we were able 
to obtain in 1898 are of very rude workmanship and many 
are undecorated. The laigest and finest specimens consist of 
a crescent of turtle-shell with the horns produced into an 
incipient helicoid spiral and a central shank perforated at 
the end for suspension (pla X. fig. 14, XI. fig. 9). There 
is usually a second hole near the base of the shank, and a 
third in the middle line of the convexity of the crescent. 
At this point is a pair of divergent spurs, and another 
pair occurs on each side. Incised patterns forming a band 
traverse the crescent opposite the three pairs of spurs, 
and there may be a little ornamentation on the shank just 
below the uppermost perforation. The four finest and best 
decorated specimens known to me are in the Liverpool 
(1. 10. 85. 16 and 17), Glasgow (89. 67. ck) and British 
Museums ; the first three have respectively a total length and 
breadth of 146 x 119 mm. (pi. X. fig. 14), 143 x 179 mm., and 
170 X 173 mm. The shank in the last specimen is very long; 
it was obtained at Mer by Mr B. Bruce. The example which 
I obtained at Mer in 1889 (pi. XI. fig. 9) and gave to the 
British Museum, measures 146 x 138 mm.; it was made by 
Qadodo's &ther. I obtained at the same time a somewhat 
smaller specimen (pi. XI. fig. 10) in which the horns are 

continuous with the shank (1S8 by 140 mm.). In smaller specimens the horns, as in 
the last mentioned, are in the same plane as the rest of the ornament, and they exhibit 
greater diversity of form. The specimen collected by me in Mer in 1888 (pi. XL fig. 7) 
is shaped like an anchor, but it is evidently connected with the former type. In a 
specimen figured in the Album (I. pi. 338, No. 1) two hooks appear to spring from 
a common shank, and this is more evident in the rather rough specimen shewn in 

Fill. 6fi. FUh-bone pendAnta worn 
by nuuried women ; Met. A and 
B ftre i nat. aze and are in tli« 
Ounbiidge HoMom. is in (he 
British Moaeam. 

The great majority of specimens are hook-Hke objects, decorated or quite plain, 
and frequently with single or double spurs. Some are narrow and may be quite flat 
or slightly rounded, others again are broad, sometimes extremely so. The book may 

d by Google 


be a simple bend of the efaank, or it may form a large recurved loop. The workmanship 
may be of fine quality, or the object may be coarsely faafaioned, of uncouth form, and 
without ornamentation. The variationa are so great and 
individual that nothing can be gained by a more detailed 

As was previously mentioned, various objects were worn 
by girls in Mer as part of their marriage outfit; I am not 
sure how far this custom extended in the Straits, but I think 
we can safely eay that it reached its maximum in the Murray 
Islands. A complete set of a Miriam girl's betrothal outfit, 
besides such trinkets as dibidibi, nose-sticks, small cowrie and 
ovulum shells, strings of kua (Coix seeds), necklaces of shells 
and especially of dogs' teeth and so forth, would include 

triangular pieces of polished cone-shells, o, turtle-shell bodkins, ' Bimm. lotm ' ' 

(*r (pi, XI. figs. 11 — 17), turtle-shell fish-hooks, mekek, and 

Sfdfofforar. Some of these objects are purely decorative in character; othera such as the 
bodkins and fish-hooks are useful. The fish-hooks were used in pairs, each being fastened 
to the end of a thin string; as they are slender (pi. XI. fig. 1) they admit of very 
little decoration. One would suppose that if it were attempted to ornament them, they 
would have to be so much broadened in order to take a pattern as to render them of little 
use as fish-hooks, the object shewn on pi. XI. fig. 3 may however be a decorated fish- 
book. The desire for decoration, combined with the traditional wearing of fish-hooks, 
led, according to this view, to the adoption of a purely 
ornamental hook. Once this step was gained the further 
development into a large and handsomti ornament could 
be readily accomplished. Analc^us examples of the evolu- 
tion of the forms of objects are everywhere to hand. The 
duplication of the hook to form the anchor-like variety 
is also a perfectly natural development. Two fish-hooks 
lying back to back, as they often would do when hanging 
down a girl's back, would suggest the apposition of the 
ornament; and two similarly placed simple sahagorar 
(127 mm. in length) are seeu in pi. XI. fig. 8. It is only 
a short stf^ to make one ornament instead of two fastened 
together. The specimens shewn in fig. 66 and pi. XL fig. 7 
are very suggestive of this supposed transition. I do not 
know what is the significance of the spurs. The specimen 
of fig. 10, pi. XI. is clearly a degradation of such a form „ 

as fig. 9. If my hypothesis be correct, we have here a ^^ t^«.ri«i]. M.r; Brit. Mn.. 
primitive form (the fish-hook), intermediate types, the Total length 8« mm.; length of 
highest specialised derivative and its degenerate modifica- turtle-Bhall 97 mm. 
tion, all being worn at the same time^. 

In 1889 I obtained at Mer a necklace (fig. 67) of ten warns, two dibidibi, and 
1 This hypothMia wm firat pabLibed la mj Dteorativt Art of Britith Ntis Quiiua, 1S94, p. 85. 

H. ToL IV. 7 

d by Google 


four marginal pktes of turtle-shell to which last the Dame of sabofforar wsa given, 
although they differ entirely from the manufactured ornaments of that name. It formed 
part of the marriage decoration of a bride, but doubtless was worn at later festivals. 

Tia. 68. Sauad, kn artifioUlly deformed boar's tiuk Fto. 69. 3ava4, idth tell]' notohea, kod 

Tom ft* an otnunent, with two pigs' t»ili, Mer. two piga' tula, Her. 

The aauad, which were worn at the conclusion of the initiation ceremonies by the 
important men of the Bomai-Malu fraternity of Mer, were very valued objects. They 
are tusks from the lower jaw of a boar, whoae upper tuska have been knocked out; 
the lower tusks, thus having nothing to grind against, keep on growing, and in time 
would grow into a complete circle. The animal is killed when, or just before, this has 
taken place and the tusks extracted; the base is then pierced and they are suspended 
from a thin necklace made of twisted coco-nut fibre. Two savad that we collected had 
attached to them a couple of pigs' tails and one had three; it is possible that these 

d by Google 


were originally always present. One of our specimens (fig. 68) forms a complete circle 
10 cm. in diameter. Another is a larger tusk, 10*7 cm. in diameter (fig. 69), but the 
circle is incomplete, its chief interest lies in eight short incised lines near the point, 
which are tallies memorialising amatoiy adventures (Vol. VI. pp. 261, 283). 

The sauad were of great value as their production was not an easy matter and 
they had to be imported from New Guinea. Perhaps it was for this reason that imitation 
ones, nasir lauad, were made locally out of laige Trochus shells, nasir. One very fine 
specimen (pi VII. fig. 5), now in the British Museum, was collected by me in 1889, 
it is 126 mm. (6 ins.) in diameter. There is one specimen in our collection in process 
of manu&cture which is 129 mm. in greatest diameter. 

Artificially deformed boare' tusks are greatly prissed throughout New Guinea and 
Melanesia generally', and imitation ones are cut out of the giant clam (Tridacna) in 
the south-east archipelago of New Guinea. , 

1 O. Finich, "AbiionDe EbeiluiiMr, Pretioaen in aihrnaok der BildBee-YOlker," Mitt., Anth., 6«i«U. Wien, 
XTH. 18S7. 


d by Google 


Other pendants of boare' tusks are described in the section dealing with war 

Also connected with the same ceremonies (tl 294) was the custom of wearing a 
human lower jaw; one specimen (Tl. pi. XVII. fig. 8) collected by us was provided with 
a thick fringe of Coix seeds, with some small sbellB intermingled among the seeda, and 
a loop of cord for suspension round the neck. 

A Miriam widow, when in mourning, wore many objects, bud lu, suspended &om 
necklets, these are described in Vol. Tl. pp. 157, 168. They consisted of talliea of the 
length of the nose, and models of certain fingers and limb bones of the deceased, these 
models may have been substituted for the actual objects, but the dried tongue, palms 
of his bands' and soles of his feet were actually worn (fig, 70), in addition to various 
objects (such as armlets, leglets, groin shell, or ornaments) worn by him, or implements 
or nkaterials which he had recently employed. Fig. 71 illustrates a pit totiar (nose 
foshion) or pit autare (nose measure), 95 mm. long, filled with shreddings of the deceased 
husband's clothing, instead of the usual hair, and decorated with bead tassels, lamar 
kSrup or tarpor ktrup, together with a boar's tusk which the husband used as a scraper. 


Crossed shoulder-belts or baldrics were worn as part of the accoutrements for war 
and for dancing. The simplest were merely strips of coco-nut leaf, tu (W.), «u (E.) 
(pis. I. fig. 6, V. fig. 10), or of pandanus leaf, others were like ordinary belts, but with 
the two ends joined. They are called kamadi or naga (Muralug), maiei or moie (Tutu), 
naga (Mabuiag), wagogob ^ wak, belt, gogob, ring of rope or string (Mer). The Miriam 
mourning wagogob (fig. 70) was made from the bark of the aem tree (Hibiscus tiliaceus). 
Some naga from Mabuiag in the British Museum have a length of about 1 m., the 
plaited band being covered with two or more rows of Coix seeds. Coloured wool or 
yellow orchid skin may be sewn in as chevrons as in ordinary belts, and tassels of Coiz 
seeds may be added {Albttm, i. pi. 341, No. 4). 

Belts, vxiJcau (W.), wok (E.), are worn by both sexes, but much more frequently 
l^ the men. The narrower kinds are of local manufacture, but the broad belts, which 
consist of a plain check foundation with designs in appliqu^ picked out in colours and 
furnished with a fringe, are imported from Kew Guinea. The technique of the various 
kinds of belts is dealt with in the section on Textiles. 

The following descriptions refer to the first type of belt. Frequently they are 
decorated by overlaid wefts made of the bright yellow skin of an orchid (Dendrobium) 
(fig. 72 D, E). To some Coix seeds are attached, but this is more usual in the baldrics. 
We collected several to which were fastened the mouths of cowries (Cyprrea annulus, etc.), 
uxa (W.), pet (E.), hence these are termed pei uiak by the Miriam. A firinge of shells, 
usually Barbatia, to serve as a rattle was often added to the dance belt, tepe wakau (W.), 
1 Brookott in hit Narrativt of a Voyage, eto., 1886, Myi: "In some of the hnU [of Met], *b saw akini of 
liucU vhlob were bMiging ap; thMa the natives wear m o>-.uneDte □□ the daji of lejoicins" (p. 26). Id the 
Naia. Stag. Tt. 1SS7, p. 7M, it ia itated thkl the ikios of 'Jie hands of departed friends are worn b; the Uiilam 
women "on feitive or foooral oocaaions." It waa onk^ when in moDraing that a Mwiam widow wore theM 
objeota, they were not worn b; anyone else, moreoTer moomera did not attend feative gatherings. 

d by Google 


aerpa wok (E.). I was iaformed that terpa was the general name and tik or tig the 
special name for these shells. These belts vary in length from about 80 cm. to a little 
over I m., but about 90 cm. is a common length. 

Fio. 73. Sb«ll belt^ Mer. A— O, pet teak; D, E, $erpa vak. 

The width of the bands with one row of cowries (fig. 72 a) varies from 9 to 20 mm. ; 
those with two rows of cowries (figs. 72 B, c, 88) from 23 to 35 mm. ; the one with 
three rows is 55 mm. In some cases the fringe of shells is attached to a narrow band, 
serpa wak, which is fastened on to the pet wok (fig. 72 B, e). Occasionally the plaited 
band is covered with some European stufiT, probably originally of a bright colour, on 
to which the cowries are bstened. In one belt (fig. 72 c) the spaces between the shells 
are filled with tuils or rosettes of red or black wool. 

All these belts are worn by the Miriam men in the ordinary dance, kab; the terpa 
took making an agreeable rattling noise when the wearer is dancing. 

In a folk-tale written down by Paai (VI. p. 62) a poto wak is mentioned, which, 
fix>m its name, may be a belt decorated with opercula, but it may be intended for 
pet wak, as I have not seen a belt so decorated. 

The kiu wak was a belt (fig. 73) that was worn by the kiai (or lads during 
initiation) at the Bomai-Malu ceremony (VL p. 292). It was composed of whole Coix 

d by Google 


seeds, between eveiy dozen or bo of which waa threaded a portion of the antenna of 
a cmjfiah, kaier (Palinurus). In one of the apecimena made for ue the antenna-beads 

Fie. 78. Stnuid of a kiu vak. Her. 
are replaced by tubes of the organ-pipe coral (Tubipora musica). A belt was composed 
of several of these strands, and there was a side tassel composed of loops of the stnuid 
and five wada beans. A similar strand composed of Coix seeds and antenna-beads, 
about 2'5 m. (8 ft.) in length, is in the British Museum (Kennett ColL C.C. 6931). 
It is called a " necklace " (which is improbable), and is said to have been used at Cape 
York and Mount Ernest (Nagir). 

Objects attached to Belts. 

The groin shield was worn when fighting (it is fully described in the section on 
Warftire) or when dancing (pi. VIII. fig. 2) but not on ordinary occasions. Macgillivray 
says (It. p. 15) that at initiation the lads among the Kowraregas [natives of the Muralug 
group] are "decorated with a profusion of ornaments which are worn until they drop 
off, and wearing in front a small triangular piece of shell as a distinguishing mark " 
(cf Vol. V. p. 217). I regret that I have not confirmed this statement, but I think 
1 should have heard of it, had this custom occurred elsewhere; at all events at Mer 
I obtained two small specimens, 121 x 82 and 116 x 63 moi., which are labelled as "boy's 
ebeneattp," and also one imperforate one of the shape of an isosceles triangle, 144 x 60 mm., 
the upper portion of which is decorated with crab's-eye seeds; it was probably a toy. 
It is possible Macgillivray alludes to the ordinary groin shield, but it seems probable 
that the objects of which Gi'om informed him were smaller and in that case would 
resemble the o described on p. 47. There is no reason to believe that women ever 
wore any object similar to a groin shield as a covering, certainly they have never done 
so since they came under the notice of Europeans. 

There is in the Museum fllr Volkerkunde a belt (fig. 74) consisting of six pearl- 
shells fastened to a double chord which evidently was worn round the waist. One of 
these shells has an engraving of a sucker fish, gapu, and another has a turtle and two 
sharks. There is little doubt that this came from one of the western islands, and it 
was probably worn ceremonially. 

When dancing, there was inserted at the back of the belt a large tuft of cassowary 
feathers, nadua, kaba-nadua, satnera or sauma (W.), kolber kolher (E.) (pis. VIII. fig. 6; 
XXIX. fig. 1 and Vol. Vl. pi. XXVI. fig. 1) or a bunch of leaves, those of the croton, 
w« (E.), or other brightly coloured leaves being in most favour. 

I obtained in Tutu in 1888 a large ktJni'nadua which consists of a narrow mat, 
over which is placed a thick bundle of the "grass" of which the Mawata petticoats 
are made, both are tied together to form a stem which is wrapped round with calico 
{Allmm, 1. pL 343, No. 3). The free portion of the mat measures 11 x 40 cm., it is 

d by Google 


a Bimple plait and ite outer face is completely covered with a series of horizontal fringes 
of shredded banana leaves. The total length of the specimen is 685 mnL (27 in.). 

Fio. 74. B«lt ol pewl-Bboll*, fietUn Hasemn (ti. 657). 

Abmlkts and Leqlets. 

A band, vmsn/r (W.), put (K and occasionally in the W.), encircling the upper arm 
was formerly constantly worn by both sexes, but more frequently by the men (pis. I,, IL, 
in., v., VIII.; Alhwm, i. pi. 341, Kos. 7, 8, 14); there was usually one on each arm. 
Small objects were carried in them, and sprigs of scented plants, brightly coloured leaves, 
and flowers were frequently inserted therein. They are narrow plaited strips of ratan, 
bamboo rind or similar material (pi. XIV. fig. 4), or narrow or broad plaited bands 
(figs. 92 — 97 ; pi. XIV. figs. 2, 5—8) which in Mer are made from the fine mid-rib, 
he lid, of the coco-nut leaf Mr Bruce says that the Western Islanders make armlets 
from a fibre or vine, ferey, which they get from New Guinea; the same material is 
used by the Miriam for making the kadik worn in dancing. 

The broad flexible armlets were often ornamented with Coix seeds or the mouths 
of cowries ; under European influence beads, brightly coloured 
worsted, bits of calico and the like were frequently added 
(pi. XIV. figs. 3, 6—8). 

I collected in Mer one or two armlets, gir put, made 
of boars' tusks, gir. Fig. 75 represents one, obtained from 
Maroai, made of two tusks fastened together at each end 
by turkey-red twill with five nicks near the apex of one 
tusk which are tonar eniam neur emeret (fashion steal girl 
olden time) (VL pp. 251, 295). An artificially deformed boar's 
tusk that I gave to the British Museum has a wrapping 
which unites the two ends, and it is probable that it was Fio.TS. Armlet mode of two !><»»' 
converted into an armlet bom being a pendant. tnskswith UUynowhes; laom. 

Arm-rings were sometimes constructed out of the large ' ""' 

top shells, nofi or nasir (Trocbus niloticus), they were called iMsir. 

In the British Museum (6529) is a "bracelet" made of a human lower jaw, the 
articular ends are connected by a stiff string cord and are decorated with cassowary 

d by Google 


feathers, to the symphysis are tied a feather aad an ovulum shell. A decorated staff 
was " used with the jawbone bracelet," which indicates that the latter was employed 
ceremonially. They came from Bobo (Brietow Island) which 
lies close to New Guinea. 

The most precious arm-ring (fig. 76) was that made 
from the large cone-shell, wauri (Conus Utteratus var. 
millepuDCtatus). It was called waim by the Western 
Islanders and wauri by the Eastern. Even small specimens 
were of value, but one large enough to go on the arm was 
the highest unit of exchange in Torres Straits (p. 44 and Fia. 76. Cooe-shdl um-riag. 
Vol. V. pp, 230, 293; Vol. vi. pp. 185, 186). These objects 

are also greatly prized in various parts of New Guinea and are sometimes ornamented 
with seeds of the wild banana, or in other ways, but so &r as my experience goes 
the Torres Straits specimens were quite plain. I was 
informed in 1889 that only one man in Mer made wauri. 

A wristlet, perta urukam (W.), was sometimes worn, 
as in the initiation of lads at Saibai. The tiapuru (wrist 
string) was a special string wristlet which was plaited by 
a young Mabniag man and sent to the girl whom he 
desired to many (T. p. 223). 

The arm-guard or bracer, kadig (W.), kadtk (R), was 
worn on the left forearm exclusively by men, it formed an 
essential accoutrement in war and was almost invariably 
worn when dancing (pis. V. fig. 10, Vllt. fig. 1, XXIX. 
fig. 1). As is stated in the section on Textiles, there are 
two kinds of bracera : (I) of strong, rigid, coarse plaitwork, 
apparently made of ratan ; this is the ordinary form (pi. 
XIV. figs. 9, 10) which was used in fighting and was 
also worn when dancing; (2) a rare type only worn in 
dances, of which I saw two specimens in Tutu, one old 
specimen imported from Daudai is in the British Museum, 
the other is shewn in pi. V. fig. 10, and we collected one 
or two, iinp leadik, in Mer (pi. XIV. figs. 11, 12). It 
consists of a continuous series of long splints bound to- 
gether by twined weaving. Mr J. Bruce informs me that 
the kadik is made from the tereg vine, and from boz, 
Flf^llaria Indica. Those made from he lid, mid-ribs of 
coco-nut leaves, were the most pliant, and the tereg vine 
makes a more pliant kadik than the boz and is used in 
preference to the latter when procurable. The kadik 
are made by men only. The Miriam zogo kadik (fig, 77, 

see VI. p. 295), which was purely ceremonial, was con- p,^, „_ ^^^^ ^odii, Mer. 

structed in a similar manner. It is made from the stem 
of the puar or poar, an epiphytic aroid (Vocab.), a parasitic vine that grows on coco-nut 

d by Google 


palms and other trees, like an orchid with a white flower (J. Bmce). This kind of 
arm-guard ia also found in the Fly river district of New Guinea. 

On festive occasions a plume of caasowary feathers, paupuaa (W.). tagelu (E.), was 
frequently inserted into the bracer (pi. VIII. fig. 1). It consisted of a stick on to which 
were tied numerous cassowary feathers ; the total length in one specimen is 69 cm. (27 in.). 

Flo, 7S. Poor tagi tu Irom Her. 

A remarkable ornament called kadig tam, arm-guard branch, or sometimes kadig 
tang (W.), tage la, forearm thing (E.), is inserted in the arm-guard for the war- and 
other dances. In its simplest form it consists of a single or a double loop of a strip 
of ratan or bamboo rind, the free ends of which are tied together and bound round 

H. Vol. IT. Q 

d by Google 


with string or strips of calico, thus forming a handle or shaft, which ie infierted into 
the arm-guard, the loops projecting tar beyond the elbow of the wearer (pis. V. fig. 10, 
XXJX. fig. 1). Two strips are also usually added to the loops, the ends of which are 
often bent round and twisted round themselves and then back again so that the free 
ends are once more terminal, the double twist forming a fairly regular series of over- 
crossings OR the central strip. The loops are usually ornamented with feathers or shreds 
of calico; not unfrequently the free ends of the two strips are tipped with a bunch 
of cassowary featheis. The decoration may be even more lavish. 

On first seeing these objects in 1888 I was puzzled for a long time as to what 
they could signify, till one day, on re-reading d'Albertis' book on New Guinea, I came 
upon the following passage relating to some " men of the coast near Kataw " of the 
Mawata district on the mainland of New Guinea: "I remarked no ornaments, except 
the bracelet worn to protect the arm fii>m the bow-string. They use this also as a 
bag or purse, and put tobacco or a spare string for their bow, and other little things 
in it" (II. p. 173). On re-examining the specimens I had collected I came to the 
conclusion that the kadig tarn represents a spare bow-string, which has been modified 
into a functionless dance ornament The two loops and the two tree ends were originally 
one doubled-up bow-string, but the ornament is now built up of several* elements. 

In 1898 I found that this ornament had undeigone further modification. The four 
examples from Mer (fig. 78) will suffice to illustrate some of these variations. Most of 
the tage lu which we collected in Mer are bound with coloured wools. A belonged to the 
Mamoose of Mer, it has a total length of 77 cm. ; the four diamonds threaded on the 
loops are a new feature; the black rod consists of cassowary feathers tightly bound 
together. In c there is an addition of two small loops, the single rod is deflected in 
the drawing; it is 63 cm. long. B consiste of four loops and four rods; it is 68 cm. loog. 
D more nearly reverts to the original type as it consists of two loops and two rods, 
which however are not all in one piece, as they should be ; it is 56 cm. long. 

Finger rings, getau gugain (W.), tag makamak (E.) {&g. 79), were plaited from the 
surface roots of the coco-nut palm, which were prepared and 
plaited by both sexes. The rings were worn as a decoration 
without distinction of age or sex, on any or all of the fingers, 
but not on the thumbs; as many as five rings might be 
worn OQ a finger. People might wear them on uiy occasion, 
but they took them ofiF when working in the gardens or going 
out to fiah or swim. Rings were handed down as heirlooms in 
a family. This is the Miriam custom and doubtless it was ^'^J^' ^^^a^^J"^' 
general throughout the islands, but, as I said in 1890 {J. A. 

I. XIX. p. 370), I never saw any in the western islands, and no one had previously 
mentioned them. Finger rings are now sometimes made from turtle-shell ; they may 
have been introduced by South Sea men, at all events the Samoan teacher on Mer 
gave me some which he had made. 

Fringes, biBuab or biswab (W.), bisiwam (E.), of shredded sago leaves usually dyed 
red were worn on the arms and legs when in mourning (fig. 70, see also v. p. 262, 
VI. pp. 157, 168). The long fringes on the upper arms and below the knees were called 

d by Google 


hy the Hiriam tag put and teter put respectively, and the shorter tHoges on the wrists 
and ankles tag mtw and teter mue. The butwom shewn in fig. $0 is a fringe worn 
on the leg by Miriam men when dancing. Each fringe consists of a string 74 cm. 
(29 in.) long and doubled round at one end, in the middle of which for a distance of 
215 mm. (SJ in.) is a row of tufts, alternately red and brown, forming a fringe 40 mm. 
deep. These are all imported from New Guinea. 

Fio. 80. Biritnim, Her. 

Fto. 81. A, bent view, uid B, b«ok view of method of tying on the tafti of a bitivam. 

The legs of the natives were but simply decorated. The makamak was the universal 
leg ornament; it consisted of a bundle of thin plaited rings worn above the calf and 
below the knee (pis. III. fig. 3, V. fig. 10, XXIX. fig. 1); sometimes a very large 
number would be worn, 130 have been counted on the legs of one man'. They were 
worn on all occasions and at any time by the men but only when dancing by the 
women. They are made of the same material and in the same way as the tag makamak, 
and like them were handed down from generation to generation. 

Leglets and anklets, bilrua or brua (W.), are made of the sprouting leaf of the 
coco-nut palm, tu (W.), «m (E.), for temporary use in the dance (pis, V. fig. 10, VIII. 
figs. 4—6). 

The dani kuk or dani kukur and dani makamak (W.) are anklete and leglets made 
of dan(%) fibre (Ficus). 


Except for the ornaments they wore, or the special costumes for war and for ceremonial 
occasions, the men went entirely nude. I believe that in the secular dances the men 
wore an extemporised petticoat made from the yellow immature leaves of the coco-nut 
pahn, at all events I have frequently seen this done (pi. VIII.). 

At the present day they wear a variety of clothes, the simplest costume being 
a waifitcloth of white or coloured calico; this is usually combined with a thin cotton 
vest or singlet. Increasingly the men take to trowsers, shirt, and coat When they wish 

> A preoiael; liDilar ornamsat is worn in tbe ume manasr by muiy of the interior trib«e of Borneo; it ii 
•tiled utiu uid ia made from a wild palm, aping (Sea Dayak), limak (Kayan). 


d by Google 


to appear "flash," noidi or zantiak (W.), peror (E.), they wear a &Dcy shirt with tumed- 
down collar and a brightly coloured necktie, a pair of trowsers, a jacket, and often 
a helt or a gaily coloured cummerbund, a straw or felt hat. and occasionally boots 
or shoes. 

The war accoutrements and dancing costumes of the men are described under their 
respective headings. The groin-shell must be regarded as armour though it was frequently 
worn in dances. The coco-nut palm-leaf petticoats or other leafy coverings worn by men 
when performing certain ceremonies were merely ceremonial garments and were never 
worn on ordinary occasions, but I believe they were often worn when dancing ; descriptions 
of or allusions to them will be found in Vol v. pp. 253—8, 262, 340, 344, 348, 349, pis. 
XIV., XVm., XIX., Vol VI. pp. 156—9, 275, 289, pis. XXIV. fig. 5, XXIX., XXX. 

No cloaks were worn by either sex, nor any protection for the head from sun or rain, 
nor any coverings for the bands or feet. 

Formerly the women and elder girls invariably wore apetticoat, zasi (W.), nesar (E.), 
made of split leaves or bark fibre. Young girls wore a small petticoat when they were 
approaching the age of puberty, till then they went nude. Macgillivray (ll. p. 19) says: 
"Not only at Cape York but throughout Torres Strait the males use no clothing or 
covering of any kind. At Cape York and the Prince of Wales Islands grown up 
females usually wear a covering in front, consisting of a tuft of long grass, or flag 
(Philydrum lanuginoeum), or split pandanus leaves, either hanging loosely or passed 
between the legs and tied to another behind; over this a short petticoat of fine shreds 
of pandaaus leaf, the ends worked into a waistband, is sometimes put on, especially by 
the young girls, and when about to engage in dancing. This petticoat, varying only 
in the materials from which it is made, is in general use among the ferosles of all 
the Torres Strait tribes, except the Eowrarega [those who inhabit the Prince of Wales 
group, i.e. Huralug and the neighbouring islands], and much labour is often expended 
upon its construction." 

The petticoats were usually of ample size extending from the waist to about the 
knee and might he thin or thick, frequently two or three were worn at the same time. 
The band was generally quite narrow and ended in a loop on the left side of the 
wearer and in two strings at the other end. In most cases the fringe was continuous but 
in some of the western islands, more especially Saibai and Dauan, a gap was left in 
the middle of the petticoat (pi. XVIII. fig. 4), thus a portion of the right thigh of 
the wearer was exposed. This &shion must be put down to influence from New Guinea, 
as the typical garment of the women of Daudai and Eiwai consisted of a discontinuous 
petticoat with two veiy long fringes, the smaller one being in front; these are tucked 
up BO as to leave the hips exposed (pi. XXI. fig. 2). The ome nesur worn by Hiriam 
widows (ti. p. 167) was of the ordinary form, but the other female relatives when in 
mourning did not tuck the stmnds of the ome nemr between the legs, as the widow 
did (fig, 70), but cut pieces fix>m it at the sides so as to leave the sides of the thighs 
bare. By this means it was assimilated to the fore and aft petticoat of New Guinea. 

The different kinds of petticoats were named after the material of which they 
were wholly or mainly composed, thus teger nemr, kiaJn nesur, etc. 

The materials mtrat commonly used in the manufiicture of petticoats were: the leaves 

d by Google 


of a water-wort or flag, Philydrum Unuginosum, tagar or tagur (W,), teger (E.) appear 
to have been simply split and dried. The leaves of this plant were formerly employed 
in Australia "for the girdles of aboriginal women" (Hooker) (cf. Maiden, Vse/id Ifative 
Plants of Aitstralia, p. 629). Jukes (I. p^ 171) says that in Erub, "the grown-up woman's 
petticoat, or 'nessoor,' was formed of the inside part of the large leaves of a bulbous- 
rooted plant, called by them ' teggaer,' of which each strip was an inch broad." 

The inner bark or bast of the root of the dani (W.), ome (E.) tree, a species of 
Ficus (near to F. Cunninghamii), was beaten out; it formed a very coaree fringe, the 
fibres of which were sometimes allowed to remain felted. This rough kind of bark cloth 
was called /u gegur, bark thing. This is probably the western vuie as in pi. XVIII. 
fig. 1. It was also employed in making the nagar, the long fringe that extended from 
the neck to the knees and was foatened round the waist by a belt, worn by Miriam 
widows when in mourning (fig. 70, see vi, p. 157), 

The inner bark of the Hibiscus, vrakar (W.), kokuam (K), was used for a similar 
purpose in the western islands and also for making the toge (v. p. 262). A pounding 
stone or hammer for making the bark fringe was called mog lu, lump thing (K); I have 
not seen a specimen as the work is not now done, but evidently it was merely a smooth 
stone; the board on which the bark was hammered was called patpat lu, flat thing. 

The finely shredded leaves of the banana, dawa (W.), kaba (E.), were firequently 
employed either for entire petticoats (pi. XVIII. fig. 3) or with other material (pL XVIII. 
fig. 4), Jukes (I. p. 171) says, "The girls' 'nessoor' was made of much narrower strips 
from the inside of the leaf of the plantain, which they call 'cabbow.'" 

Other materials were: the leaves of the kiaki (E.) vine, and of the teid (E.) plant 
(pi. XVin. fig. 1). The tabom (W.) is a long petticoat, probably made of pandanus 
leaves (tarn, branch; bom, pandanus), perhaps the tt^m of Muralug is the same word. 
A pandanus (abai) leaf petticoat, abal n«gur, was also worn in Her. Pandanus leaves 
were also employed in making the itagar of the Miriam (vi. p. 157). In the Vocabulary 
petticoats are stated to be made of the leaves of bameg tree (but probably this was 
only a dye, p. 62) and of isu (W.). 

The finely shredded leaves of the sago palm, bin, were used in making the long 
fringe, aoge, pL togeal (W.), nagar (E.), worn on the chest and back when mourning 
(v. p. 262, VI. p. 157) as well as for the fringed mourning armlets and leglets (p. 59). 
They were also used in the bid, a pendant fringe representing the foetus, worn on the 
abdomen during pregnancy in Saibai (v. pp. 194, 196). 

The three kinds of petticoats worn by the legendaiy Gawer (vi. p. 26) were said 
to have been made of kud, ar and tamad (breadfruit tree) leaves, but this is the only 
record I have of these leaves being employed for this purpose. 

The dance petticoat worn by men, tu sati (W.), gu naur or i* kupi nesur (E) 
was made of the split pinnules of the pale yellow, sprouting leaves of the coco-nut palm. 

Sometimes the materials of which the petticoats were made were left in their natural 
sear condition, as for example the banana leaves, but fi:>equently they were artificially 
coloured. A red colour was produced by dyeing the material, usually dani or ome, with 
a decoction of the bark of mangrove root, which was pounded and the juice rubbed 
on the fibre. Bin (p. 58) was also stained in this manner. Dr Seligmann has given 

d by Google 


me the following note from Mabaiag: "The roote of hameg, kerikeri (Zingiber sp.) or 
keriker, aubau (Morinda sp.) and wigir were all pounded tc^ther and mixed with lime, 
mangrove juice, and salt water" for the red colour; "yellow was made in a similar way 
but leas mangrove juice waa used." Ab a matter of feet bameg and kerikeri are used 
in conjunction with the suffix adgamvlnga as colour names for yellow (IL pp. 59 — 61). 
Hibiscus bark was blackened by applying black paint, obtained by rubbing down a stone 
called kubibud (W.) and mixing the powder with water. One method of blackening 
fibres was to bury them in the black mud of a mangrove swamp. Probably the tolop, 
a petticoat made of blackened leaves wom by the xogo U of the Meket airiam (vi. p. 274), 
was coloured in a similar way. 

At the present day the women wear self-coloured or figured calico gowns of all 
colours and diverse patterns. 

Sometimes, when in church or on festive occasions, they wear a scarf round the 
neck. On these occasionB also they wear hats, usually made of straw and often gaudily 
decorated with artificial fioweiB. These and the calico for their gowns they obtain &om 
storekeepers. They have been taught by the lady missionaries and South Sea teachers* 
wives how to make their garments, and many now own sewing machines. 

Calico or cloth is called duma wahi in the west, the plural form duma waktd 
signifying clothes; waku is a mat, but I do not know the origin of duma. The Miriam 
call calico or cloth and clothes in general watt ; am-wcdi is a dress from amile, v. clothe ; 
gem-waH, a shirt or chemise, frnm gem, body; mog-vtaU, a towel, from mog, a piece; 
and paper is eiau-wali (or jau-ioali) from eiau, the dura mater, connective tissue or the 

d by Google 


Basketby and Plaitwobk. 

Basketry and plaitwork are the most important of the native arte of the Torres 
Straits Islandeis, though here also, as is found to be the case with so many other 
arti&cts now in use, importations Jrom New Guinea are met with ; these are, however, 
very ezceptionaL Owing to the importance of this branch of the textile art it is fitting 
that it should be dealt with at some length, especially as no adequate account has 
heretofore been given of this industry in Oceania, with the exception of that by 
Dr Walter E. Roth (North Queensland Ethnography: Bulletin No. 1, 1901). 

The basketry and plaitwork may be grouped under the following heads: 1. Mats. 
2. Belts and bands. 3. Baskets. 4 Handles of baskets and other objects. 6. Petticoats. 
6. Ropes, strings and knots. 

Of all these the baskets are the most complex and involve the employment of 
nearly all the technical processes of the native art. Id order to save repetition the 
"strokes" employed in the simpler objects are described in the section on baskets, and 
thus the details of technique are kept together. 

It is interesting to note that string bags, so widely spread in Australia and in 
New Guinea, are not made in Torres Straits. 

All the specimens referred to are in the Cambridge Museum, except where otherwise 

Olossary of the principal terms employed. 

One of the difficaltiee in the description of the technical proceBsee of basketry or plaitwork 
is the absence of any general term for the essential factors of the technique, which it ia here 
proposed to c&lt "wefts" (ct glossary below). Since these may consist of such varied materials 
as a strip of leaf, a splint of wood, a piece of cane, a rigid rod or a wisp oi thread, and 
all may be interwoven either singly, doubly, trebly or in bunches, it is very difficult to find 
a. woid that ia in all cases at once descriptive, accurate and convenient. In describing an 
actual epecimeo it is easy to refer to leaf-strips, lengths of cane, rods and so forth, but 
concrete applications cannot be used in describing technique. 

Writers on the technology of basketiy have met the difficulty in various ways, and the 

perplexities are well illustrated by reference to the standard works of Otis T. Mason*. In 1901 

' 1901, "The Teclinio of Aboriginal Amerioan BuketTj," AmtrUan Anthropologitt, N.S. m. 1603, "DirMtiDDH 

for CoUaoton of AmariuBD Buketr;," PaH P. Bull. S9, U.B. Nat. Jfiu.; "Aboriginal Ameriosn Bssketrf," Rtp. 

n.S. Nat. AfM. (1904). 1906, " Vooabnluj ot HsI»7iUa Bssltetwork," Proe. U.S. NaL Jtfw. iixv. 

d by Google 


the wefts &re referred to as "warp elements" and "weft el^nenta," or (when the distinction 
fails as happens at the outeet, p. 110) merely as "element&" In 1902 "strand" is given in 
the glossary as the general term. In 1908 "strand" has a special meaning, and in complicated 
descriptions the "warp and weft elements" are distinguished as "horizontal," "vertical," 
"sinistral" and "dextral," and are referred to as "horizontals," "verticals," and so on. 

"Weaving element," "weaver" or "strand," the general terms used by other writers, are 
all open to objection, and since in elucidating an intricate process it is of great importance 
to have a word which by its literal interpretation, either popular or scholarly, can meui 
nothing but what it is required to mean, it is here proposed to nse the term "w^ts" for 
the technical factors in woven basketry, i.e. for rod, stake, splint, or strip of any material, single 
or compound, interwoven in the construction. 

Braid. Synonymous with Plait. 

Broken ninet. Fig. 9S. 

Check. Each weft passes alternately over and under each consecutive crossing weft, p. 79. See 

pL XV. fig. 1. 
Coiled haeketry. The coiled foundation is sewn tt^tber by some flexible material, each stitch 

croaaing two coils and interlocking with the one immediately below it. See fig. 117, and 

pi. XVII. fig. 4. 
Cord. Two or more wefts twisted t<^ther in the same direction. 
Decoration. The addition of elements, which are not essential to the construction, for the 

purpose of ornament. 
Dengn. The general effect derived from the pattern or conjunction of patterns. 
Dextral. (Wefts) leaning t«wards the right. 
*FUch. Two wefts twisted together in the same direction one under the other, enclosing a 

cros^g weft at each half-turn. 
Oblique Ihreee, etc., fig. 112. 
Overlaid weaving, twining, etc. Wefts not essential to the construction, inserted for decorative 

purposes, figs. 89 and 120—2. 
*Pair. Two wefts twisted together in tbe same direction one over the other, eaclosing a crossing 

weft at each half-turn. 
Palteme. The surface effects produced by various strokes. See Check, TtoHl, etc. 
Plait. The regular interlacing of not lees than three wefts to form a continuous baud or 

surface. A plait, cord or twine is described as three-ply, four-piy, etc, according to the 

number of wefts. 
*£and. The interweaving of single wefts over and under more or less rigid uprights. The 

term is used in wickerwork instead of Check. 
'Jiods. The horizontal wefts in wickerwork. 
Sinistral. (Wefts) leaning towards the left. 
''Slakes. The uprights in wickerwork. 

Stiteh. A completed movement in coiled basketwork, corresponding with stroke in plaited work. 
*Stroke. A completed movement in plaited basketwork corresponding with stitch in needlework. 
Three-ply cord, plait, etc. See Plait. 
Tvnlled ttooa, threes, etc. Each weft passes over or under more than one crossing weft, p. 79, 

figs. 108 to 113. 
Twine. Two or more wefts twisted together in the same direction enclosing crossing wefta 

at each half-turn, fig. 116. 

d by Google 


*Wale. Three or more wefts tiriaed in the same direction enclosing a crossing weft at each 

half-tnrn, fig. 116. 
Warp. A Terticai rod forming a passive element in wickerwork. 
Weft. A technical factor in woven or plaited basketry, corresponding to strand, rod, splint, 

etc. A wefc may consist of a single strip of material or of a number of strips. 
Zigzag ArMs, fig. 113. 
Ziffzag nintt, fig. 96. 

It may be noted that the terms are mostly unknown to the basket-maker's craft, which 
poBaessee a peculiar vocabulary of its own. For the true basket-making terms, marked with 
an asterisk, I am indebted to Mr Thomas Okey who has most generously answered numerous 
queries on the difficult subject of terminol<^ ; for illustraUons of the strokes see his ajlicle 
"Basket" in the Bneyeloptedia BrUanrtica, llth ed., 1910—11. 


The plaited mats, viaku (W.), moder (R), of Torres Straits are simple in construction 
and serve a variety of purposes. They are convenient for sitting on, for sleeping on 
or nnder, and for various other domestic uses. Formerly large mate, measuring about 
3'66 m. (12 ft.) by 1*52 m. (6 ft.), were used as canoe saile. Smaller and finer mats are 
used to wrap things in, and as mats for babies (pi. XIII. figs. 5, 6). Such are the Miriam 
iturt, the Malu name for which was viakot (llL p. 61) which is the Western word waJeu. 

A ceremonial mat made by interplaiting the pinnules of a coco-nut palm leaf, bei, 
was employed in the Bomai-Malu ceremonies (VI. p. 294, p. 301, n. 6, and p. 36), but 
it was never in common use. A similar technique is seen in the screens deiar or &c» 
mueni or hat moiaini, used in the rain-making ceremony on Mer (vi. p. 199, pL VIII.), 

In the Vocabulary papek (E.) is a mat made of enau leaves (Mimusops). In a 
Western folk-tale (v. p. 79) the following mats are referred to : humr (probably made 
of ban, Flagellaria indica, see also p. 105), vb, made of the straight-grained bark of the 
Tea tree vbu (Melaleuca leucadendron), minUai {minarlai) and pot (see below, p. 66), and 
in the Vocabulary a kind of mat is called t(^i. The process of plaiting is called umai (W.). 

Fia. 82. Fesrl-shell ioi^wr, Habaiag. Fio. 83. Shewing method of amahing large plaiUd mat, Haboiftg. 

The leaf-strips used in the makiDg of mats and baskets are rendered supple by being 
scraped with simple implements, wakabi (W.). These are usually semilunar pieces of 
pearl-shell of which the convex edge is blunt and smooth through use (fig. 82); their 
length varies from about 10 — 16 cm. 

The large plaited mats are made of leaf-strips, tvamim (W.), and are worked all 

over in plain check or twilled twos* ; some of the twilled examples have lines of twilled 

fives at intervals, suggesting a gapu design (see p. 66). The method of finishing off 

differs in the larger and smaller mats. In the former the sinistral wefls' are bent over 

■ For teahnioal teniis, see Gloftarj, p. 64. 

H. Tol. IV. 9 

d by Google 


downwards, doubled back, and interlaced up again under one crosaing weft, thus coinciding 
with the correaponding destral wefts; all the ends are then cut off together at the 
edge. The method is indicated in fig. 83, but it is practically imperceptible in the 
actual specimen without close scrutiny. A. fiirther point shewn in the same illustration 
is the method of finishing off the group of wefts left at the comer of the mat The 
bunch of loose ends is plaited together and the plait is threaded through the mat once 
or twice and fastened in a knot. 

The kuri mats vary in size from 665 cm. (26^ in.) in length by 38 cm. (15 in.) 
in width, to 48 cm. (19 in) by 20 cm. (Sin,). The plaiting begins at one end and is 
finished off at the other, the wefba being interlaced under and over a few crossing wefts 
and left to form a fringe on the outside. Sometimes the fiinge is on the inside, and 
occasionally the ends are all cut off short so that no fringe appears. The method of 
construction is seen in the sampler, fig. 84. 

Fia. 81. Sampler shewing oonHtmetion of taranud mat, Habuiog. 

The mats are divided into two groups. Saramud are worked all over in plain 
check and this admits of no design. The sampler seen in the figure is a saramud. 
One of these worked with very broad strips of leaf is called pot. The other group, 
minarlai (having patterns, from minor, pattern), shews more diversity, as these are all 
worked in twilled strokes. Some are worked all over in simple twilled twos, but usually 
the surface is broken up into rectangular areas by lines of twilled twos running in 
horizontal and vertical directions. 

Dr C. O. Seligmann collected some of the native names for the designs which are 
illustrated on pi. XIII. Thus a twilled pattern with a somewhat irregular appearance 
is gitalai sanatunga, crabs' footprints, pi. XIII. fig. 1, An effective chevron design is 
obtained by the use of double or treble wefts (pi. XIII. fig. 2). The under ones are 
worked in plain check, but the uppermost ones in twilled threes or fives, which thus stand 
out prominently on the surface. It represents an Isopod which is parasitic on sharks and 
is called baidamau ipi, shark's wife (see Section on Decorative Art). A similar design 

d by Google 



IB seeQ in fig. 7 on the same plate. The zigz^ shewn in fig. 3 has the name tidai 
maril (muril), bending spirits (cf, v. p. 369). Another design, pi. XIII. fig. 4, is buiulunga, 
like a glass bottle, it represents a square bottle with a short neck. 

Decoration is added by means of emphasising the varieties uf the patterns with 
dots or lines of blue, red or white paint, and by fringes of loose ends left projecting 
on the surface of the mat. In the mat illustrated on pi. XIII. 
fig. S the vertical lines of twilled twos are picked out in red and 
white paint, and red and white lines or dots run all round the 
border. Decorative fringes are seen in the mat illustrated on pi. 
XIIL fig. 6. Here the pattern consists of two strips of twilled 
twos separated by a median vertical line of the same stroke. Short 

horizontal lines of frayed ends of wefts project on the sur&ce of Fio. 8B. DeUilofdeoon- 
the mat (for detail see fig. 85), and ar« coloured red and blue. "o" in S"^ i>»t- Mw. 
Lines of red and blue dots also run down the centre of the mat and down the sides. 
As two or three wefts are used in the plaiting the number of loose ends does not 
interfere with the solid consistency of the 

Another effective mat is the one whose 
lower half is illustrated on pi. XIII. fig. 7. 
This is worked in double or treble wefts, 
and the greater part of the surface is in 
plain check. The whole area is divided by 
vertical rows of twilled twos into four strips 
and these are crossed at intervals by rows 
of twilled threes, made more conspicuous by 
being worked with the upper wefts only, and 
also by being bisected longitudinally by a 
line of blue paint and outlined with red, the 
colours being occasionally reversed. This is 
the same design as the gapu or baidanmi 
ipilnga design, referred to above. As only 
the upper wefts are used to make the design 
the mat is not revereible and only four strips 
of plain check shew on the back. 

Another entirely different type of mat 
used in Torres Straits, kai (W.), ka or ker (E.)*, 
is mode of strips of pandonus leaf, abal (E.), 
sewn together, fig. 86. It is an importation 

from New Guinea and is in only occasional Pm. 86. Folded mat made of pandanas lesTSB BBwn 
use. When employed as a tent during the toKether,ioi(W.), Jo (E.), imported from New Guinea. 
initiation ceremony at Tutu (v. p. 209) it ^oui length 3 58 m. (8 ft. 8i in), breadth 168 m. 
is called sobera. 

The technique is simple and effective, though difficult to describe. First pandanus 

d by Google 



leavea are split longitudinally into long atripa. Two of these are laid back to back, 
with their shiny eur&ces outside, and the two edges are stitched together all down one 
side. These may be called the A strips. Two more stripe, B, are taken and Wd above 
and below the first pair, with their dull sur&ces outside, and all four strips are stitched 
together down the hitherto unstitched edge. The free edges of the B strips are then 
folded together, as if one were to bend back the covers of a book until the edges met, 
and now their shiny sur&ces are outside, and the row of stitching is hidden. Another 
pair, C, is next sewn on to the free edges of B, just as B had been attached to A, 
and so on until the mat is complete. The result is that the shiny surfaces are all 
on the outside and none of the lines of stitching are visible. The sketch (fig. 86) shews 
how the mat is finished off with stitched Vandykes along the upper and lower edges, 
which, in the specimen illustrated, are decorated with tufts of wool. In the figure the 
mat is doubled in two and the horizontal insertions shew how weak pUces in the sur&ce 
can be repaired. 

Belts and Bands. 

The belts, wakau (W.), toaJc (E.), tall naturally into two groups, which differ, however, 
more in effect than in construction. , 

One group contains the belts consisting of a narrow plaited band, decorated and 
sometimes entirely covered with patterns of overlaid wefts and orna- 
ments of shells or seeds (figs. 72, 88, 89). 

In the second group the belts are wider and the plain check 
foundation is bidden under diagonal patterns in overlaid weaving, 
the ends of the decorative wefts usually forming a fringe all along 
the lower edge (pL XJV. fig. 1). These are probably all imported 
fivm New Guinea. 

1. The belts in the first group measure 7 cm. to 9 cm. (27^ to 
35^ iiL) in length, and 2 cm. to 5 cm. (1 to 2 in.) in breadth. The 
foundation consists in a strip of plaitwork, the wefts varying in 
number from 9 to 29, terminating in two three-ply plaits at each 
end; occasionally instead of terminal plaits the ends are twined into 
a two-ply cord, one end is passed through between the wefts of the other, and both 
are knotted separately (fig. 87). Over this foundation are patterns usually composed of 

Fio. 87. Tvirt ud 
knoti Rt the end of 
a belt, Hoislog. 

overlaid wefts of contrasting colours, caught under the foundation wefts at intervals, 
forming zigzag or other designs (fig. 89). Another example is seen in the belt illustrated 
on pi. XII. fig. 1, and the coco-nut water vessel handles {fig. 122) shew the same technique. 
In one example the zigzag is worked partly in yellow orchid stem, partly in strips 

d by Google 


of a species of puidanus, partly with red wool and stripe of scarlet twilL Sometimes 
the decoration consiats of shells or Coix seeds sewn on to the foundation (fig. 88). 

Crossed shoulder-belts or haldrics, naga, kamadi, maiei or moie (W.), vxigogob (E.), 
differ from the other plaited belts in being joined at the end to form a ring about 
80 cm. to 1 m. (31^ — 39^ in.) long. The ends may either be taatened one above the 
other mailing a continaoua band, or, when a loop of a sufficient size has been made, 
they may be joined back to back and both sets of wefts then plaited on tt^ther, 
terminating in a tassel In construction, patterns, and decoration they shew the same 
characters as the other plaited belts, the foundation being a plain plait, with overlaid 
wefts forming zigzag patterns, as in fig. 89. Some are decorated with Coix seeds, either 
sewn on thickly and forming a flat surface, or threaded on strings and attached at 
intervals round the belt, and many are ornamented with various shells. See also p. 52. 

IL All the belts in the second group shew the same methods of construction and 
vaiy but slightly in appearance. 

The foundation coasistB of a band, usually about 5 to 10 cm. (2 to 4 in.) broad, 
in plain check (pL XIV. fig. 1). The length varies frvm S'6 m. (about 6 yards) to 43 cm. 
(17 in.), the average being about 1 m.; the former are carrying-bands and the latter are 

The foundation sur&ce is almost entirely hidden under patterns in overlaid twining 
and plaiting (fig. 119) in alternate dark and light wefts. In the smaller belts the ends 
of these wefts form a fringe all along the lower edge. The design usually consists 
of rhomboid shaped areas, two sides of which are parallel with the margin of the belt. 
Each area is outlined with overlaid twining, and filled in with parallel rows of overlaid 
plaiting. These rhomboids may cover the whole belt, or they may occur in groups 
separated by various zigzag or fret designs (fig. 90). 

Fio. 90. Subbing of & belt eollMted b; d'Albeitia at HawaU in 1876. Borne (261S). Eaeh group 
of two L-shaped spaoei ia punted altemntel; red and blaok. i nat. aiu. 

In some examples, the whole surface is covered by zigzag designs, sometimes beautifully 
regular in construction, sometimes very indefinite. In the long fringeless belts, wavy 
lines run along the whole length. 

Whenever the sur&ce of the foundation shews in between the patterns it is coloured 
red or blue, and these colours are used in outlining the rhomboids and other patterns. 
Tufba of various coloured wools are occasionally added as decorations. The upper and 
lower edges of the belts are bordered by rows of overlaid black and yellow twining, 
which vary in appearance, according to the number of wefts used, their contrasting 
colours and the length of the stitch. 

Bands which encircle the upper arm vary considerably in size and form. They are 
called musur, occasionally put (W.), and put (K), see also p. 56. 

The narrow armlets usually differ considerably from the broader ones and several 

d by Google 


distinct types may be distinguished. The simplest are made of strips of ratao or of 
a stiff vine plaited in the same manner as the arm-guards (pL XIV. fig. 4). This type 
is also made in pliable materials and the ends are left projecting to form a long untidy 
vertical fringe. This (with the help of some examples of unfinished specimens from 
New Guinea, fig. 91) serves to shew the method of construction. The wefts are first 
securely knotted to a foundation thread formed by a strip of leaf, and then interwoven 
together in a loose phut, of 10 to 20 strapds. When a plait of sufficient length has 
been made to form the circle of the annlet, the free ends proceed again round the 
circlet, being plaited on in and out of the original simple plait, which is thus converted 

Fio. 93. Flut«d armlet, Her. 

into a surface of twilled twos. Having completed a second circuit the wefts are inter- 
plaited once more round the armlet, and the final pattern is thus twilled threes. For 
the second and third interplaitings a. bone or wooden skewer is necessary to form the 
intricate path for the wefts. In one example a further complication is introduced by 
the division of the plait into two parallel strips half-way round (fig. 92). Other armlets 
which may be included in this group are woven in various patterns in the same way 
but with their ends concealed. One shews a zigzag design with yellow orchid skin laid 
over the foundation wefts (pi. XIV. fig. 2). An exactly similar one comes from the 
Mekeo district of British New Guinea. 

Fio. 93. Detail of armlet, Maboiag. Fto. 94. Annlet from Mar. 

A Specimen of another type comes from MabuiE^ (pi. XIV. fig. 5). The foundation 

is made of the split raid-rib of the leaf of the banana, and a strip of banana leaf is 

bound round this and interlaced over and under two narrow strips of the rib on the 

outside. This is illustrated in detail in fig. 93'. A broader example of the same type 

1 An example ol thii pattern, bat with three horizontal stripa laid over the foundation, is figured by W. E. Both 
{North Qiuensland Elhnagraphji : BulL No. 1, Ian. 1901, pi. V. flg. 6, Briabane). These "Pandaniu eomponnd- 
anolete" oame from " the Embley Biver, at the HoTeton, and on the highei reaohee of the Batavia River" (p. 11). 

d by Google 


is worn in Uer when dancing, but it waa said to have been imported from New Quinea. 
Over a firm foundation is woven a pattern in oblique rows of twilled fours and threes, 
the vertical wefta passii^ over four or five and the horizontal wefts over three crossing 
wefts. A narrower example is decorated with two interwoven strips of twill, laid over 
the horizontal wefts, and another, also from Mer, has a decoration of a tuft of coloured 
wool (fig. 94). 

Another type is a musur from Mabuiag, which is simply a broad band of manu- 
&ctured webbing or elastic, decorated with vertical rows of strings of beads, attached at 
the ends only, two red strings alternating with two blue strings (pi. XIV. fig. 3). 

Vta. 9S. Broken nines. Fia. 06. Zigzag niaeB. 

Of the broad armlets the largest are about 15 cm. (6 in.) deep and 27 cm. (10^ in.) 
round the upper edge {pi. XIV. figs. 6 — 8). The pattern of the greater part of the 
surface ie twilled threes in horizontal rows, varied by horizontal bands 
of broken nines or zigzag nines (figs. 95, 96) which usually occur on 
each side of the decoration. The decoration forms the distinguishing 
feature. The ends of the wefts are left fairly long, and they are 
brought down together in a bundle or ridge and left projecting like 
a brush at the lower edge of the armlet. In one example the ridge 
is covered with a row of mouths of cowrie shells (fig. 97 ; pi. XIV. 
fig. 7); in another the ridge is covered with red twill, and bordered 
with two rows of cowrie sbelb, others are decorated with rows of 
Coil seeds, rows of beads, and strips of twill ; strings of beads or tufts 
of wool often form a tassel at the larger end. Part of the pattern is 
occasionally picked out in pigments. 

The ordinary arm-guards or bracera, kadiff (W.), kadik (E.), are plain, 
cylindrical in shape, made of strands of ratan, various vines or coco-nut 
fibre, and plaited in twilled patterns (pi. XIV. figs. 9 and 10). They 
vary in length from 20 cm. (8 iiL) to 32 cm. (12J in.), and in circum- 
ference from 195 cm. (7f in.) to 30 cm. (12 in.) at the upper edge. 
These plain kadig are seldom decorated, but one has a pearl button 
at the upper edge and another, as seen in the illustration, has a small ^"'- 3'- Awiie' with 
bound handle. See also p. 56. . „™ Jt" ° ''*'*"® 

Two dance kadig are figured on pi. XIV. figs. 11, 12. The smaller 
one is 21 cm. (SJ in.) in length and 285 cm. (Uj in.) in circumference at the upper edge 
narrowing down towards the wrist; the other specimen is 29*5 cm. (11} in.) by 26*5 cm. 
(lOJ in-X The stroke is a close rand, finished off at either end of the bracer by a row 

d by Google 


of plaiting in fibre. The decoration in the longer kadig conflists of lines of red and 
blue paint and a taesel of red beads and a ahirt button attached to the npper edge. 
The shorter kad^ is ornamented by threads of wools in various red and orange shades. 
These are laid on alternate foundation rods, and the wefts cross over and under them 
in consecutive pairs. The zogo kadik (fig. 77) is made in a similar manner, 

Leglets, or makamak, and finger rings, tag makamak, are made of the roots of the 
coco-nut palm, phtited in a continuous (usually five-ply) plait with invisible ends (fig. 7Q). 
This is worked with a single weft on the plan of the Turk's head knot. The leglets are 
often decorated with tufts of wool or beads. 

There is no special difference between the ordinaiy baskets of the western and 
eastern islands. The general name in the west is iana or ietia, and in the east, epei. 
The following names of baskets were collected : 

Western Islands: — ixUboi, a woman's basket (fig. 99), bcU, across, box, young coco-nut 

leaf; boi is also the name given to the temporary baskets made of cooo-nnt 

leaf (fig. 98); li, a woman's basket made of pandanus leaf (pL XVI. fig. 3), 

and kuta (pi. XVII. fig. 1), also a woman's basket made of the same material; 

mugagud (Muralug) in Vol ill. p. 112 is equated with hoi, and is probably 

connected with magi, small, and gud, mouth; walai (Muralug), watiii (Tutu) 

(pi. XVn. fig. 6), is the name of a basket made of the rush, walai, Xerotes 

Banksii Maiden (JJaejul Native Plants of Australia, p. 634) says "the leaves 

[of Xerotes spp.] are used for basket work by the aboriginals." 

Eastern Islands: — epgi is the general name for basket in the east, hence we have 

gerer epei (pi. XV, fig. 2), a man's basket, for fishing line, etc. from Mer, made 

of g^er, pandanus leaf; and bugtli epei (pi XVLL fig. 2), one made of Flogellaria 

indica; aipus (pi. XV, fig. 6) is a basket made of u lam (coco-nut leaf). 

Before beginning a detailed examination of the baskets of the Torres Straits Islanders 

it may be well to give a general description of the collection, and to distinguish the 

chief varieties that occur. All baskets fall naturally into one or other of the two main 

groups, plaited or coiled, and they can be further classified according to the stitches 

or strokes used in their construction. (A stitch is a completed movement in coiled work 

and a stroke is a completed movement in plaited work.) 

All the Torres Straits baskets, with one exception, are plaited baskets, and they 
form three distinct technical groups: plain check, twilled and fitched or twined. 

A. Check. 
The simplest form is that known as boi (W.) (fig. 98), This is very quickly made 
from the split green leaf of the coco-nut palm, and is used for temporary purposes only. 
Varieties of this form of basket are widely spread in Oceania. Hedley (Mem. III. 
Avsiralian Museum, 1897, p. 291) figures one from Funafuti He states that "the pinnule 
tips, instead of being knotted at both ends of the basket as in New Guinea, are plaited 

db, Google 


along the floor and knotted in one bunch ioaide. A second Bpecimen haa the knot 
outside the basket." According to Mr J. Bruce one form, called bos epei, "was introduced 
by . South Sea men to Murray. It is of an 
oval shape and made in different sizes from 
a large one, for carrying kumelas and yams 
from their gardens, to small ones for putting 
small fruits into," The three examples col- 
lected in Torres Straits shew three varieties 
in the finishing off along the base (for details 
of which see p. 77), but neither of these is „„„.,, ,, ... 

„ . Fio. 98. £ai, basket made of OMount palm iMf, 

the New Guinea type. M»bni»B. Hdght SOom^ draomftmiiMlSOcin. 

The only other baskets made entirely in 
plain check are a kuta (woman's basket) for carrying yams, collected in Habuiag, but 
said to come from New Guinea and Saibai (pi. XVIL fig. 1), and a large basket (pi XV. 
fig. 1) of pandanus. 

B. TvnUed. 

The greater number of the Torres Straits baskets belong to this group, and they 
shew great variety of appearance owing to the different strokes used and decorations 
added. Among the simpler examples are a coco-nut leaf basket (pi. XY. fig. 3), a pandanus 
basket (pi. XV. fig. 5) from Uabuiag, a halhoi (woman's basket) (fig. 99) of coco-nut 

Fia. 99. Baiboi, womu's Bah buk«t, MkboJag. H«lght 31 em., oiramnferenoe 62 em. 
leaf, also from Mabuiag, a Mer basket, gerer epei (pi. XV. fig. 6) and a stiff saucer-like 
tray (pi. XVI. fig. 4), Among the more decorative twilled basketa are those with zigzag 
patterns, such as the little bnrua iena^ from Uabuiag (pi. XVI, fig. 2), the gerer epei 
from Mer (pi. XVI, fig. 1) and the It from Moa, or Banks Island, figured in pi XVI. 
fig. 3, shewing a peculiarly bold design. Others, such as the lined basket (pi XVI. fig. 5), 
the bunli-epei, made, as its name implies, of Flagellaria indica (pi XVIL fig. 2), and a 
pandsLDUs basket, both from Uer, and the highly decorated one illustrated in pi XVII. 
fig. 3 are all included in the twilled group and are noticed severally under Design and 
Decoration below, pp. 82, 83. 

> Bunia from bunt, dirt;, probably tetut (o the dark mitt. The ipeoinien is new and noiued. 
H. Vol. IT. . 10 

d by Google 


C. Fitched or Twined. 
There are very few of this type and they all come from the western islands. ' Two 
examples are illnstrated on pi. XVH figs. 6 and 6. They are made of the rush Xerotea 
Banksii, from which they receive their name of wain (Muralug) or wa»ili (Tutu). The 
technique is characteristic of many Australian baskets, and it is by no means certain 
that they have not all been imported from Cape York. They are said not to be made 
at Mawata. Except that the whole sur&ce ia 6exible, the technique might be described 
as wickerwork, for the vertical wefts are parallel or crossed, and are connected at intervals 
all the way up by rows of pairing or fitching as in ordinary wickerwork. 

The baskets will next be reviewed in detail, under the headings Shape, Site, 
Construction, Design and DecoraHon, 


The Torres Straits baskets are generally oblong with sloping or approximately vertical 
sides and a slightly curved base (fig. 100, 1, 2 and 5); some have vertical aides and 
a square base, 4 ; and one rigid example is 
shaped like a shallow saucer, 6. Of the two 
fitched or twined examples, one is bowl-shaped, 
3, and the other approximately hemispherical, 7. 
The coco-nut leaf basket illustrated in fi^. 98 
is asymmetrical. Mr J. Bruce says that the 
latter type, as originally made by the Murray 
people, was of a di£Ferent shape [approximately 
fig. 100, 1 inverted, with concave sides] and 
smaller than the lai^^t sizes of the introduced 
South Sea examples. Another specimen in the 
Cambridge collection ([364] not figured) is square 
in shape, with the same dimensions (104 cm., 
41 in.) round the base and the upper e<^e. 

In size the baskets vary from a large hemi- 
spherical specimen (pi. XVII. fig. 5), 1-165 m. 
(45f in.) in circumference and 37 cm. (14} in.) in 
depth, to a minute wallet 10 cm. (4 in.) round the upper edge and 6 cm. (2} in.) irom 
top to base. An average size would be represented by an example measuring 63"5 cm. 
(25 iiL) round the upper edge and 22'8 cm. (9 in.) fixim upper edge to base. 

All baskets may be placed, according to the method of their construction, in one 
of two technic groups: — 

I. PUited or woven basketry, in which the wefts are interplaited. 

n. Coiled or sewed basketry, when a pliable weft is sewn or wound round a rigid 

coiled fotindation. 
The examples of baskets from Torres Straits alt, with one exception, belong to Qroup L 

d by Google 









•I 3 
J I 


.3-S " 





d by Google 


L Plaited Baskets. 

The plaited baskets may be classified according to their most aalieat features, 
which aie: — 

1. The method of beginning. 

2. The method of finishing oS*. 

3. The strokes employed. 

1. There are two distinct methods of beginniag the basket. A. Begimiing at the 
base. B. Beginning at the upper edge. 

A. The simplest baskets are made fiom the base upwards. The base consiats of 
a number of parallel or crossing or interlaced wefts (pi. XVI. fig. 6), and these are 
plaited together u[\ the sides of the basket, finishing off in various ways at the upper 
edge. This apparently is not a native process, but has been introduced from Samoa. 
It seems probable that it has now superseded method B. (6) below. Occasionally this 
type is begun at one side and the bottom is seamed together. 

B. There are at least two methods of beginniag at the top. In the simpler of 
these (a), the weaver takes a piece of mid-rib split down the middle, and coils it round 
to form the rim of the basket; the attached pinnules (fig. 101) or spUt strips of leaf 
(fig. 102) form the wefts, which may be supplemented by other strips not attached to 

Fio. 101. Intide of npper edge. Fta. 103. Ftajed oat leaf for the begiDoing of a buket. 

Ct pi. XV. flg. 8. 

the coil. These are all plaited down to the base of the basket where it is finished 
off in various wajrs'. A variety of this method is shewn in fig. 103 representing one 
side of a finished basket. In this example short lengths of mid-rib with pinnules attached 
are plaited vertically instead of horizontally. 

The other method of begimiing (6) is less obvious, as it starts with no initial coiL 
It is impossible to deduce the process from a study of the completed baskets, as they 
have the appearance of being finished off both at the base and at the upper edge 
(fig. 104) and no information could be obtained in the islands. Unfinished baskets from 
the Solomon Islands apparently shewing the same technique were collected in various 
stagee of incompleteness by Dr W. H. R. Rivers, and are described in Afan (VoL X. 1910, 
No. 93), "The Solomon Island Basket." It is possible that the Torres Straits baskets 

' For the boi bukets [fig. 98) a single ooil forma the appei rim. The ordlnar; method for finei bMketa is 
to mkke two ooili whioh are attached ti^ether and loose wefla plaited in where DBoeaBaT7. This oaa be oleari; 
•een in the oompleted baiketi illQitrated on pt. XV. figs. 8 and 6 Mid pi. XVI. flg. 4. 

d by Google 


in this group are made in the same way. Fig. 105 shews the detail of the croeaing 
of the wefts at the upper edge'. Eliactly the same result could be obtained by beginning 
the basket as in B. (a) above and, after plaiting the body of the basket, cutting off the 
initial coil and finishing off the ends in one of the methods described below ^ 

Fia. lOt. Soft wallet of banuiB leal, MabnUg. 
Height 10 cm., ciroamferenoe 39 am. 

Vm. lOS. Drtall of npiwr edge of baeket. The dextral wefti are diitiogniahed hj diagoiial ihading; the 
nnittral vefti are plain. The part to the right in (he diagram ia left oovered to the* the ooDitraotioD. 

2. There are two distinct methods of finishing the basket. A. Finishing off at the 
base. B. Finishing off at the upper edge. 

A. Very little variety is seen in the basal terminations, as, except for the boi made 
of coco-nut palm leaves, and one or two others, the same method is employed for all. 
The appearance is not by any means uniform, on account of the variations in material, 
but the intention is always the same. 

a. In the temporary coco-nut leaf baskets, boi, the pinnules are twisted or plaited 
together when they meet. In the example illustrated in fig. 98 the finishing off consists 
in a plait on the inside of the basket, running from left to right, turning at the right 
band lower comer and continuing back on the outside from right to left, Eresb pinnules 
being caught in all the way along in both directions. The ends are knotted on the 
' For 'the pnrpoaea of deioription thia fringe ia treated aa a "finiihtDg off," and dealt with below. 
* Dt Q. Landtman who proonred some unQniehed apeoimenB of haaketa begun in method B. (a) oonld obtain 
no inlonnation aa to the beginning B. (b). He wm told that ftome old women knew of other kindi oE indigenous 
buketa, " but on oloier inqnir; nobodr aeemed to know anything fnrther." 

d by Google 


oataide and the knot is tucked in between the pinnules a little way up the side. In 
the two other examples a twist takes the place of the plait, the pinnules being wrapped 
round each other where they meet. In these the twists are both on the insides of 
the baskets; in one the knot is also on the inside, in the other the ends are twisted 
on outside and the knot tucked in, as in fig. 98. The sune method of finishing occurs 
in the basket illustrated on pi. XVTI. fig. 3. It terminates very neatly in a five-ply plait 
all along the base on the insid& The plait continues on the outside on reaching the 
edge of the base, and passing up the side ends in a knot, concealed under the strip 
of cane. 

b. In all the other baskets, as soon as the base is reached a rapid decrease is 
made by bunching bother several wefts and interlacing together these thick wefts in 
plain check. When the wefts meet in the median 
line they cross each other and continue under 
and over the successive crossing wefts until the 
opposite sides are reached; they are then cut off 
on the outside. The result is sometimes rather 
clumsy and indistinct, but generally it is definitely 
symmetrical, as is clearly seen in fig. 106, and 
can be distinguished in the side view of the 
same basket in pi. XV. fig. 2. 

T, nn. -L-i-i.- r - ^- Fio. 106. Bue of bMket to ihsw tnattuxl ol 

B. There are many poesibihties of variation ^^^^^,^ ^f. pL xv. flg. a. 

and of decoration in finishing off at the upper edge. 

a. The wefts are all doubled down diagonally towards the outside, and interlaced 
under a few rows of crossing wefts. The ends are all cut off and left projecting in 
a fringe on the outside of the basket, one fringe slanting to the right and the other 
to the left> (fig. 104 and pi. XV. fig. 5). 

b. A different effect is obtained when all the wefts running in one direction are 
cut off, and only those running in the other direction are doubled diagonally down and 
interlaced. This results in a single fringe, the ends of which all slant in the same 
direction (pi. XV. fig. 2). 

c. In one example each pair of meeting wefts is taken together and doubled back 
and interlaced in the same direction. This gives a zigzag edge 

to the upper margin (fig. 107; pi. XV. fig.. 1). Here the fringe is 
on the inside. 

d. Occasionally there is no fringe at all; all ends are hidden 
under the crossing wefts (pi. XV. fig, 4). 

Some of the baskets are finished off at the upper edge with ^^^ jg^ DoDer edoe of 
a binding formed either of a strip of leaf, or occasionally of cotton & buket shewiDg mathod 
stuff; this binding is sewn or button-holed on (pi. XVII. figs. 3 "^ fimahing. cf. pL xv. 
and 6), and as it is rather decorative than constructional it is ^ ^' 
dealt with under Decoration below (p, 83). 

3. The distinctive character of each basket is most intimately bound op with its 
' In (be Solomon liUnd* bukets thia fringe ooonra on the iniide of (be buket; in (be apeoinieni ooUeoted 
in Tonea Btniti it is nBoall; on (he ontaide. 

d by Google 


weaving pattern, which is, as it were, the expression of its maker's individuality, and 
the device of her peculiar skiU. It is in this domain, more perhaps than in any other, 
that uncivilised fingers shew their snperiority over mechanical science. 

Each of the Torres Straits baskets in the collection has a distinct character 
of its own, and no two are alike, and yet, when their patterns and component strokes 
are examined, it is seen that they are all included in the three groups : A. Check, 
B. Twilled, and C. Pitched or Twined, and that by far the greater number fall under 
the first two of these, which may be considered as the simplest of all technical con- 
structions. But out of these most elementary methods of interlacing, many elaborations 
have been evolved, some of which are so complicated as to require carefiil tuLalysis, and 
the distinctive effects derived from variety in material and varied combinations of 
contrasting wefts, both in the construction of the baskets and in their ornamentation, 
give a marked diversity to the collection. 

A. Check. 

This is the simplest stroke in basketry, each wefb passing alternately over and under 
each consecutive crossing weft. It is generally used for the foundation of the baskets 
which start from the base, and for the finishing off of the baskets which start at the 
upper edge. Even with this simple pattern great varieties are obtainable, depending 
mainly on the material used 

The wefts may consist of whole pinnules (fig. 98) or of narrow strips of leaf only 
2 mm. wide (fig. 104) ; and they may be woven closely as in the latter example, or 
shew open spaces between the wefts as in the former. 

Varieties in effect are also achieved by the number of strips of leaf or other material 
taken to form each weft. 

B. TwHUd. 

In twilled weaving each weft passes over or under more than one crossing weft. 
There are so many distinct varieties of this pattern that it is convenient to classify 
the main types as follows: — 

Fia. 108. Twilled twoa. Fio. 109. Twilled two* ftnd odbb. 

L TwUled twos. o. Twos. b. Twos and ones. 

u. Twilled threes, a. Threes, b. Threes and ones. c. Oblique threes, d. Zigzag 

L Twilled twos. a. Each weft passes alternately over and under two crossing wefts, 
one of which has been crossed over by the weft above, and the other of which is creased 
over by the weft below, thus producing a diagonal effect (fig. 108). 

d by Google 


b. Twilled twos and ones. A different effect is obtained by passing all the wefts 
running in one direction over two and under one of the crossing wefts, while the latter 
pass over one and under two of the wefts which they encounter (fig. 109). 

ii. Twilled threes, a. This is exactly the same method as i a except that each 
weft now passes alternately over and under three crossing wefts instead of two (fig. 110). 

Fio. 110. Twilled Uusm. Fie. 111. Twilled threes mad ones. 

b. Threes and ones. A variety is obtained by passing all the wefts running in 
one direction over three and under one of the wefts running in the other direction, 
while the latter pass over one and under three of the crossing weits (fig. 111). 

c. Oblique threes. Another variety results from passing all the wefts running in 
one direction alternately over and under three crossing wefts, while the alternate wefts 
running in the other direction pass over one and under two, and under two and over 
one of the crossing wefts respectively. This produces a line of oblique threes (fig. 112). 

Fio. 112. Twilled obliqae tbieei. Fio. 113. Twilled Eigzag tbreeB. 

d. Zigzag threes. Twilled threes may also be worked in zigzag lines (fig, 113), 
PI. XVL fig. 1 shews a basket worked all over in zigzag threes. Fig. 3 on the same 
plate shews zigzag threes and ones. 

These patterns can be made more effective by the use of double wefts, composed 
of two strips of material. The upper strips are woven in twilled patterns, and the under 

d by Google 


ones in check. This causes the twilled pattern to stand out in greater prominence, and 
the effect is different from that produced by the simple interlacing. This may be seen 
in the detail of fig. 115. The upper wefts are woven in threes and ones, while only 
plain check appears on the inside. 

Fio. 116. D«tul ot bands. Of. pi. XT. Bg. t. 

Fid. 114. Horisantal Mid TstUeal twilled Omta. 

It is seen that various patterns are developed, as it were incidentally, at the junction 
of lines of different designs, and these junction patterns are sometimes definite enough 
to merit description. For example, when a vertical row of twilled threes meets a hori- 
zontal row of twilled threes the result is the pattern shewn in fig. 114 and pL XV. fig. 6. 

C. FUdud or Twined. 

This, called by Otis T. Mason' " twined weaving," is the ordinary wickerwork technique 
known to basket makers as fitched work. 

In English basketiy the materials are rigid ; the uprights are called stakes, and 
the wefts that bind the uprights together, rods. In the examples collected in Torres 
Straits the materials are pliable rushes, but the technique is the same, the uprights 
being connected at intervfds by horizontal rows of pairing, fitching, waling or plaiting. 
PI. XVII. fig. 6 shews a plain fitched basket. It starts at the base with eighteen parallel 
wefts which are secured by a row of fitching. As the fitching 
bends back in its elongated spiral, it fastens in a bunch of half * 
a dozen wefts doubled in the middle and radiating out into a ' 

dozen additional wefts to form the spread of the sides. Crossing 
the parallel basal wefts once more a similar bunch is inserted * 
at the opposite end. The fitching then proceeds symmetrically ^^ ^^^ ^^^ ^^ Atched 
round and round the basket until the rim is reached. More wefts .g^^, ^_ ^^^ q, three-ptj 
are added as required as the work proceeds; these are doubled twine. B, Three-pl; bnid. 
and looped over a row of fitching, no attempt being made to C- Ctoxei fitch. Cf. pi. 
conceal the irregularity. The large fitched basket illustrated on ■ fl«- ■ 

pL XVII. fig. 5 shews more variety. Here the surface is broken up by horizontal rows 
of crossed fitching, where the uprights are crossed, and held in place above emd below 
by three-ply plaitings, as seen in detail in fig. 116. 

■ Otii T. IfBHin, "The Teebnio of Aboripnol Amerieui Butketrj," Ameriean Anthropologiit, N. B. tn, 1901, 
p. 118. 

H. Vol. IV, 11 

d by Google 


U. Coiled Basketbt. 

There is only one example of coiled basketry in the collectiotL This ia the basket 
belonging to a stone top collected at Mer (pL XVII. fij^. 4). It measures 525 cm. (20i in.) 
in circumference and 6 cm. (2j in.) in height. 

The stitch is worked clockwise, beginning in the middle of the base, and consists 
of the following movements. The weft is brought from above 
downwards over two foundation rods and passes back under the 
lower one, thus forming a long stitch on the outer sur&ce of 
the basket. From behind it is brought out between the two 
foundation rods to the left of its own long stitch which it 
crosses over horizontally, and, passing behind the upper rod, it P'o- Il7- Detail of ooiled 
is ready to start again. This will be clear on referring to ^**'' ""• *"■ »■■ ^™- 
fig. 117 shewing the stitch in detail. The basket was formerly 

decorated with threads of red and orange wool caught under the horizontal stitches, but 
almost all traces of this have now disappeared. 


The technique of basketry is essentially restrictive, since, in woven examples, it limits 
all designs to straight-line effects. Some of the baskets are quite plain, but most of 
them exhibit some design consisting of either horizontal or vertical bands of a contrasting 
weave ; some have both horizontal and vertical bands, dividing the surface into rectangular 
areas, and the boldest effect is obtained by a junction of horizontal and vertical lines to 
form symmetrical zigzags. 

There are, in this collection, only a few entirely plain baskets, i.e. those in which 
the Bur&ce is worked all over in the same stroke, unrelieved by lines of different patterns 
or other variations consequent upon differences in the strokes 'employed. Here belong 
the boi, described above (fig. 98), made of coco-nut pinnules loosely plaited together, 
where the simple form is conditioned by the material; also a large basket (pL XV. fig. 1), 
made of double strips of pandanus, and a tiny wallet measuring only 112 by 63 mm. 
In these no design is seen, unless the fringe of split ends, left after finishing off at 
the top (see p. 78), can be so described. The pandanus epei (pi. XVL fig. 5) may 
also be grouped with these. It has a double surface, the lining being worked in large 
plain check, and the exterior in twilled twos. It is entirely plain save for occasional 
strips of wider material inserted vertically. 

Fl. XV. fig. 3 shews the simplest use of horizontal bands. The body of the basket 
is made with coarse strips of coco-nut palm leaf woven in oblique threes, but it is 
begun at the top and finished off at the base with lines of twilled twos. The balboi 
(fig. 99) shews the same design, bat here the main body of the basket is in rather 
irregular twilled twos in vertical rows. 

Of the two fitched baskets one (pi. XVII. fig. 6) should be classed with the plain 
examples, as the horizontal rows are essential features of the construction, and can 
scarcely be regarded as design. In the other (pi. XVII. fig. 5) the surbce is varied 

d by Google 


by decorative horizont^ bands of crossed fitehing, and these are picked out in colour 
to add to the effect. 

A simple form of the combining of horizontal and vertical rows is seen in the 
example in fig. 104, where the fine strips of banana leaf are worked in oblique threes. 
A little below the centre comes a horizontal band of twilled twos; below that is an 
area of the same pattern in vertical lines, and this is separated by another horizontal 
hand from the base. The fringe of loose ends all round the outside below^ the upper 
rim may also be regarded as part of the design. Another example is seen (pi. XV. 
fig. 3) where areas of vertical twilled twos are crossed by three horizontal bands of the 
same pattern, the lowest band being continued to the base. The basket illustrated on 
pL XV, fig. 4 starts from a base of plain check and shews externally horizontal bands 
of twilled weaving varied with areas of plain check up to the top. The bands are worked 
in twilled threes and ones in vertical rows, but they derive an unusual effect from the 
(act that the whole basket is worked with double wefia; the inner wefls are plaited 
in plain check and serve as a foundation for the threes and ones of the outer wefts; 
thus the latter stand out with greater prominence, while only plain check is seen on 
the inside. On pi. XV. fig. 6 is seen an example of the effective result to be obtained 
from the simple combination of honz^tal and vertical lines. The whole basket is woven 
in alternate bands of twilled threes, worked first in one and then in the other direction. 
The effect is enhanced by the use of lines and dots of red and blue to emphasize the 
design. In one basket (pL XV. fig. 2) the alternations of horizontal and vertical rows 
are so arranged as to form rectangles. 

The most effective design is the zigzag which occurs in three baskets of the collection 
(pi. XVI. figs. 1 to 3). Fig. 1 shews a man's basket for fishing hues, etc from Mer, 
and the design, which is very evenly carried out, is outlined in red and blue lines. In 
the burva iena from Mabuiag (fig. 2) the design is more or less obscured by the contrast 
between the brown and the straw-coloured strips of pandanus. The third example, the 
It from Moa, collected by Dr Haddon in 1888, is the most effective basket in the 
collection; the zigzag design stands out boldly and is emphasised with lines of bright 
red and black paint. 


The decorative impulse shews itself in various ways, but since the Torres Straits 
baskets were all made for use, a simple ornamentation contented their makeis and the 
greater number are entirely plain. But even in the plain baskets one ornamental feature, 
which is characteristic of the Torres Straits plaitwork, is rarely absent, and that is the 
decorative use of frayed out ends of wefts. It is characteristic of civilised basketry and 
plaitwork generally that the ends of the wefts should be concealed, but in this collection 
not only are the ends not concealed, but in baskets, mats, belts and armlets, they are 
utilised for decorative purposes forming . fringes on the outer eur&ce of the weaving. 
Besides this use of fi^yed ends, decoration is added by means of superficial decorative 
stitches, contrasting bindings, paint and coloured stufis. 

Frayed ends are found at the upper edge of the baskets, forming a single or double 
fringe, and also on the edge of the base (figs. 104 and 106). In one example (pi. XVII. 


d by Google 


fig. 3) tufts of frayed ends which are stained pink are attached aa decorations to the 
upper rim. 

The busUi epei illustrated on pi. XVII. fig. 2 is woven in plain check, but diagonal 
and zigzag decoration ia obtained by the use of added wefte of contrasting colours. These 
are employed in three ways. 

1. Overlaid inteiiaoinff. The decorative wefts are laid over the foundation wefts, 
and woven with them, or tucked in aftier the basket is finished, the ends being more 
or less concealed under the crossing wefts (fig. 118 a). 

2. Overlaid tvnning. Two or more decorative wefts are twined diagonally over the 
foundation wefts (figs. 118 b and 119 a). 

3. Overlaid braiding or pltutinff. PI. XVII. fig. 1 illustrates a woman's basket for 
carrying yams, etc. Here the plain check sur&ce is decorated with four horizontal lines 
of overlaid twining and two of overlaid plaiting (the materials used including strips of 
cotton stuff). In the latter stroke four or more wefts are plaited diagonally over the 
foundation weftis (fig. 119 B). This decoration only occurs once in the baskets, but it 
is very common on belts and bonds. It is worked usually with alternate pairs of wefts 
of contrasting colours or material. The wefts of each pair cross each other over the 
junction of two foundation wefts, and pass diagonally behind the next two foundation 
wefts below, these latter being covered by the next pair of decoration wefts. 

Fro. 118. DaUU of bukat to sbaw deoontive 
■titohei. A, Ov«Tl&id interUaing. B, Over. 
Ikid tvining. 01. pi. XVII. fig. 3. 

Fia. 119, Dugram to ihew deooi&tire stitohaa A, OTWlaid 
twining. B, OverUid braiding or pUiting. 

Some of the baskets have decorative Inndings round the upper rim. The one illustrated 
in pL XVII. fig. 3 has a strip of leaf (stained pink) bound on with native two-ply cord 
in button-hole stitch. Another (pi. XVL fig. 5) has a binding of red stuff. The li 
(pi. XVI. fig. 3) is bound round the upper edge with a strip of leaf, button-holed on with 
black cotton, and the large walsi (pi. XVII. fig. 5) is bound with cotton stuffi^ 

The use of colour has already been mentioned. The designs on the basketfl illus- 
trated in pi. XV. figs. 2, 3, 5 and 6, and pL XVI. figs. 1 and 3 are all accentuated 
by lines and dots usually in red and blue, and in the last example also in black. In 
the walsi (pi. XVII. fig. 5) the horizontal bands are coloured blue, and the adjoining 
rows red. In the basket illustrated in pL XVH. fig. 3 a pink colour is used as decoratiorL 
The tufts of frayed ends along the upper edge are stained pink, and the strips of cane 

d by Google 


which are sewn on like a belt and outline the base and Bides are bound round with 
broad strips of leaf stained with the same colour. 

Coloured Bta£b are used not only as wefts in decorative strokes, as mentioned above, 
but also as binding along the upper edges of the baskets. One of the walsi (pL XVII. 
fig. 5) has a binding of this description made of pieces of black stuff, red twill and 
pink ^Kitted print in regular sequence. Shells, seeds and similar objects are not employed 
in the decoration of baskets. 

Handles op Baskets and otber Objects. 

The handles are invariably separate in construction from the baskets, and they are 
usually very roughly attached. They are commonly made in the same 
material as that used in weaving the body of the basket, and the most 
usual type is a simple three-ply plait (pi. XVII. fig. 6). In many cases 
where the material in which the basket is constructed is unsuitable for 
a plaited handle, the latter is made of a different and more pliable 
material. And often a rough handle is made by threading in strips of 
cotton stuff, or a piece of string; these are interlaced once or twice 
through the sides of the basket, and are retained by a knot on the 
inside (pi. XV. fig. 4). 

The most ornamental handles are those in which a broad plait of 
5 — 10 or more wefts is used as the foundation for overlaid decorative 
patterns, usually of wefts of contrasting materials, which are caught in 

at intervals under the foundation wefts (pi. XVI. figs. 2, 3, 5). ^ .^ a i, 

_, , _ 11 ■ 1 I J 1. ■ Pio. 120, Handle 

These decorative wefts generally zigzag across the band formmg ^f tngket. Ha- 

the handle, in parallel lines (figs. 120, 121). The technique is similar to boisg. Cf. pi, 

that employed in the belta (p. 68). ^™- ««■ ^■ 

The method of attaching the handles to the baskets differs veiy 
considerably, and generally it is of the clumsiest description. In many 
examples the handles are simply interlaced down the sides of the baskets 
once or twice, and knotted on the inside. Occasionally the knotted ends 
of the plait which forms the handle are sewn' into a small tuck at each 
comer of the basket (pi. XV. fig. 1). 

Sometimes the wefts of which the handle consists are taken separately 
and stitched over and over the upper edge of the basket, and then knotted 
tt^ther on the inside (pL XVII. fig. 1). In some examples they are 
sewn on with string. In one basket a piece of ordinary string serves 
as a handle. It is threaded horizontally through the sides of the basket, 
a little below the upper edge, and knotted on the inside. The string ^ ^^. n„Ji. 
is drawn through so that the handles spring from the inside, and on of badet. Ha- 
the outside only the horizontal loop of string is seen (fig. 99). buiag. CL pi. 

A stronger and more elaborate handle is seen in the basket (pi. ^^' '^^ "- 
XVn. fig. 3). Here a stout bundle of fibres is wound round with string in a button- 
hole stitch. 

d by Google 


Designs in overlaid wefts similar to those seen in basket handles, occur also in 
some of the handles for coco-nut water vessels (pi. III. fig. 1). These may be made 
in a nine or more ply plait, ornamented with a varied number of decorative 
wefts. A variation of this design is seen in one of these handles made 
in an 11-ply plait with eight decorative wefts, which are interwoven with 
the foundation wefts in the centre of the plait, but pass over two wefts 
at the margins (fig. 122). This pattern is also found in a seveu-ply plait 
with five decorative wefts. In all cases the fonndation is painted blue 
or red where it is visible between the decorative wefts, with a very 
efiective result. 

The handles, keddup (E.), for the coco-nut water vessels are 60 — 80 cm. 
(24 — 31 in.) long and 2 — 3 mm. broad, divaricating at each end in two 
plaits of the foundation wefts alone, 13 — 26 cm. (5 — 10 in.) long. The 
same pattern is found in the belt on pi. XII. fig. 1. 

Handles were made for carrying the large Fueus shells, used as water 
vessels (fig. 150). One of these is 42 cm. (16* in.) Ions, terminating at ^"- ^**. Handle 
, j' , . . /o. ■ \ ■ J- ,. . -.1. oroooo-nutwsiet 

each end in a loop or nng. 6 col (2J m.) m diameter, to receive the ytime\ llei 
umbo and siphon of the shell respectively. The thick bunch of coco-nut 
fibre forming the foundation ia bound round spirally with a fine two-ply twine of the 
same material. 


The petticoats, zazi (W.), netur (&) (pi. XVIII.) of the Torres Straits women form 
one deep busby continuous fringe all round the body, festening on the left hip by means 
of a loop in &ont, and a long free end, which is slipped through it Occasionally a 
gap of an inch or so occurs on the right side (pi. XVIII. fig. 4). This is probably 
due to the influence of the Daudai type of petticoat, see below. 

The materials are mainly banana leaves, kaba (R), and the bark of the root of 
ome, a species of Ficus. PL XVIIL fig. 3 shews a petticoat with a fringe made entirely 
of kaba, and fig. 2, one entirely of ome. One Miriam petticoat is made of the leaves 
of a species of flag, teger (Philydrum lanuginosam), and another (pi. XVIII. fig. I) is 
made of otne and strips of teicf. Strips of yam or of cloth are also used, and one 
example is made entirely of yam. The banana leaves are split down the mid-rib, and 
the surface of the leaf is frayed out, forming a fine curly fringe; occasionally the centre 
of the leaf texture is frayed into fine stripe and the edges are left intact. The ome 
bark is used sometimes as a wide tace-like mass, sometimes it is shredded and forms 
strings, and it is either left in its natural colour, or stained in various shades of red 
and brown with a decoction of mangrove root. 

In size the neaur vary in length round the waist from 88 cm. (34^ in.) to 46 cm. 
(18 in.), and in depth from 75 cm. (29^ in.) to 33 cm. (13 in.). 

The construction is the same in all examples, a continuous fringe being bound on 
to a foundation consisting of one or more cords. 

The stitch by which the fringe is bound on to the foundation cords varies slightly 
in the difierent petticoats. To illustrate these variations the groups of material forming- 

d by Google 


each knot are separated in fig. 123 so that the stitch may be seen, but in the petticoats 
themselves they are pressed so closely together as entirely to conceal the method of 
construction. In the simplest type the foundation is a single cord of stout twisted fibre, 
round which the bunches which form the dependent fringe are doubled and knotted, 
and kept in place by a series of wefts as seen in the detail (fig. 123 a). Other petticoats 
are made on a foundation consisting of two cords of twisted fibre, and the variations 
in knotting may be seen in fig. 123 B — F. The foundation cords continue beyond the 
ftinge on either side, one end forming a loop and the other a free end to pass through it*. 

Fio. 133. DataiJs oF knotting in fringei of pettiooftU. Mer. E and F, front and book Tiews of the ume Bpedmeii. 

In Daudai and Kiwai petticoats, wapa, of a different type are worn. These consist 
of two long narrow firinges of erepupu leavea, one forming the back and the other the 
front, leaving the thighs bare (pi. XXI. fig. 2). The small fore part is tucked between 
the legs in front, then the broader part is brought between the legs from behind, over 
the front fiinge, and is tucked under the front panel tcom below and the free ends 
hang down over the panel in a long fringe. 

Of the two fringes forming the wapa the back piece is the wider, being 17 cm. 
(6^ in.) wide, and 1*08 m. (42} in.) deep, while the front piece is 10 cm. (4 in.) wide and 
rather over 1 m. deep. Both fringes are made on the same foundation cords, which, 
in a plait 21-5 em. (8^ in.) long, join the front and back panels across the right hip, 
and terminate in a loop at one end, and a long free plait about 1 m. long at the other, 
to fasten on the left hip. The foundation pieces form solid panels, consisting of four 
or eight rows of twisted fibre, over which the bunches of erepupu leaves, which form 
the fnnge, are doubled and knotted round and kept in place by strips of pandanns. 
The narrower panel, which forms the foundation for the front part of the petticoat, is 
illustrated in fig. 124 and shews the method of construction. This panel is entirely 
plain, as when worn it is covered by the back fringe. The broader back panel (fig. 125) 
is covered with decoration consisting of zigzag and other patterns in overlaid twining, 

■ Tbe loop ia on tb« left bknd b 
with Ooeidentkl women. Thaie ia oe 
CTer right ta with Oooidental men. 

le of the wearer in front, bo that the foatening ia from right to left bb 
example, howevit (posBibly from Mawata), in whioh the &atening is left 

d by Google 


in alternate dark and light rowB, with a row of overlaid plaiting at the upper and lower 

Fio. 124. Front ptuiel of itapa petticoat, Daodii. 

Fio. 13fi. Worked-np rubbing of back panel at a ttapa pettiooat ; ) naL site. 

Design and Decoration. 

Since the neaur are all made on the same plan, with alight variationB, moet of which 
are only variations in construction invisible in the completed petticoat, there is little 
scope for originality in design or decoration, and the most striking differentiation lies 
in the use of contrasting materials. 

The neaur made of ome and teid, illustrated on pi XVIII. fig. 1, shews a variation 
in technique. The ome is knotted over the foundation cords as usual, 
hut the teid is differently treated. It is split into narrow strips, 
about 10 to 15 cm. (4 to 6 in.) long, and six or seven of these are 
taken to form a bunch. Each bunch is doubled in the middle, and 
inserted between the strands of a two-ply twine of the same material, 
so that the ends hang down in a continuous tassel, completely covering 
the central cord. In fig. 126 the bunches are lifted out of the way 
to shew the method of construction. 

The same technique is seen in a gaudy Miriam sporran made 
of cotton strips (pi. XII. fig. 1). The dependent fringe consists of Fiq. lae. Det^l of 
eight strings, about 50 cm. (19i in.) long attached to the centre of '•"eta ol uid petti- 
the belt These are made on a foundation of fibre twined into a ^rVr ^' , ' *" ' 
two-ply cord. At every third turn, a narrow strip of cotton stuflF is 
inserted with its ends banging down on each side. These cotton strips form a con- 
tinuous tassel concealing the central cord, and they are grouped into bands of indigo, 
white and scarlet alternately, terminating in &ncy strips of various colours at the ends. 

d by Google 

Ropes, Strings and Knots. 
By A. C. Haddon. 

Many of the ropes, uru, urukam, kurdai or kwodai (W.), lager (&), and lashings 
used for various purposes by the Torres Straits Islanders consist of the natural stems 
of vines or creepers without any artificial preparation. Such are vietkep (EL) the stem of 
the vine Pueraria phaseoloides and nr«b or airip (E.) the stem of the Queensland bean 
(Entada scandene) which is used as a rope for thatching and for canoe cables (vi. p. 47) 
when it varies fixjm 6 — 12 mm. in diameter. At Saibai a species of Calamus, Hrenga, 
and a smaller cane, genagen, which was imported from New Guinea, were split and used 
for house-building and as cables for the stone anchors. The stem of the tesepot (E.), 
a species of Clerodendron, is used as a rope. The stems of the " lawyer vine " (Flagellaria 
indicaX but or frtm (W.), hot (E.), are used in house-building, tying fences, etc. It is 
generally split into strips, 7 — 15 mm. in diameter, and forms an extremely tough yet 
pliable ropa 

The husk of a coco-nut is split into lengths which ore knotted together at the 
ends in reef knots (fig, 127), thus forming a rough kind of rope, ked (E.), which I have 
seen used in Her for tying up bananas (pi. XXI. fig. 1). 

Fio. 137. Bope fonoed ol oooo-nnt hnak, Mer. 

At Yam and Saibai rushes, gai (a sp. of Juncus), are dried in the sun and twisted 
into rough ropes which are used for canoe lashings and especially by the women for 
tying and carrying loads of firewood. At Mabuiag they were said to be used for lashing 
the poles of a dugong platform, and I was informed that a galai gaipapi was tied round 
the tail of a dugong after it was harpooned. 

The strings most commonly employed are those made of coco-nut fibre, the prepara- 
tion of which in Mabuiag is thus described by Dr Seligmann. " The husks, muH (W. E.), 
are soaked in water until soft, which generally takes a fortnight. They are then beaten 
out with a (^lindrical piece of hard wood, buru tut, after which the individual fibres 
aro pulled out of the softened mass and scraped with an ahd (Cyrena) shell on the 
instep of the foot of the squatting operator, the remnants of broken fibre, bum, being 
rejected. This process is known as muti pudan, or more commonly luwaian. The prepared 
fibres, mutian kunil, are left to dry in the sun for a short time ; they average in length 
from 30 — 35 cm. Wood ashes axe then smeared on the bands and instep of the squatting 
operator and the fibres are roughly twisted into strands, kosar taman. The commonest 
form of fishing line consists of two such strands twisted together, arigal kupmani, the 
strength of the line being determined by the number of fibres used in forming the 
original strands. The best and strongest strings or cords are made by plaiting, umai (W.), 

H. Td. IT. 12 

d by Google 


three or more strands, the individual fibres of which are carefully selected and matched, 
and are not even loosely twisted before being combined into strands. The three-ply braid so 
made, mvi-umaizinga (iffa in Saibai), ked (E.), is used for innumerable purposes. Jgal, 
or iffalai, kupmani of Saibai, is a four stranded line so plaited as to be round, the 
others are flat or oval in section." Igal is a common Western name for a twisted 
string made of mutt which is employed in catching turtle or in fishing. Ariga (W.), 
ariag (E.) is a fishing line. 

MacgilHvrsy states (ii. p. 20) that the fiishtng line of the natives of Uuralug "is 
neatly made fivm the tough fibres of the rattan, which are first scraped to the requisite 
degree of fineness with a sharp edged Cyrena shell, then twisted and laid up in three 

Finer and softer strings, wali (W.), vmlt lager (R), are made of the inner bark 
of the wcdi tree, Pipturus argenteus. This is the Queensland grass-cloth plant or native 
mulbeny of Australia; Maiden states that the tree is not endemic in Australia and 
that it "affords s fibre of fine texture and great strength; it is, however, rather difficult 
of preparation" {Useful Native Plants, p. 630); of dum, Apocynacea, others are usually 
twisted in a two-ply twine. Dr Seligmann noted that " at Mabuiag both urakar (Hibiscus 
tiliaceus) and avpodar (Dianella ensifolia and ? Hsemodorum coccinium) are used by women 
for domestic purposes. The cortex of the former is soaked in water for a day or two 
when its softer internal portion is stripped easily fi»m the harder external layers, it is 
then twisted into serviceable cord. The stems of supodar when shrunken and softened 
by drying in the sun are used in a similar way. At Saibai roughly plaited ropes were 
made fivm Hibiscus and fi^m kausar (Pandanus pedunculatus)." We collected at Saibai 
a beautifiiUy woven cord made of urakar and used by women. It starts at one end 
in a two-ply cord for a short distance (18 cm.) changing into a three-ply plait for 42 cm., 
broadening into a five-ply plait which continues for I'l m. (41 in.) until the central belt 
is reached. This is woven in a 13-ply plait and is ornamented with a zigzag pattern 
of four overlaid decorative strands of kauaar in the design illustrated in fig. 120, p. 85. 
The rope then narrows into a five-ply plait, then to a three-ply plait, and then to a two-ply 
cord, as at the beginning, thus forming a symmetrical whole. 

Fia. 128. DngoDg rope, Mftbnitg. Fio. 139. Worked ooid. 

The running line or rope, am, amu (W.), used in dugoag fishing, is made of stems 
of ruku (a creeping and climbing plant, Apocynacea), which is not prepared in any way. 
It is made of a four-sided, loose, eight-ply plait, 45 mm, in circumference, plaited with 
sixteen strands (fig. 128). On account of its buoyant nature the am is preferred to 
European rope. 

Some of the worked cords, such as are employed in making petticoats, shew a more 
complicated pattern. Fig. 129 gives the detail of a double cord interlaced with two wefts 
of contrasting colour. 

d by Google 


Mr W. I. Pocock has been so kind as to identify and remark upon the knots, 
kabu (W.), mitinA (E.) (fig. 130), collected by Mr J. Cowling ; they were made by Tom and 
verified by Waria of Mabuiag. The names obtained with the knots have been placed 
in inverted commas and their actual or probable signification in brackets; the words 
do not appear to be the true names of knots, but it seems as if they were given to 
explain the use of the several knots. 

No. 1. " Ariag" is a reef knot used to increase the length of a fishing line, ariag, 
by adding another to it. 

No. 2. "Amu" is the fisherman's or Elnglishman's knot, and is used to fasten two 
ropes together, or to mend a dugong rope, am. P. N. Hasluck {Knotting and Splicinff 
Aopes and Cordage, Cassell & Co., p. 22) and J. Tom Burgess {K-nots, Ties and Splices, 
Routledge & Co., p. 41) agree (1) in laying the left-band end above the right-hand 
end, and (2) in making either tie by passing the end just over and then under the 
standing part. In amu condition (1) or condition (2) seems to be reversed, but not both. 

No. 3. " Oalagipab " {galai gaipapi, see section on Dugong fishing) is the running 
Dooee that is tied round a dugong's tail after it has been harpooned. 

Fio. 130. Knots from Uabniag. 

No. 4. " Kabtoedan " (kabu idai, to &sten, tie a knot), a is fostened to a canoe 
and B — B is a mai^^ve stake stuck in the beach (it should have been drawn in a 
vertical position). This is a lark's head knot crossed and stoppered with overhand knot 
of fi^e end round standing part. 

No. 5. " Muiai mukub" (mui, fire; muku poidai, fasten, tie) is a knot for tying 
up a bundle of firewood, or for tying up belongings. It is nothing more or less than 
the single bow (tied reef-knot fashion), though by the specimen it seems to have been 
formed by passing the end through a half turn in the standii^ part, making a reverse 
half turn round the standing part and passing a bight instead of the end. It may 
have got into the shape by hauling on different parts of the knot. 

No. 6. "Sur" {sunt or sur in Daudai, a pole for punting canoes), a is the painter 
of a canoe, B — B a mangrove stake to stick in the beach. The cord passes twice rouod 


d by Google 


the mangrove stake which givea the idea that it is intended not to slip. Perhaps 
the simple twist is enough to prevent the cord slipping down the stem ox letting the 
noose run. The slip knot is probably a stoppering as in No. i ; most likely the practice 
is to draw the slip knot taut against the twist. This would be a good tight fastening, 
easily undone or removed from the stake. 

No. 1. " ladi mukaind" (iadi, stone anchor, and ? mtutu poidai, the hawser of 
a canoe) is fiistened by a reef knot like No. 1 at a to the bridle that holds the 
stone used as an anchor ; the knots at B — B are the same as No. 6. The hitches 
of Nos. 6 and 7 are like simplified rolling hitches. The reef knot, it may be noted, 
being made on a bight is really a plain lark's head as would appear if the "bridle" 
were pulled taut at the knot. 

The knot used in leashing the sucker-fish is described in the section on Turtle Fishing. 

d by Google 

By a. WILKIN and A. C. HADDON. 

Me Wileik made a special study of the differeDt types of houses which he saw in 
his travels in British Kew Guinea, but, with the exception of those of the Fly River 
district and Torres Straits, his notes were not written up. His information concerning 
the older types of houses in the western islands was scanty so I have incorporated 
my earlier observations. The description of the eastern round houses and of the 
modem houses is practically entirely his, but I have added some notes of the old Miriam 
house recently sent to me by Mr J. Bruce. As no satisfactory account has been 
published of the houses of the Fly River delta and Daudai I have included Mr Wilkin's 
descriptioD of these together with his remarks concerning their relationship to the 
Torree Straits houses. The descriptions would have been more detailed and lucid bad 
Mr Wilkin lived to see them through the press; all the diagrams were constructed 
by him. [a. c. H.] 

This section deals with: (1) Southern, Central and Northern types of Houses of 
the Western Islands. (2) The Houses of the Eastern Islands. (3) The Modem Houses 
of Torres Sb^ts. (4) The Long-house of Eiwai and some Daudai Houses. (5) Notes 
on Houses in Torres Straits and Western Papua (British New Guinea). (6) Fences. 

Western Islands. 
It has been noticed by Melville {Sketclies, pi. 16) that there was a "gradual 
improvement in the huts and in the general condition of the people towards the 
northward." He "first noticed at Mount Ernest [Nagir] huts superior to any seen 
in Australia." In 1888 it was evident that the houses of the Western Islanders improved 
in quality passing from south to north, and that roughly speaking they fell into three 
groups which correspond with (1) the Kauralaig, (2) the Malulaig and Kulkalaig, and 
(3) the Saibftilaig (v. p. 2). The first of these peoples inhabit the Prince of Wales 
group (Muralug, ete. and Moa), the third inhabit Saibai, Dauan and Boigu, while the 
second belong to all the remaining inhabited islands of the western and central divisions. 
For the sake of clearness we may speak of the three groups of houses as the southern, 
central and northern respectively ; but it must be remembered that this distinction 
belongs solely to an order of things now long passed away and that even in the past 
there was probably no rigid line of demarcation between the construction and dis- 
tribution of some of the types of houses. A somewhat variable South Sea type of 
house now prevails everywhere. 

d by Google 


Southern type of Home. 

Hacgillivray (ll. p. 19) states that " the huts which the Eowraregas and Cape York 
people put up when the rains commence are usually dome-shaped, four to eis feet 
high, constructed of an arched framework of flexible sticks, one end of each of which 
is stuck firmly in the ground, and over this sheets of tea-tree (Melaleuca) bsrk — and 
sometimes an additional thatch of grass — are placed until it is rendered perfectly 
water-tight." Brockett (p. 37) says of Wednesday Island : " Their houses were not so 
neatly made as the huts in other parts of the Straits, and they were built in a different 
shape, somewhat resembling that of a tent." 

Fio. 181. Three hnU, mUgi mud, covered with te*-tre« buk, eketobed ftt Aigiuisui, Unralag, b; A. 0. H. 
in Sept. IS88. 

The only houses seen in Muralug in 1888 that had the appearance of being 
original were of two types. The simpler (fig. 131) consisted of a series of slender 
poles or rafters surrounding two or three sides of an oblong area about six feet in 
length, the upper ends were fastened tc^ether so as to form a roof-like framework, 
and in two instances a longitudinal pole served as a ridge-pole. The rafters were 
strengthened by being lashed to a purlin. In one instance a short pole was tied across 
the upper end of the entrance to brace the first two poles. Strips of tea-tree bark 
were slung over the framework, and in one ease were tied by rope to keep them in 
place. The sketches do not show any vertical poles to support the ridge-pole. The 
other type' (fig. 132) consisted of a sleeping platform raised two or three feet above 

» Dr W. E. Both (■' North Qaeeniland Ethnograph;, BulletlD 16," Record* of the Amtralian Miueum, vin. 1910, 
pp. 5C— 66} gives an aocouat of North Qneeosland hats, moit of whioh are droolar in plan. The forthsBt north 
at which he has observed the oonponte hnt (mj second type) "was on the Emble; Rivei, io the neighboarbood 
of the juDctioD o( the Palmer and Mitchell Rivers"; but in this case there are two ridge.poles, BDpported bj 
four [orked poles (fig. 42 and pi. XIII. Ag. 3), Di Roth s^b: "the ridge-pole designa with torked uprights ara 

d by Google 


the ground and supported on forked poles. Around and above this was erected a 

roof with a ridge-pole supported hy two vertical poles, lateral vertical poles supported 

the sidea of the roof, which consisted of several longitudinal bars or purlins lashed to 
the rafters ; the whole was covered with tea-tree bark. 

Fia. 132. Two bats, iai nud, od piles, akeMbed bj A. C. H. tX AigiDiHO, Mnralog, in B«pt. 168S. 

We were informed in 1898 that the old style of bouse in Muraliig consisted of 
little more than a simple gable roof of the b^'k of the tea-tree, tibu (Melaleuca 
leucadendron), with the gables filled in, resting directly on the ground. Sometimes they 
were small and intended for one &mily, in which case they had but a single entrance. 
Others were larger, about 3'5 — 1*5 m. (12 to 15 feet) long, and about 2 m. (6 to 7 feet) 
broad, and the house was so low that a man could not stand upright inside. There 
were three very low, small entrances on one side only, to these were given the following 
names, kala pasa, hinder door, dada paaa, middle door, and kttrubad pasa, a comer 
doorway (this was probably similar to that shewn in pi. XIX. fig, 3) ; inside the house 
in front of each entrance was a fireplace and along the length of the house was a 
clear space for the mats on which the people slept. Such a house would serve for three 

According to the accounts of the Mabui^ people, the houses of the Hoa natives 
were similar to those of Mabuiag, some of them had doors, and the Moa people were 
always ready to abandon them for a time and live in the bush among the great hills 
on the eastern side of the island, like their kinsmen of Muralug or the Australians 
of Cape York. 

pnlNtblj of F4PIIU1 introdaation ; indeed, (be lattei amngemeDt is oerUinl; ooDDscted with the iqiure tmnework 
bat met with only in the PeniDBaU and in the North, bat whether oonneoted in the w&j of progreu or retio- 
gnnioD it is impouible to mj." 

d by Google 


Central ^jrpe of Houw. 

The only illustratioa of the ordinary Western type of house (probably at Damut) 
is one by Melville, the artist who waa on the "Fly"; this is reproduced in pi. XXIV. 
fig. 2. The following descriptions by Jukes explain the drawing. 

"The huta on [Tani] had the appearance of a first attempt at a house, having side 
walls about two feet high, and a gable^shaped roof rising four feet from the ground, ^ey 
were about ten feet long and six feet wide, made principally of bamboo, and thatched with 
grass and leaves" (i. p. lOS). "The hut« [at Damut] were by far the neatest and best 
erections of the kind we had yet seen. Each one occupied a quadiangnlar space, ux to 
eight feet wide, and from ten to fifteen feet long. They had gable-sbaped roofs, eight feet 
high in the centre, and sloping on each side nearly to the ground. The frame of the house 
was made of bamboo, and thickly covered or thatched with grass and palm leaves ; the 
front and beck walls were also made of small bamboo sticks, upright and fastened close 
together, the front wall having a small triangular opening for a door, over which hang 
loose strips of palm leaf. The door looked into a little court-yard, of about ten feet square, 
in front of the house, strongly fenced with stout posts and stakes, interlaced with palm 
leaves and young bamboos, and accessible only by a very narrow opening between two of 
the strongest posts. In this court-yard was the cooking fire, llie different bute and fences' 
were rather irregularly disposed, but placed closely together, so as to leave only narrow 
winding passages between them. They occupied a space fifty or sixty yards long, by ten 
or fifteen broad. Behind them was the open place of meeting, on the other side of which, 
against an old tree, was a semicircular pile or wall of dugongs' skulls about three feet 
high. the middle of this was a conical heap of turtles' skulls,... At the south end of 
the huts we came to a building much superior to, and different from, any of the rest. It 
was like a Malay house unfinished, or one of their own smaller huts raised on poets to 
a height of sis or seven feet. The point of the gable was at least fifteen feet from the 
ground, the roof being supported at each end by two stout posts, about a yard apart*, 
having their tops ornamented by carved grotesque faces, painted red, white, and black, with 
much carving and painting below. The lower part, or ground-floor, of this building was 
open ^1 round, except at one end, where a broad, rudely-constructed staircase led to a platform, 
from which went the entrance to the upper story : this was floored with stout sticks, and 
at this end covered with mats ; this part was also partitioned off from the other by a bamboo 
screen. Under the roof hung old cocoa-nuts, green bougbe, and other similar things" (i. pp. 161—3). 
"About half a mile from the village [on the south side of Masig], wc came to a single hut, of 
a different shape from any we had yet seen. It was just like a great bee-hive, ten or twelve 
feet in diameter at the base, and the same in height, having a thick thatch of grass. A pole 
protruded from the summit on which was a large shell (fusus), and a small hole or door, 
partly covered by a board of wood.. .At Damley and Murray Islands almost all the houses 
are of this form, so that this had either been erected in imitation of them, or by some people 
of those places when on a visit to Masseed" (i. pp. 167, 168). 

' The fenosB have been omitted in Melville's diawiug, probably because they would have obsoored the hats. 

' These lateral posts recall the Eiwai arohiteatore. The term lateral is eudgested toi those posts whioh 
sapport a root on eaeh side of the median line, they are separated by a oonsiderable interral from the walls of 
the house. 


L Google 


MiiGgilliTre; found that the village on Nagir "consists of n single line of hut«, which 
probably wonld fumiab accommodation for, probably, 150 people. It is situated on the 
north-west, or leeward side of the island, immediately behind the beach, and ia front of 
a belt of jangle. The huta are long and low, with an arched roof, and vary in length 
from ten to twenty feet, with an average height of five feet, and a width of six. They 
consist of a neat framework of strips of bamboo, thatched with long coarse grass. Each 
bnt is usnally situated in a small well-fenced enclosure, and opposite to it on the beach 
is the cooking place, consisting of a small shed, under which the fire is made" (ii. p. 35). 

The inhabitante of Tutu live in a small cluster of houses of the South Sea type, 
more or less surrounded by fences at the southern end of the island, but at the 
Dortbem end we saw two huts (pi. XIX. fig. 3) which probably were constructed in 
the olden fashion. One consisted of a simple roof resting on the ground and supported 
by two central posts, one end being open. The other was better built and had a doorway 
at one comer. 

Macgillivray (ii. p. 41) noticed in Waraber, natives "sitting down under temporary 
sheds made by stretching large mate — the sails of their caooes — over a framework of 
sticks." Probably this was the usual custom when parties went to uninhabited sand- 
banks on turtling expeditions. 

FiQ. 133. Head-bonHe, Mabmsig, drawn b; OizQ. 

Three kinds of house8,.^were seen on Badu in 1888: (1) huts consisting of little 
more than two sloping walls meeting like a roof, evidently an indigenous structure; 
(2) a small house on piles, of the New Ouinea pattern ; (3) a large well-built grass 
house with neatly thatched walls, and a long verandah raised from the ground, this 
was erected and inhabited by South Sea men. The Badu people were to all intents 
and purposes identical with the natives of Mabuiag and had similar dwellings but no 
head-house of their own. Fulu (v. pp. 3 — 5) being the common meeting pUce for the 
inhabitants of both islands, and the centre of their military and religious life. 

Before the arrival of foreigners in Mabuiag the natives say that they lived in houses 
built on the ground and we have sketches by natives confirming this. From their 
account the houses varied in length and were generally not more than 1-8 m. (6 ft.) 
higL The floors were of white sand covered with layers of grass and mats, hence 
a bed, toie, was sometimes called apa-sik {apa, ground). The roof-walls were composed 
of grass and tea-tree bark. Some of the largest of them had as many as ten doors 
which were so low that it was necessary to go in crouching almost on all fours. They 
were built on a framework like those of old Mawata, and their section was simiUr 

H. Vol IV. 18 

d by Google 


to that of a Gothic arch. Women and children lived in these houses, the men, or at 

all events the bachelors, sleeping in the kwikwi-iut and kwod. With regard to the 

structure of these houses it may be confidently said that it was of the simplest. The 

median posts, aani-kag, were in vogue, not the lateral posts as at Kiwai, at any rate 

in the ordinary houses; the ridge-pole was called tbd. 

The name kuriktvi-iut signifies head-house in both senses of the word. The description 

of the head-house by Gizu was supplemented by a drawing by him (fig. 133), and one 

by Tom is reproduced in fig, 134. It appears that the bouse was very similar to the 

long houses of Mawata in 1888 and to those of the Fly delta. It was built on piles, 

and had a door at either end, and reddened skulls and jaws were bung around. (In 

1888 one of us saw bunches of skulls hanging underneath houses in Mawata.) Oco-nut 

palm leaves were placed on the roof, gable ends 

and eaves, and the posts, pasi kag, were painted 

red. It seems that there were originally two 

of these houses in Mabuit^, one probably belonging 

to each phratry {v. pp. 172, 306), in them arms 

were kept and the skulls were those of enemies 

, .|i 1 . -nr JUL . Fio. 1S4. H«»d-hoaM, Mabniag, after % aketob 

killed m war. Women and small boys were not ^ '.^.^^ ^' 

allowed to enter them; for further information 

concerning the use of these houses the reader is referred to Vol, v. p. 306. 

Fio. 18G. Kttod kt Bm, Mabniag, sketobed fa? A C. H. in 1BS8. 

A kwod (fig. 135) was still to be seen in Mabuiag in 1888, it consisted of four 
walls with one wide entrance and a light flat roof made of coco-nut patm leaves; it 
was about 75 m. (25 ft.) long, 4'5 m. (15 ft.) broad, and 1'8 m. (6 ft.) high. Several 
authorities declared that the kwod served as a place of meeting when the kwikwi-ivi 
was too hot or overcrowded, and that the men and boys of a certain age slept in 
either indifferently. In October, 1888, a number of natives from other islands visited 
Habuiag, and the bachelors took up thoir quarters in the kviod. 

d by Google 

The head-house, eo br as onr knowledge goes, was confined to the western islands, 
at Mabuiag we were informed that there was formerly one similar to theirs in Saibai ; 
the house on piles in Damut (p. 96) and the hut on Aurid (v. p. 378) and to a 
certain extent the much modified totem shrines on Yam (v. p. 373) were probably of 
a similar nature. The kwUcwi-iut iB unmistakably related to the club-houses of New 
Guinea. The kwod hut at Bau in Mabuiag was a much less pretentious shelter house 
for the youths and unmarried men, and may be regarded as being originally a sort 
of annex to the head-house which survived the destruction of the latter as it served 
the useful social purpose of a club-house at which young male visitors could be 
put up. 

For the significance of the Western kwod and the use to which they were put 
the reader is referred to Vol v. pp. 3—5, 366—7, 370, 373—7. 

We had a circumstantial account of an aijitai^ or round house, but this was 
afterwards denied. It may be worth while, however, to repeat what was at first said. 
The ai-mai were very few in number at Mabuiag, they were constructed, so fitr as we 
could understand, on lines similar to those in use at Mer, had a very small low door, 
were Scored like the other houses and had the tops of their centre poles where they 
projected above the roof carved into figures representing men. Moreover it was asserted 
that they were the ancestral types of one or two s<]uare South Sea houses at present 
existing. A peculiar St Andrew's cross ornament of bamboo which occurred in two of 
the old houses at Mer did not so occur, it was said, at Mabuiag. It is just possible, 
supposing our informants to have invented the aimai at Mabuiag on the lines of those 
which they had had opportunities of seeing at Mer, that these peculiar but conspicuous 
additions to the inner frame had escaped their notice — but it does not seem probable, 
and if so they should have appeared in their description of the aimai, or, at least, 
would not have been definitely rejected after a leading question on the subject. Again 
the centre poles were said to have been carved like a man, and no central pole that 
we saw or heard of at Mer was ever so carved. Oizu made a drawing of an admai for 
us which evidently represented a round house on piles and a ladder leading up to it. 
Altogether, though the existence of the aimai was denied by some, we appear to require 
further proof of its non-existence. On the other hand we have no confirmation trom 
English sources of the presence of round bouses on Mabuiag. 

Northern ^pe of House. 

The old style of house of the most northerly of the western islands, we believe, was 
invariably built on piles; in 1888 the South Sea type of house was beginning to 
snpplajit the other, but some of the older type of pile dwellings are still to be found 
in Saibai (pi XX. figs. 1, 2). In 1871 Mr Murray found that on Saibai and Dauan 
the houses were "built on stakes eight or ten feet high." All the houses which he 
saw on the islands and on the mainland were built mainly of bamboo, and the roofe 
were thatched and the sides enclosed with the pandanus leaf (Forty Tears' Mission 
Work, 1876. p. 456). 

> Aimai seems to be the plnnl of ai, anoeitors. In tha ToMbnlsiy iaiead is a Tonnd house. 


d by Google 


The general character of the northern houses can be seen from fig. 136 which 
is a sketch of a house in Saibai, made in Aug. 1888. At that time the natives of 
Boigu had heard a rumour that the Tugeri pirates were coming on a head-hunting raid, 
and, knowing that they could not withstand them, 6ed to Saibai. There was not enough 
house-room in Saibai for these visitors, and so the under portion of this house was 
roughly wattled with leaves of the coco-nut palm for the accommodation of some of 
them. At that time this was the only two-storied house in Torres Straits. 

It is very probable that the small house, with the roof coming down to the Boor 
and a door at one end (pi. XX. fig. 2), is the original type; a precisely similar one 
was seen in Badu in 1888. The larger oblong houses with a side door (fig. 136, 
pL XX. fig. 1) rather look as if they were a cross between the old type and the 
South Sea type of house. 

FiQ. 1S6. Pile dwelling at Saibai, with the piles tempanril; BnrroDDded bj a wattle of palm leaTes; 
sketched h; k. 0. H. in Lag. IRSS. 

Two houses of the old type in Saibai had respectively the following measurements: 
length, 31 ft., 32 ft. 5 in. ; breadth, 16 ft. 9 in. each ; height of wall, 5 ft. 3 in., 5 ft. 
6 in. ; height of gable, 16 ft. in one house ; width of door, 2 ft. 6 in., 2 ft. 3 in., in 
each case the door was not quite in the centre of one side ; height of piles, 4 ft. 7 in., 
6 ft. 8 in. 

One house was for the use of unmarried girls, this was described as "mission 
fiishion"; we believe that this was instituted by the missionaries in other islands, and 
we know that the MacFarlanes had one in Mer. They said at Saibai that in the old 
days there was a men's house, loto, the ordinary word for house is lag. 

d by Google 

Eastern Islands. 

Among the natives of the eastern islaDds, Uga, Brub, and the Murray Islands, 
another form of house, rneta, was found, confined so &r as we know to this particular 
intermanying group. Some of the earliest English voyagers in Torres Straits mention 
and illustrate the bee-hive round houses, kaubkaub meta, of these islands, but their 
descriptions are as a rule too meagre to be worth reprinting \ At present (about 
July, 1898) there is a small one at Damley, and a large one is said to exist on the 
north or north-west of the island. At Mer there is only one of moderate dimensions 
and this has recently been vacated by its owner, and probably soon ceased to exist. 
There is also a small find deserted hut which has features in common with the round 
house and which may possibly form a connecting link between it and the oblong 
structures of the western islands. 

The dimensions of the round house at Ulag (pi. XX. fig. 3), Mer, are as follows. 
Circumference outside (at base) about 62 feet. Radius inside 9 ft. 6 in. Height about 
12 ft. 6 in. inside; the central pole of wood projecting above the roof to a total height 
of some 14 ft. 6 in.' Round the top of the centre pole a spiral of slender vine is 
wound and this forms the only exterior ornament at the present time. Formerly a large 
Fusus or Cassis shell would be set on the top, and, in some cases at least, the elaborate 
structure shewn in pi. XIL fig. 3 was added. The thatch was composed of dry grass, 
and the bundles were gathered in together at the top from which the centre post 
projected about eighteen inches ; it was laced on to the laths of the framework by 
long creepers, one man being inside, another outside, and passing them through by 
means of an eighteen inch needle of bamboo, arem tu or atviar lu, notched at one 
end^ A feast was held in connection with this thatching, and often a number of friends 
would take part and race side against side. 

The first operation in the building of the round house* was the delimitation of 

' Aooording to Jnkea (l p. 197) "The whole *hore here [N.W. dde at Mar] was lined with a oantinaonB 
TOW of hoDses, each in ft Bmall eonrt-yard ot from ten to twenty yards sqaare. The houses were the [p. 198] 
■aine u those of EiTO(}b....Tha hoDses [are] perhaps larger and more oomplete than at Erroob. They seemed 
Tery clean and neat inside, with raised platforms, ooTered with mats tot bed-plaees....The tops ot the booses, as 
also the fences of the coort-yard, were ornamented by large white sbells, and oocasionally a skull or two 
was sasi>ended somewhere near the honee, ot placed on the stamp of a tree and painted t«d....Hete and there 
between tbe fenoet of the hots were left nattow passages, giving aoeess to the land at the back, where there 
were some small plantations." Bampton saw in Emb a boose, SO x 15 ft., raised on piles 6 ft from the ground ; 
it "was the sole hat in which there were no sknlls ot hands" (Flinders, Voy. i. p. ixivi). 

* A ronnd boose at Lai in 1669 had the following dimenaioos: oireamferenoe, 66 ft.; diameter at doot, 
SO fL S in.; transreiM diameter, 18 ft. 6 in.; height. IS ft. ; door, 3 ft. 6 in. high and 3 ft. wide. 

■ The bamboo tbatohing needle which we obtained in 1SS9 is 91 cm. long and is pieioed with a hols 
(iKKm, I. pi. 333, no. 10). 

* Jakes (i. p. 245) gives the following description of the baitding of a house in Emb : " Eight oi ten 
stent posts abont five feet high were driven into the gtound at eqaal distances, forming a cirole of fiReen 
feel in diameter. Bonnd these, at eqoal heights, were fastened three hoops ol bamboo, both inside and onlside 
the posts, bat a space in the lower hoops was left between two of the posts, where the low door woold come. 
They afterwards fasten tall poles of bamboo aptight to the hoops pretty closely all roond, and bringing their 
ends together, tie them to a stont centre pole, which rises np from the interior and protradea throogh the 
roof. On to this framework, they weave and fasten a very thick thatch of gtass, and palm leaves split into 

d by Google 



the circumference. This was done by simply making a trail with one foot in the 
sand. Twenty stout posts, teter, were then put in at intervals of about three feet so 
as to form a circle, they projected above ground five feet. Four split bamboos were 
fastened round these posts, the uppermost band being near the top. To these Icosker 
teibwr (fig. 137) were then tied the pik in four sections, each section being done by 
five men. The general direction of the pik is transverse Irom side to side rather than 
fi^m bottom to top. The wooden main-post, seaeri, was then dragged in through the 
opening left for the door and set up in the centre. The piJc alone, arranged as they 
were in four groups whose ends overlapped and formed diagonal patterns, gave the 
house almost the appearance of a rounded quadrangle. There were fifteen pik in all, 
four a side, the number being reduced to three on the eastern section which formed 
an arch over the door. The ends of these fifteen split bamboos were tied outside the 
first two [? the upper two] rows of kosker teihur and must have added enormously 
to the strength of the roof Next the wooden lemlem (which correspond to the rafters 


Fio. 187. Qeneral and det&iled Beetione of a Miriam 
lOmtd honae, bj A. W. Boale 8 It. to 1 in. 

Fio. 186. k and 
roDDd honae; i 
eud iwla on Me 

, Elevation and plan of ■ Miriam 
and D, Eleration and plan of an 
, b7 A. W. Scale, 16 ft. to 1 in. 

of the Eiwai house, p. 112) were fostened to all the kosker teibur so that the lower 
ends rested on the ground while the apper converged and were tied to the centre 
pole at the top. The underlying pik which they crossed served to keep the curve 
true and symmetrical. Laths {turn pik) of split bamboo were added and the frame 
was ready to receive the thatch. The door was triangular and faced the sea (which 
lay to the east). It measured 2 ft. 3 in. high x 2 ft. 3 in. across the base, and its aide 
posts were neatly bound with grass. Such a doorway could only be entered on all-fours. 
At the centre of each hemisphere enclosed by the pik was a St Andrew's cross of 
split bamboo measuring about 3 ft. high x 1 ft. 6 in. across (between perpendiculars). 

I imall opening I 

' tbe door, to enter whioh the; mnit oroDoh on th« bands 

tbin itripa, leaving only ( 
and kneu." 

The cicoolai hnta of the North Qnoenaland natiTea dcMribed by Dr Walter E. Both (I.e.) are ol mncb n 
■imple oongtrnction. 

d by Google 


The thatch at the bottom was called maiau, that at the top akur (the former term seems 
to be used for any kind of rough shelter). 

This bouse bad originally contained three sleeping platforms, one of which filled 
in the whole of the western half and was composed of a bamboo tied to two of the 
side poets and to the centre pole at 2 ft. 6 in. from the floor, with the ends of four 
transverse bamboos resting on it, their other ends being tied to points in the western 
wall These transverse bamboos were fortber supported by another which, crossing 
under them at right angles, was tied at one end to a wall poet and at the other 
to the platform which lay to the left on entering. This second platform covered the 
fiist in part (see fig. 138 B) and its frame was supported by the same bamboo that 
carried the latter, and by the side posts. It was thus an inch or two higher than 
the fiist pUtform and was completed for a depth of eighteen inches by a number of 
slips of bamboo. The second platform did not come up to the central post by about 
two feet^, and the remains of a corresponding frame on the right side seemed to shew 
that it too bad been so constructed as to leave an open space in front of the doorway, 
which thus measured about nine feet long by four broad. In it was the fireplace. 
The natives said that these houses might contain six platforms for sleeping and storing 
goods, three being added at a height of about 2 ft. 6 in. above the lower, their frames 
resting on the tope of the side posts (teter) and the highest of the four kosker tabur. 
They added that the round houses were much warmer than the South Sea buildings 
in which they now live. 

The door was closed in this case by a slab of wood. Outside it was a solitary 
post about 10 ft. high, the survivor of six which had formerly stood three each side 
{tera-cUatva-lu). Groups of these houses were enclosed in an uteb or compound and 
a high fence served to keep out the wind. Formerly a bunch of mottled cowrie 
shells (mdfcep) were tied over the door to give notice of the entrance of a visitor. 

There is no reason to suppose that the Miriam have materially altered their 
domestic arrangements simultaneously with their houses, and probably a large round 
house might have accommodated more than one family just as some of the South Sea 
houses do now, though the majority were and are the exclusive property of one man 
and bis belongings. 

The only other old house at Mer, that of Kopam (8 a) at Sobatir, an area in 
Kop, bad been long deserted and was in a ruinous condition. Fortunately enough of 
ite structure remained to connect it for certain with the Miriam round houses, and 
possibly with the type found at Mabuiag and the western islands. Its ground plan 

' In the sketch b; FrotMsoi Haile; ot the interior of a hoaae in Erub (M«agilliT»7, ii. p. 47) there ii a 
well-made platronn oppoeita the eotrsnoe which eitenda aoross the hnt and np to the oentrel post, it appeara 
to be raised nearly four feet above the ground. To the left is a mach lower one, the bamboo flooring of which 
extendi as far as the level of the oeDtral post, bot the platform does not come close op to the post, as in Gg. 
138 B. At a height of aboDt sii feet several poles are tied to the central poet, the other ends being apparently 
sapported by the framework of the hoase ; from these bags and varions objects are enapended. 

According to Ifr J. Bmce the six beds were in two tiers, the two long ooes extended from side to aide 
Ixiiittd the central pole, and it wsa od their tcaai supports that the flsh-speara were aaapended horizonlallj. 
Od each aide of the entranoe were two shorter beds. The flieplaoe was between the central pole and 
tlM entranoe. 

d by Google 


was •rectangular with one rounded end (fig. 138 D, and pi. XX. fig. 4). In that end 
all the structural peculiarities of the round house repeated themselves. For the rest 
the name eud meta or " old house " well described it. The chief measurements were : 
length 14 &,., breadth II ft., height 8 (t. Nine poets 3 ft. 6 in. — 4 ft. high supported 
. the main &ame, three being in the rounded end, three to each side. There was only 
one ridge-pole and this was carried on the crossed upper ends of the rafters. A section 
of the side wall (fig. 139) shewed two courses of kosker teihur on which were the 
rafters {lemlem). Three longitudinal epatired hi strengthened the rafters between the 
kosker leibur and the ridge-pole. The ends of these were bent down where the curve 
was reached at the end — its depth being about 2 ft^ 6 in. — to meet the koaker teibur 
on their outer edges. The whole of the round end was precisely similar to the round 
bouse down to the St Andrew's cross, but a glance at figs. 26 — 30 
will shew that the sides too were not really constructed on a difierent 
plan. A shell, Tiasir (Trochua niloticus), decorated the intersection of 
the arms of the St Andrew's cross, the longest of which was just 
under three feet. Formerly the open front was screened towards 
the sea by a coco-nut leaf hung on either side of the arch. Behind 
this house about four feet from the back was a slight platform on 
poets 4 ft. 8 in. and 5 ft. high. Its length was 5 ft. 6 in. ; breadth _____ 
2 ft. The long bamboos at the top were fitted into forked posts ir,o. 139. Seetion of 
and projected beyond them a few inches. Five bars across bad the roiuided end of 
originally supported six large giant clam, miakor, shells disposed in »•» eud meta od Mer, 
pairs, of which four were left. These big clam shells were used to toiia 
catch rain and to hold the supplies of fresh water for daily use. The 
fireplace of the house had been made anywhere, but the freshest ashes were in the 
toiddle though slightly nearer the closed than the open end. If the Kop house does 
not form a connecting link between the dwellings of Daudai and those of Murray and 
Damley it certainly has all the appearance of being one. 

Mr Ray in the course of his linguistic investigations obtained the following order 
of procedure in building a house: 

Le meta ikeli, man makes a house ; e tonar detar kikem tetem, he first draws 
the plan with his foot ; e daivti a teter ekoa, he digs holes and erects side posts ; 
e kosker teibur lageru didbar, he ties on the horizontal bars with rope ; e ditimeda 
totge bakedida s^e a pik didbar, he begins at the top and goes down (arrives) to 
the bottom and ties on the uprights [arched bamboos] ; e seaeri ekos, he erects the 
centre post ; e lemlem eniir e ditimeda totge, he puts in the thin upright rafters, beginning 
at the top; e turn, pik egawi, he. ..the horizontal laths behind; e akuru derem, e ditimeda 
sebge, he laces on the thatch, beginning at the bottom ; e sHe bau didbar, he ties together 
the framework of the bed; « marep epat sikem, he... bamboo for the bed place. Meta ikerer 
emetu, the house is made. 

Mr J. Bruce has recently sent the following description of the old type of Miriam 
house, or kavhkaub meta, "round house." The frame was made of bamboos, marepu 
darot, these were fixed in the ground in a circle and bent over and fostened to a 

d by Google 


stout ceDtral wooden pole, aeseri. Laths of split bamboo were laehed horizontally to 
the bamboo frame to bind it together and for the attachment of the thatch. The 
lashing used for fixing the bamboo frame to the centre pole was a strong vine, boe 
(Flagellaria indica), and diy pandanus leaves, ahal, were employed for tying the laths to 
the frame. The thatch was made of a strong grass, akur; the men pulled or cat the 
grass on the grass lands, but only the women tied it into sheaves and carried it to 
the building site. The stem of Elntada scandens, siret or strip, after being beaten and 
twisted to make it pliable was used for tying on the thatch. The door, pao or pan, 
for closing the entrance was always made from the side of a broken canoe, pao or pau. 

The houses were all built on the ground, and the interior was generally fitted 
with six or eight bunks for sleeping on or for holding their treasures and implements, 
thus serving instead of boxes. One bunk was placed above another on each side of 
and opposite to the door. They were made of bamboo poles, and layers of split bamboos 
were laid across and spread out hke laths, marep epat or sik epat. The lower bunks, 
wt aik (sand bed), were raised about 150 mm. (6 in.) above the ground and the upper 
ones about 1*2 m. (4 ft.); between the two was a rack, on which spears, bows and 
arrows, etc. were placed. Dancing gear and other small articles were hung on the 
walls. Painted trumpet shells, maber, were hung on the wall over the entrance, m^a 
taier etkoper (house decoration). Afr«r a death the desiccated body, keber, was fixed 
to the central post, facing the entrance (vi. p. 148). The fireplace was situated between 
the central pole and the doorway. 

A dozen or so of saplings, adige meta etkoper (outside house decoration), were stuck 
in the ground, six on each side of the door, and on the top of each a painted trumpet 
shell was placed. A. very large painted trumpet shell was placed on the top of the centre 
pole, or sometimes one or two large Cassis shells, as in the Erub house shewn in 
pi. m. fig 3. 

The houses were made of various sizes according to the number of families or 
individuals who would probably occupy them. A small house would be about 4*6 m. 
(15 ft.) in diameter and about the same in height. A house for occupation of two or 
more families might be as much as 91 m. (30 ft.) in diameter and twenty feet or 
more in height. 

In the Vocabulary there occurs the word makerem meta, house for young men. 
Unfortunately we cannot give any further information on the subject, it may refer to a 
house in which the young men, khsi, were lodged during the initiation ceremonies. 

The Modern Houses op Torres Straits. 

With regard to the modem houses in which the natives of Mer and Mabuiag 
now live there is no necessity to say much. They are all constructed on models 
introduced by various Polynesian and Melanesian strangers, and differ fundamentally 
from the old types in nearly every respect. At Murray there are a few houses built 
on piles constituting a combination of South Sea and New Ouinea architecture, and 
somewhat resembling houses at Hula and elsewhere in New Guinea where foreign 
infiuence has been felt. There is no continuity in the evolution of these bouses. A new 

H. VoL IT. 14 

d by Google 


type was iotroduced and immediately accepted — not so much because it was better than 
the old (for such was not the case in Mer at any rate), as because it was asaociated 
with the new order of things that the traders and miBaionarieB brought with them. At 
Mabuiag the influence of the "teachers" from Mare, Lojralty group, may perhaps be 
traced in a few square houses with a central poat. 

The following is a list of the terms' employed in house-construction, the Mabuiag 
word comes first and the Miriam equivalent is in brackets : 

Beam of wall, inner horizontal : bal-ua-pui (kerem-teter), — outer horizontal {kosher 
teibur, mui pHe). Cross-beam: baX kaputai pui. Purlin of roof: bal-iai-pui (mtii piJc). 
Door : pasa (pau). Doorway : pasa, pasa-gud (te or meta te). Door-jambs : pasa-ffudati 
tttda [or tbd] {te-lu). Poets outside door: (t^a atatmi tu). End of house, comer: 
kuruhad (kop). End of large ridge-pole : jrmi, — of small : kvt. Lath supporting thatch : 
bera pui (turn pik). Lean-to house, porch, verandah or eaves: »wi {maistt). Lintel; 
poBo-gudau tuda [or tbd]. Median or central poet : aaru-kag {seseri). Pile : pasi kag. 
Platform-bed (aik or sik bau). Posts of inner aide of house: pari kag (teter), — outer: 
pasiu pui (teter temlem). Rafter : aau, — inner : barbat or babado puidai bera pui (am» lu), — 
outer; kadaka tra pui {lemlem). Ridge-pole, inner: tdd (tot). Roof: toit or tbd (tot 
or meta turn). Thatch : borda or frurdo, kamug, mugud {ahir, maieu), — inner thatch : 
bai or bdi, — thatch-bands : bordau pui {marep pik). Tie-beam : bal-iai-ptti. Wall : past 
(bir). Window: arkat. 

I ' 


Fio. 140. DebulB ol the oonstrnctioii oC modem honaw in Torrei Stnuta. A, Mer; B, Msbniae, by A. W. 

The mui pik, which should perhaps have been placed higher up in fig. 140 A, is 
laahed to the oris lu with neud lager (neud has not been identified); the kerem teter 
is very securely lashed to the oris lu with bot lager, a strong rope made of the 
stem of Flagellaria indica, it evidently prevented the beams from slipping off the 
posts, teter. 

At Mer there was a building with a central post and diamond-shaped ground plan 
(fig. 141 A, b), but there was no evidence whatever to connect it with the round house, 
1 The following ii a trftnBlation or Mtas oC these worda, we have not foond & meamng for the othere. 
Mabniag: arlcat, hole; babado, poM. of babat, toftti's Bister or womao's biothei, etc.; bal, ewtom; bera, rib; 
bordau, bardo, thstoh, u, ot; gad, mouth; tai, lie Klong; kadaka, opward; tag, post; In, thin^; pati, side; 
jmi, etlok, pi. puidai; tra, ridge. Uer: irmn, head; koftcr, woman; marep, bamboo; muf, iniide; Uibia, 
inside; te, month; tOer, lover leg oi toot; (urn, (op. 

d by Google 


and it was very likely, what its owner Dick claimed for it, a freak of his imagination. 
Its greatest length between comers was about 35 ft., and a line connecting them would 
have lain roughly parallel to the seashore. Its greatest breadth was about 21 feet. 
Each of the four sides measured 20 ft. and there was an opening in the two which 
&ced the Bea about 3 ft. from the extreme ends, 4 ft. 6 in. x 2 ft, closed by ordinary 
wooden doois set in frames the height of the side walls. The height of the house 
to the topmost point would be about 14 ft., and the central pole projected somewhat 
beyond. Several margp pik bands confined the thatch, and palm stumps bad been 
utilised in the construction. Inside were two platforms (see plan) about 2 ft. 6 in. high. 

&i > 1 1.,)' 

■;■.■. ii.iii|ii..u. 

Via. HI. EleTRtion and groimd plui of two modem Hiriam hooMi, b; A. W. Scale 16 ft. to 1 in. 

The Miriam house shewn in fig. 141 c, D, measured 15 x 11 x 9 ft. and was some- 
what peculiar in having the ridge-pole the same length as the side walls so that 
there was no roof at the two ends, the wall being in one piece. Two median posts 
as usual carried the roof, and there were marep pik to &sten the thatch. The door was 
5 ft. X 2 ft. and opened outwards leaving a narrow .space by which to enter. The three 
platforms were 18 in. high and the fire about central (fig. 13). 

Fig. 142 A represents a Miriam house not yet finished, but the skeleton made it 
easy to take accurate measurements and to observe the construction which was typical 
of all the "South Sea" houses of Mer, though the height of the side walls was 
somewhat abnormal. It measured 28 ft. x 16 ft. broad. Its height was 14 ft., that of 
the walls 8 fb. The house was parallel to the sea though some hundred yards fi^m 
it, and its back was turned to the road. On the sea front was a door 5 ft. 6 in. x 2 ft. 9 in. 
at a distance of 11 ft. 9 in. from the north comer. The median posts were three 
feet in circumference at about a yard fix>m the ground. They stood about 8 ft. 6 in. 
from the end walls and in the middle line. The maia ridge-pole was mortised into 
their tops and projected 18 in. at either end. The second and exterior ridge-pole 
rested on the upper ends of the aaia lu {cL fig. 140 a), while the lemlem converged at 
the top and were tied to the interior main ridge-pole, flush with the asis lu. The 


d by Google 



kosker te^mr were three aside, and the mut pik two. The end walls were similar in 
construction. The side posts were hardly inferior in solidity to the median posts and, 
as this is the rule in Mer houses, the superfluous strength of the side walls would 
seem to account for (he absence of defects consequent on the neglect of tie-beams 
other than those afforded by the end walls. The construction of the side walls which 
is &r more massive, though hardly more complicated, than that of the round house, 
is best understood by reference to the section in fig. 140 A. The thatch was laid on 
in bundles, beginning at the bottom where a different name was applied to it, maitu, 
from that used for the upper portions, akur'^. This method did not seem to be due to 
anything more practical than the dictates of fashion. Windows did not exist before 
the introduction of "South Sea" architecture, and they do not vary much in size or 
position from the examples here quoted. The side walls in fig. 142 a are higher than 
usual ; ordinarily dooi^ are of the same height as the walls, and they may or may 
not have a wooden fi^me and threshold as in pL XXIII. fig. 2. 

A yT 









Flo. 143. EleT^tiottB of modern Miriun boiue*, \>y A. W. Soate 18 ft. to 1 in. 

Fig. 142 B is a typical " South Sea " house (that of Jimmy Rice and Debe ' Wali). 
Its dimensions were 40 ft. x 18 ft. x 13 ft. high. The side and end walls and doox were 
5 ft. 2 in. high. The doom were in the middle of the sides facing the sea and the 
road respectively, and were about 2 ft broad. That facing the road had been temporarily 
closed up as it was not much used. The walls were of layers of plaited palm Reaves 
which overlapped one another, while the roof was of the ordinary thatch. The ends t^ the 
ridge-pole were 9 ft. from the end walls of the house, thus making a hipped roof as in 
pi. XXVni. fig. 2, and the thatch over the ridge-pole was finished off neatly with a ^trong 
plait. Two windows &ced the sea, close under the eaves, 2 ft. x 18 in., and in the piCeition 
shewn in the figure. Inside the bouse was a floor raised 15 inches above the sand. Piles, 
hau, supported cross pieces, mux hau, and these transverse planks, sik bau, both the- latter 
being of split bamboo. This platform covered the whole floor of the house except 
a gangway four feet wide from door to door. The two halves of the hau where* they 
fixinted the gap were finished off with a single bamboo, hau, laid crossways. In the 

' A bamboo icmfFold wkb used in tbatobiag. A lingle pieoe 16 — 20 ft. long is tied to two npriglita abont 
13 ft. hi^ Bt a beigbt of perhspa 9 ft. Th« whole Issna agaimt the eide of the hoiue and the thatober itands 
on the homontftl bar. 

d by Google 


middle of this central gajigway was the fireplace. In most Kouaee the platforms were 
higher and confined to a more limited area, which seemed to vary entirely according 
to the taste or the requirements of the inhabitant. 

A very similar, but rather larger, house was that of Sisa. As usual it faced the 
sea and had its back to the road. Dimensions, 44 ft. x 28 ft. x 13 ft., side and end 
walls measured 5 ft. 9 in. high. A door 2 ft. wide and two windows 1 ft. x 1 ft. 6 in. 
were on each of the sides, all in wooden frames. Here the windows were placed 6 in. 
below the eaves. The tnarep pik band consisted of two horizontal strips of bamboo 
between which were short strips arranged trellis fashion (pis. XXIII. figs. 1, 2, XXVIIL fig. 2). 

A few houses at Mer were built on piles, probably, as the natives themselves 
explained, in imitation of the New Guinea buildings which many had seen in the course 
of their voyages in trading schooners or mission boats. Of these, Marau's house, fig. 143, 
is an example, and one is also seen in pi. XXI. fig. 1, 31 ft. x 23 ft;, broad over all 

Fia. 14S. Side and end elevation at Marau'B pile dwelliag on Mer, bj A. W. SoaJe IB ft. to 1 in. 

X 13 ft. high. The piles were 4 ft. 6 in., and palm tree stumps had been utilised. 
There were two verandahs, one -at the back facing the road, the other at the fi^nt 
which &ced the sea. They were 3 ft. 6 in. wide. The doors (that at the back closed 
up with palm leaves) were in the middle of each long side, 3 ft. wide and 4 ft. 6 in. 
high. The roofe of both verandahs were added outside the main roof of the house 
and were obviously an addition to it. Though they doubtless formed part of the original 
plan, the clumsiness of this overlapping arrangement (which would allow the rain to 
&11 off the main roof into the verandahs) shewed that it was new and unfamiliar. 
The verandah eaves overhung about 18 inches. The median posts projected 2 ft. 6 in. 
above the roof and, being of bamboo, were incapable of decoration. The thatch was 
normal. A six-rung ladder led to the door at the back. Under the eaves ran the 
usual marep pik band. The floor beams were of bamboo, and the top of the thatch 
plaited together. 

Two other similar houses belonged respectively to Capsize and Sali, and to. Beni, 
Dela and Qisur. They stood on an eminence above the road facing the sea. Neither 
had a verandah at the back as the hill-side was too steep to admit of it. The median 
posts projected about 2 ft. and were decorated, so it was said, after the old style 
(pi. XII. fig. 3). The doors were of split cane fastened in wooden ftames and presented 
a very neat appearance. There were also shutters to the windows. One house had 
side walls of plaited palm leaves, the other had the usual grass thatch. Both houses 
were examples of divided tenancy. 

Fig. 142 D is an end view of an unfinished house (Komaberi's) 30 ft. 4 in. long 
X 11 ft. X 13 ft,, high. The house faced the sea. The side posts were 5 ft. 6 in. high. 

d by Google 


The door, 13 ft. from the north comer, was 4 it. wide. The roof began to &11 bora 
the ends of the ridge-pole 6 ft. 6 in. from either end. The median posts were 9 ft. 
and 9 ft. 6 in. from the end walls. The ridge-pole was thus about 20 ft. long and 
3 ft. 6 in. in circumference, mortised into the median posts (2 ft. and 3 ft. 6 in. in 

In Fig. 142 C is shewn a small store-bouse (of Jimmy Dei) 15 ft. x 10 ft. x 9 ft. 
high. The thatched walls were 4 ft. high and had the usual marep pik band. The door, 
in a wooden frame &cing the sea, was 2 ft. wide and only 2 ft. from the north comer. 
In front was the kar, fence, 2 ft. 6 in. from the front. The ridge-pole ended 2 ft. 6 in. 
from the end walls, and was carried on a single central post. All the frame and posts 
were of bamboo and the Mer lemlem and lemlem were absent, as the weight to be 
supported was small. 

Within were three platforms disposed about the end furthest from the door. That 
on the south-east side (that of the door) was 7 ft. 6 in. long x 4 ft. broad ; the end 
platform was 10 ft. x 2 ft. 6 in., and the remaining sik hau 2 ft. x 3 ft. All were a foot 
above the sandy floor, and were apparently used chiefly to cany firewood. The house 
was interesting chiefly because of its simplicity of structure and because it was one 
of the few examples of a specialised out'house. Formerly, in the round house days, 
a rough lean-to, eud meta, sufficed for a similar purpose. 

Most of the houses at Murray Island were grouped together in compounds, ut«6, 
within a fence, and this arrangement seems always to have prevailed. On the other 
hand, at Mabuiag, the houses form regular streets and are arranged with a considerable 
amount of regularity. The old village may hkve been arranged in a somewhat similar 
manner, but we know that the missionaries brought practically all the families together 
to form a single village in the island. Oiam or mvdau giam is the name for a house-site 
and the cleared space or street between the houses is called giigu. The evgu is common 
property, when necessary it is cleaned by the women, each woman tidying the space 
in front of her house as &r as the middle line; the coco-nut palms growing in this 
sugii belong to their several planters. 

Fig. 142 B is representative of the South Sea houses at Erub and Mabuiag, though 
on the latter island occur a few examples of houses similar to fig. 141 A, but square, 
which were said to be descendants of the aimai (p. 99). 

The above notes on the South Sea houses of Mer are equally applicable to Mabuiag 
except in BO &r as the Mabuiag houses have a proper median post roof and tie-beams, 
which however do not tie the side walls together so much as support the kag, posts 
(cf. fig. 140 B). In all other essentials the framework is really identical. Bed platforms 
are rare in Mabuiag bouses, and the thatch is put on alternately "head up tail up" 
to avoid leakage. Two kinds of grass, mugvd and kamug, are mixed together to render 
the fabric more durable, kamv^ being the coaraer and stronger of the two. Tea-tree 
bark is also used in the thatch. The dimensions of a typical house (that of Tom, 
fig. 140 B) were as follows : length 38 ft., breadth 18 ft., height 14 ft. Walls 5 ft, 6 in. 
high. Two doors (in the middle of the sides) measured 5 ft 6 in. x 2 ft 6 in. A single 
window in the end wall was a foot square. Various mats were used to cover the floor. 
The tie-beams, haUiai-pui, "the wood that stretches across," although they obviated the 

d by Google 


necessity for clumsy median posts, threatened the heads of unwary visitors and their 
stractural advantages were not apparent. 

The law of succession is the same in Mabuiag for house-sites as for gardens (v. p. 284). 
Sometimes if a man had two or more houses, an eldest son would get one to himself, 
but usually the children all kept together. The loaning of houses or house-sites in 
Her is described in Vol. vi. p. 166. Mr J. Bruce has recently sent the following 
information : 

Newly married couples in Mer rarely possessed their own house, but had a house 
lent them or more frequently shared a house with their parents. The reason for this 
was that formerly marriages were arranged so hurriedly that there was no time to see 
about a house. When they decided on building, the husband prepared the bamboos and 
vines, and he and his wife carried them to the building site. When all was ready be 
gave notice to bis friends what day he proposed to erect the frame, but usually his 
village group or relatives talked it over first. If he was on good terms with his wife's 
friends they would come to assist. A feast was given when the frame was erected and 
another on the completion of the thatching ; very frequently a third would be given 
when the house was in order, Toeta dilik lewer. The law is now enforced that a young 
man mast be provided with a house before he is allowed to many, this was introduced 
as a sanitary and moral precaution. The Miriam are an extremely sociable people and 
love to converse in groups, and they were inclined to overcrowd their houses with persons 
of all ages and belonging to different fomilies. 

Houses of Kiwai and Daudai. 
The Lon^-houM of Klwal. 

The bouse, moto, at laea, of which the following is a fairly detailed description, may 
probably be considered fairly characteristic. The Eiwaians have been subjected to few outside 
influences. Their retention, for the most part, of the old native dress and of great bundles 
of arrows in the houses shews, on the one hand, that the missionarieB have hardly touched 
them; and, on the other, that the government has had no cause to interfere with their afbirs. 
We may therefore dismiss altogether the possibility of Folyneeian or European influence in the 
construction of their dwellings. 

The house (fig. 144) in which we stayed was that of the Soko-korobe (Nipa^crab; Vol. v. 
pp. 156 — 7, 190), clan house at lasa, and had the following external dimensions: 

Length over all (including veraudabs), 285 ft Breadth, 32 ft. Height, 26 ft., tapering 
to about 20 ft. at the ends. It ran about east and west, the main entrance being at the 
west, and faced the sea, the "bush" being close to it behind. A high tide would probably 
reach nearly to the piles on which it was buUt, and the numerous mangrove bushes growing 
in thick mud all around attested the unwholesome character of the situation. 

There were no lees than nine doors, of which one, barara, was at each end ; of the others, 
moto edia, two were on the north side (not apparently in use), and four towards the beach. 
To all of these access was had by means of rough staircases, loto, some being placed parallel to 
the main building, but most leading straight to the doors. The stairway at the western entrance 
was formed as follows. There was practically no verandah unless we reckon the inner compartment, 
mole, for the men as constituting one; from each side of the doorway at a distance apart of 

d by Google 


some seven feet a large ten or eleven-foot log stretched to the ground, and as the pilee were 
only some six or seven feet high at this point, the slope of the staircase was gentle. Crossing 
these two were numerous smaller logs or rungs bound together in four bundles so as to make 
a fiurly easy step, up and down which dingoes ran without di£Sculty. Files, tapering as the 
steps descended, supported the whole structure, and the visitor found a rude handrail on his 
right carried in the fork of six-foot poles. The top rung did duty for a doorstep, the doorposts 
and lintel being made of split sago palm. The other staircases were similar in construction 
and all somewhat slippery for booted feet Those at the side gave access directly to the central 
gangway of the house. 



^^y=^= r' ' ' T^^T^' 

Fio. Hi. Elsvation and plan of the Soko-konbe olaa house at lasa, bj A. W. Soale U ft. to 1 in. 

The piles on which the house rested were from six to eight feet high according to the 
ground. They were planted in four rows, though somewhat irregularly, and varied in size from 
4 in. in diameter to nearly a foot. The top of each bad been worked into a concave form so 
as to give support to longitudinal beams (fig. 145). Transverse joists crossed these at intervals 
of about two feet, more longitudinal poles, te ditlnt, were tied to these joists and upon them 
was laid transversely split nipa palm ', the convex surface being uppermost. The floor, te-ere or 
te Are, was thus exceedingly compact and somewhat resilient. 

The roof came down to within some two feet of the top of the piles, leaving only a little 
of the wall, opokara, visible. It was a lateral post roof, being carried by two rows of piles, 
earo, set in the ground at a distance of about nine feet from the sides at ten-foot intervals 
from one another. The top of these piles — they were perhaps on the average 20 ft. high from 
the ground — was generally forked so as to cany the purlins or longitudinal rafters, moo. Upon 
these purlins rested the rafters, kareuota or kararuso, of the slightly arched roof. The rafters 
met at the top and were fastened to a ridge-pole, gimini, which was supported by them and 
served in it« turn to keep them from springing apart. Another and smaller ridge-pole ran 
along the top outside, while an additional purlin ran on each side of and close to the inner 
ridge-pole, serving just to keep the rafters in their proper relative positions. Upon this frame- 
work was laid directly the thatch, were, of sago palm leaves, the mid-ribs, pat, of which were 
fastened to the rafters at intervals of about a foot. The principle of this form of thatch 
is precisely similar to that of a tiled roof. A few odd sago leaves were laid on the top of 
all — probably to cover leaks, while the slight gap between the courses of thatch at the ridge- 
pole was made good in the same way. The various layers of thatch were sewn together with 
needle-like pieces of coco-nut leaf ribs, and on the whole it seemed a fairly efficient covering — 
though by no means to be compared, except in principle, with' the beautifully neat thatch we had 
seen further east. 

The wall, opokara, of which only a couple of feet are visible under the eaves of the roof, 
rises perpendicularly on either side for about three feet from the floor. It is then deflected 
' Sir William Haogregor mi;s the floor ia made of the tt patm, this mar ^ another name for the nipa 
pslm, $oko. 



at aa oblique angle and oontinued into the carve of the roof for aboat another 2 ft. 9 inches. 
It wafl made of saf^ mid-ribs tied together and fastened to the superincnmbent roof above 
and to a number of poets at the sides of somewhat 
Blenderer proportions than the piles. So far as we oould 
see this side wall and its attachments took the place of 
the r^pilar raftere which appeared higher up. 

On entering by the west door, barara, which was 
6 ft. 6 in. X 2 ft. 6 in. and cut in a aago leaf end wall, 
a vestibule, mote, was entered which appeared to be 
Qothing but a verandah converted into a room. It 
extended the whole breadth of the house, but was cut 
off from the interior by anothei- partition, tamokaaio, 
of sago leaves laid together like the thatch, only in 
vertical instead of horizontaJ rows. The depth of this 
ver&ndsb was about 17 feet, that corresponding to it at 
the other end a foot less. 

To the left was a platform of sago 4 ft. 3 in. high 
x7ft longx 2 ft 6 in. broad, at a distance of 8 ft. Sin. 
from the west (out«r) wall. This platform covered 
a fireplace which was oblong and of the same length 
and breadth. These fireplaces (of which there were 
three in each vestibule) were formed of pieces of wood 
covered with matting and so arranged as to be easily 
dropped through the floor to the ground. A corre- 
sponding fireplace was on the right and a third occupied 
the middle between the two doors. Its dimensions 
were 4 ft. 6 in. x 2 ft., and its distance from the inner 
door 5 ft. 6 in. An open framework 5 ft. 8 in. high and 17 ft. long stood over the right-hand 
fireplace. The arrangement of the eastern vestibule wsa almost identical, save that one of 
the small platforms was wanting over one of the fireplaces. These oblong fiieplacea were repeated 
throughout the house to the total number of 41, their shape being particularly characteristic 
(cf. Old Mawata). Doors, tpMu-rumo, were fitted to the two outside east and west doorways, 
and like the side walls, were mode of sago ribs, held together in this instance by vertical 
skewers of wood in exactly the same way as are many doors in the Central District of British 
New Guinea. It may be here noted that the fall of the roof at each end began at the level 
of the interior partition and amounted to some six feet. The roof projected about five feet 
more beyond the end exterior walls and was curved back so as to meet the floor. Many 
instances of thb recurving of the roof at the gable ends were met with in the Mekeo District 
of the Fo flpocni on, where also occurred very similar cases of verandahs turned into rooms, for 
there is little donbt that this is the history of the Kiwal vestibule. 

On entering the inner door we found ourselves in what was apparently a tunnel 240 feet 
long. What light there was filtered in through the side doors and the holes in the thatoh, 
conversely the smoke of a score of fires curled its way to the outside air, after having performed 
the important function of expelling the mosquitoes. 

A central aisle, moto-gabna or moto-gabo, 15 ft broad was flanked by rows of platforms 
and partitions only broken by occasional doorways. It was immediately apparent that the whole 
wedght of the roof was carried by the raft«ra on the lateral posts, and tiiese latter determined 

L Tol. IV. 


d by Google 


the breadth of the aisle, supporting beflidee much of the fnunework that divided the family 
stalls, moUjtato. It is perhaps hardly correct to apeak of these stalls being divided one from 
aoother. They were, rather, marked off hy the fireplaces and the framework over the fireplaoes. 
In some cases the presence of a side door sufficiently separated oae family from another. Two 
or three of the five doorways, moto edia, on the south side (which was exposed to the 
B.K. monsoon) were walled off with sago partitions — probably to keep the draught under 
oontrol. In any case the three entrances to doors on the north side were not so protected, 
and the doors themselves, being unprovided with external staircttsea, answered the purpose of 
windows and ventilators. The breadth of the gangways and doors varied but was in some 
instancee about six feet, the height not exceeding three. The spacing of the doors was very 
irr^ular, the groups of stalls dividing them on the south side being respectively (from west 
to east) 26, 46, 24, 65, 35 and 25 ft. long. On the north (again from west to east) the 
intervals were 64, 116 and 52 ft. 

The arrangement of the fireplaces in the stalls is most easily explained hy reference to 
the plan (fig. 144). 

The best instance of a complete oompartment or section occurred on the left as the exterior 
east door was passed. It contained two fireplaces or, as we were informed, places for two 
men — that is for two families, and it was isolated from the rest of the building by the sago 
partition, opokara (the vertical and horizontal sticks of the partitions are called tamokau^la), 
and the side door with its gangway. The first of the two fireplaces was separated from the 
sago partition by a clear space of three feet. It was placed longitudinally and measured about 
6 ft. 3 in. X 3 ft being flush along its side with the central aisle. At the two nearest comers 
were posts which supported one end of s platform, the other end of which was borne by piles 
projecting through the floor some six feet back from the aisle, The platform was triple, consisting 
of three shelves two feet apart. All manner of food and utensils were piled on these shelves, 
which were adapted to hold the family property, and had different names according to the 
nature of the stores that they contained (thus dow, if for wood, but periperi if for food). We 
thus have a series of shelves covering a firepla<» over their whole length. The smoke would 
do much to preserve whatever was placed upon them of a perishaUe character. A clear space 
of nine feet intervened between this fire and the next which was placed transversely. This 
alternation of longitudinal and transverse fireplaces was carried out fairly consistently and, if 
it had no other object, at least served to economise space and to give a little more privacy 
than mere darkness and smoke afforded. The dimensions of this second fireplace and platform 
were about 7 ft. x 3 ft. 6 in., but, on the whole, their proportions appeared to differ within 
narrow limits throughout the building. The same series of triple shelves occurred as in the 
first. At the back of the compartment was another rough platform about six feet or less 
from the floor and extending the whole length of the nine foot space. It seemed to be intended 
to carry food. These supplementary platforms were not, however, to be seen in all the oom- 
partmenta. As to the six foot platforms they were practically continuous throughout the bouse 
save where the gangways to the side doors interrupted them and their frames doubtless did 
something towards stiffening and upholding the lateral posts. 

Allowing five members to a family this house should be able to accommodate about 
180 persons of all sexes and ages, or about 36 families. The provision of six fireplaces in 
the vestibules would allow one male member of every family to sleep there in tolerable comfort. 
It is curious that the proportion of men's fireplaoes to fomily fireplaces should be so near 
that of the numbers of fathers to the numbers of families— one to six and one to five respectively. 
It would be interesting to know whether these proportions are observed elsewhere. 

d by Google 

HO08E8 115 

The lateral post rool we saw in no other district either in New Qninea or Torres Straite, 
and it seemed much better adapted to the long hooaeB of Eiwai and the Fly Delta than the 
ordinaiy median poet method of construction. In the short bouses of the more eastern and 
central portions of the Foesession, the end walls serve to keep the roof from excessive sagging; 
but here there is nothing — not even a tie-beam — for 240 feet, save the lateral posts and the 
weak sago-palm screen which we have been obliged to speak of as a wall. In theory the absence 
trf tie-beams in the roof — and a roof, as will be seen, of rather low pitch — is a grave fault, 
and would doubtless lead to disaster were English building materials employed ; but in practioe^ 
in New Guinea, either the fioor ties the ribs together, or the weight of the superincumbent 
thatch is sufficient of itself to force them, elastic as they are, into touch with the sides and 
piles. The floor itself acts as a tie, and the fact that the roof is supported by the mwn purlins, 
which in their turn are supported by the piles, taro, would to a very large d^ree prevent any thrust 
upon the walls. The stakes, which support the inner sides of the platform, are too slight to render 
assistance to the roof, and, though l^ey might be carried into it and fastened to the ribs, 
we did not observe a single case in which this was done. Another grave source of weakness 
would seem to be the want of di^onal struts and stays between the piles on which the house 
is built; and we think there is little doubt that the few cases in which we observed bouses 
leaning dangerously out of the perpendicular might have been prevented by their use. But, 
in general, the piles are so firmly fired in the ground, so disproportionately stout, and so 
numerous, in regard to the comparatively light character of the superstructure, that bad results 
ore far from common. One of the lasa houses had settled a great deal to one side but it 
was obviously very old and neglected. We had no reliable information as to the time which these 
long-bouses lasted : probably twelve years would be a long life for one of them, as the smaller 
houses of Torres Straits and Central New Ouinea, which are proportionately stronger, are nob 
expected to be habitable for more than ten years. 

We noticed no "shrine" like those mentioned by Sir William Macgregor at Odagositia 
and Parama, and it was by no means certain that any particular individual was recognised as 
chief except by the government. At all events lasa did not possess a citizen so important 
as to require a special oompartment walled off from the common house for his use. 

The " notched stick " or log which gave access to the long-house of Odagoeitia was here 
represented by a regular staircase. Such was also the case at Old Mawata in the solitary 
instance of a house on piles which the place contained. Bull-roarera were hung on the verandahs, 
bat no carved poets were noticed. Though the lasa long-house itself was almost immaculate, 
its environs were not particularly tempting or tidy, and this may perhaps be attributable to 
the comparative proximity of white men, for in Central New Guinea it was an invariable 
rale that the further removed was a village from white influence the cleaner and better kept it was. 

Some Daudal Hootes. 

Crossing the main estuaiy of the Fly River and proceeding inside the island of Farama 
along the coast of Daudai we found a new type of house at Old Mawata, connected by a 
single piece of architecture with lasa on the one hand, and with the Masingara, if not with 
the western Torres Straits, on the other. 

Old Mawata is so called because it was the site of the village from which Gamia and 
his fellow countrymen went forth to found the new Mawata further to the south and west. 
The present inhabitants had come across the narrow channel from Farama and were "making 


d by Google 



their gardens." They had erected a nomber of temporary shelters, the most pretentioas of 
which was an ahnoet perfect reproduction on a very small scale of the long-house of lasa. 

This house (fig. 146 a — c ; it is also seen in the background of pi. XIX. fig. 2) was 
built on pUes of 6 ft. to 6 ft. according to the ground, and had its back towards the sea 
and the south-east wind. There was only one door and that faced about nortb-weet. Its 
external dimensions were : 16 ft long x 13 ft. 6 in. broad x 16 ft. high at the centre of the 
roof, which fell, like a hog's back, about one foot in front and two behind. It contained 
one oblong Dreplace of the Kiwai type halfway down on the left side of the entrance, and 
another (moveable) fireplace of matting. As at lasa the roof was carried on lateral poets. 
Hie side walls, 18 inches high, were formed of bundles of grass laid together like the ordinary 
thatch and not put on tile-fashion as is usual in New Quinea. The ends of the floor-beams 
were carried forward about a foot in front, and, being covered with cross-strips of bark, 
constituted a narrow verandah under the shelter of the slightly projecting eaves of the gable. 
A staircase about nine feet long led to the verandah. It had four composite steps of the Kiwai 
pattern and was supported by forked posts graded to the required height. The door, in the 
middle line, measured only 4 ft. high x 1 f t. 3 in. broad. The roof and end wall at the front 
were thatched, like the sides, and crossed with numerous bands of light wood which were 
necessary to keep much of the dry grass from being blown away by the wind. These bands 
were very commonly used in Torres Straits (on the introduced South Sea houses) for a similar 

Fro. IM. Elevatioas and gronnd plaa of hoaaes a 
D— F, a gronnd hut, by A. \ 

The other houses were little more than shelters of sticks, grass, and sago leaf (pL XIX. 
figs. 1, 2). They were exceedingly untidy and had obviously been thrown together only to meet 
the exigencies of the moment. Their occupants slept on the gronnd which was generously 
oovered, inside, with a coating of dry grass. These shelters, though sufficiently humble as 
examples of architecture, were at least interesting as being almost exactly like the descriptions 
contained in the Oovemment Reports of the habitations of bnah tribed living further west, and 
practically identical with the old houses at Mabuiag if the natives were to be believed. 
All these houses at Old Mawata which possessed doors (some were open at both ends or had 
only one complete side) had at least two — one at each end, and some had as many as three 
or four, all placed on the leeward side. Platforms built on piles about six feet high were 
used for storing food and nets out of the way of rats and other marauders. In some cases 
the piles had the top beams roughly mortised into them. A few of the houses had side walls 
and roofs quite distinct. More were simply roofs set upon the ground and slightly arched both 
in the longitudinal and transvene sections. In nearly all cases, the fires, even when there were 
no distinct fireplaces, occupied an oblong space. 

d by Google 


One hoiue (fig. 146 t> — r) meaanred 33 ft. x 15 ft. x 6 — 7 fL high. The roof did not fail 
at the ends. Five medi&n poets supported a ridg^-pole (fig. 147), two of which posts were in 
the end waUa. There were three oblong fireplaces of the Kiwai pattern 
but DO platforms over Uiem. The floor was composed of dry gross. 
Hie thatch and the end walls were of grass. The two door-ateps were 
a piece of split nipa palm and a log respectively. The doors at each end 
were necessarily to one side of the middle line and both on the same side 
of it They measured about 4 fL 6 in. x 2 ft. There was only one 

Another shelter on the ground had end walU of sago palm mid-rib. ^o- ^"- 8*™**"! ^"^ 
Another house was 1 8 ft. x 1 2 ft. x 5 a high. Its front and side were ^ M^SThT a"'w 
open, but dosed towards the north-west so as to escape the wind. The g^j, 8 ft. to 1 In. 
closed end was arched or rather curved and the roof on the open side sloped 

to within 3 ft. 6 in. of the ground. The ridge was carried by three median poets. The floor 
was of grass and the arched end projected about a foot beyond the end of a line drawn 
perpendicularly through the end of the ridge-pole. 

A lean-to was 3 ft 6 in. broad x 4 f t. 6 in. high and 7 ft. long. Its ridge-pole rested on 
two posts. Its sago leaf covering resembled the Eiwai thatch and, of course, faced the wind. 
There waa a slight grass protection at the windward end. 

The framework of these houses was of the simplest (fig. 147). The ridge-pole and median 
posts have been already mentioned. The usual rafters were then put on, and over them laths 
to carry the thatch. Confining bands for the thatch have been already mentioned, but they 
were hardly ever used for the shelters on the ground. Occasionally a slight purlin was added 
inside the ribs to keep them in position, but it may be said that, simple as was this coDstruotion, 
it was hardly simpler than that of the long-house, for the grass thatch required far more 
support for the same area than did the neat layers of sago leaves. 

There are a few notices in the Qovemment Reports of the houses west of the Fly Delta, of 
which the following is a summary. It will be observed, from the few details we have, how 
closely these houses correspond to those just described at Old Mawata. Of New Mawata we cannot 
■ay whether there was a special house for the men or not There was, at any rate, a long- 
house of the Eiwai pattern. Among the Masingara, a bush tribe in the hinterland of Uawata, 
and older inhabitants of the country than the coast people, it was found that "their habitations 
were a sort of compromise between the great houses of the Fly estuary and the small family 
houses of the east of the Possession." There were houses for the men 60 — 60 feet long, 
"without walls or posts." They contained trophies of the chase. There were also many large 
bmily houses for the women and children, the sides and ends of which were closed in. The 
Badn bush tribe had six houses from 30 — 60 feet long, of the same type as the Masingara. 
At Dabolai was a temporary settlement of a tribe who were seeking to escape the ravages of 
the Tug^ pirates. There were four huts 16 ft. x 10 ft., built of bark and boughs and with 
the ends left open. The " bachelors' hut " was some fifty yards distant from the other three. The 
total population was estimated at 300 individuals. On the Mai Eusa river were found the 
sheltera of sapling frames and paper bark, erected by the Tageri pirates tJiemselves. On 
tiie Morehead river at the extreme west of the Fossession there were a few bad and rough 
■belters ctf paper bark. 

The bachelors' house of the Masingara, "without walla or posts" (piles T), bears an analogy 
to the kwod of Mabuiag, and the use of bark, the absence of piles, and the separ&te men's 
house are wor^iy of note. 

d by Google 

118 anthropological expedition to toebes straits 

Notes on Houses in Torres Straits and Western British New Guinea. 

As with the land tenure so too with the houses. Between Kiwai and Old 
Maw&ta on the one hand and Mer and Mabuiag on the other there are certain great 
differences of construction and nomenclatare. It is nob too much to say that there ia 
hardly a feature or a name of a feature common to the old round house of Mer and 
the long-house of Kiwai. There is actually more resemblance to the South Sea type 
of buildings of Mabuiag in the old style habitations than to the long-house. . On the 
other hand. Old Mawata supplies a connecting link in this case which is quite wanting 
in that of Mer, unless it be that a hint is to be found as to the origin of the round 
bouse in the solitary tumble-down hut described on p. 104. Not only is it difficulb 
to trace resemblances of form between the mainland and the islands, but the eastern 
and western islands themselves differ no less one fix>m another than Kiwai does, say, 
from Port Moresby. 

The data upon which satisfactory comparisons could be based have quite disappeared 
at Mabuiag and have almost vanished at Mer and Erub. In the former case we have 
had to rely entirely upon the philological evidence and the memories of men who were 
mostly mere children when the changes that destroyed these data were already taking 
place. And the worst of it is that some of the people consulted had seen the very 
houses with which we were anxious to compare their own, and were proportionately prone 
to draw upon their imaginations when memory failed them. 

Under these circumstances it is possible to do little more than describe houses, pointing 
oat, in a few instances, any marked similarities between the buildings of different localities. 

The long-houses of western British New Guinea Have attracted the notice of all 
the travellers who have seen them, from the memorable voyage of the "Fly" onwards. 
We bad, however, no clear description of one until Sir William Mac^regor published 
his Report of the Western Division (Ann. Report on Bri^h New Ouinsa frmn. l«f Jvly, 
1889, to 30fA June, 1890, Brisbane: C.A. 106—1890). In this Report there is a very 
serviceable account of a long-house at Kiwai-Iasa ' and of another at Odagositia, besides 
numerous notes on the dimensions and appearance of the houses throughout the district, 
in addition to other interesting information. 

The low and swampy island of Kiwai has a fairly evenly distributed population of probably 
not Ear short of 6000 souls, or, say, 50 to 56 to the square mile; the island is about 
36 miles long, and about 2} miles broad, thus coataiDlng about 90 square miles. Sir William 
Macgregor has given a valuable account of the island, and he visited all the villages; the 
foUowing very brief notes of some of them must suffice. Saguane : two bouses about 200 ft. long, 
two smaller, 30 ft. long, population about 260. (There were no long-houses remaining at the 
time of our visit in Sept. 1898.) Samari: chiefs house, 50ft. long, of which he occupied 
ten feet at one end separated off by a partition, some ten houses, 30 — 60 ft. long, one over 
250 ft. lot^; by 27 ft. broad, population 400. Mdbudamu: two houses about 120ft. long. lasaz 
six houses, 120 — 150 ft. long, besides six or eight smaller ones, population about 500 (see p. HI, 
and Vol. v. p. 190). Kvbira : three houses, 150 — 200 ft long, population about 300. Svmai: about 

> AooordiDg to Sir WlllUm the naUvei tsj thst one-half of th« viUsge is oalled Elwtu sod th« other lais, 
bnt of. Vol. ?. p. 190. To prevent ambigoity we adopt the ntme Isu for the whole villnge. 

d by Google 


half-a-dozen large houses and a few gmall ones, population may be over 500. Atili: five houses, 
150ft., chiefs quarters aa apartment of fifteen feet cut off from main hall by a doored partition; 
fifty grimy skulls on each side of this door, population about 500. Ipitia : two villages, half-a-mile 
apart, southern village contained four or five houses 200 ft. long, the northern, five or six bouses, 
none exceeding 100 ft. in length, population about 1000. 

At Odagotilia, about 51 miles from the mouth of the Fly, all the population of some 
hundreds live in one house 520 ft. x 30 ft. and two or three small ones. At Tag<^ there were 
six or ei^t houses about 100 ft. long. 

It appears that the tong-bouse proper ia not found above Everill Junction, where the 
Strickland River joins the Fly. 

From Sir William Macgregor's account we are justified in saying that the houses on the Bamu 
River form a connecting link between the baildings of the Fly Delta and those of the Papnan Oull 

While the typical long-houae of the Fly Delta is inhabited by men, women and children, 
the long-houses of the Oulf and the Mekeo District are reserved for the men. The first stage 
in the separation of the men and women and children has already been reached in houses, 
which, as at lasa, provide the men with separate quarters in what corresponds to a verandah 
with a covered end wall. At the same time, the movement towards the subetitution of £amily 
for oommunal bouses has made progress — ^perhaps because of missionary influence — both in Parama 
and KiwaL At Daru, the seat of government for tlie district, no long-houses are to be seen. 
At Old Mawata some of the houses were obviously too small to accommodate more than 
one famUy, and the people who built them came from Parama. Mawata itself has or had 
long-honsee, but the Mawata people were strangers and invaders from the Fly River Delta. 

In short the communal long-house, so far as we know, is only found among the Delta 
people and their neighbours. Outside the Delta it loses its oommunal character and becomes a 
clubhouse for the men. Inside the Delta iteelf, it is now rapidly being ousted by family houses. 


Sometimes, especially in Mer, gardens are protected by fences, pa (W.), kar (E.). 
These are generally made of bamboos stuck closely together in the ground, with cross- 
bars to strengthen them (pi. XXI. fig. 3). Fences of this kind were frequently put 
round vilUges, sometimes perhaps as a protection against enemies, but usually, as in 
the case of the village of Las on Mer (vi. pis. XXIIL, XXV.), the fence serves as 
a protection against the constant strong south-east wind. In the story of Kwoiam it 
is stated (v. p. 72) that a village on Boigu was surroonded with a fence of coco-nut 
palm leaves, and on p. 73 that a New Guinea village also had a fence round it. 

Fences are now put round graves (v. pi. XV. fig. 3), and fences were sometimes 
put round sacred places (v. pi. XXII.). In the story of Bomai we find that fences 
were frequently but inefiBcaciously put round that mysterious polymorphous personage 
(VI. 34—^8), in the Murray islands, a kegar kar, which was said to be a stone fence, 
was erected ; the only stone fences now made are the fish weirs, «it (pi XXIL fig. 1), 
the Malu word, beiear (iii. p. 51, vi. p. 298), is apparently the equivaleat of the sax. 
On Danar Bomai was enclosed in a rope fence, beriberi kar. 

The ceremonial screens, waus, of the Western Islanders are described in Vol. v. 
pp. 366. 367 and pi. XIX. fig. 2. A fence of matting is called motoat (W.). 

The trumpet shell is a very common decoration for fences, both in the villages and 
in the gardens of Her (n. 8). No satisfectory reason could be discovered for this custom. 

d by Google 


Owing to the elemeatary nature of the islanders' housekeepiog their domoErtiio utensils 
were few in number and of simple constructioD. Certain implements, such as scrapers used 
in mat-making, thatching needles, etc, are described when dealing with the particular 
industry in which they are employed. 

There were no moveable articles of furniture, for the beds of the Eastern Islanders were 
permanent erections, the only exception being simple plaited mats (p. 65); these were of 
var3ring size, and were used as bedding or floor mate and the larger sizes might be used as 
canoe saila Mats were not habitually laid upon the earthen floor, but were used to sit upon 
occasionally, especially when feasts were given, in which case the feeding took place outside 
the houses. Small mats of finer construction and softer to the touch than the ordinaiy 
mats were employed as sleeping-mats for babies, or for wrapping up things (p. 65). 

At the present time, at all events, the women keep the houses clean and the space 
round the house is kept clear of dirt and refuse, for which purposes simple brooms, piwul 
or piwal (W.), bei lid, weeker (E.), are employed. They are made of a bundle of the mid-ribs 
of the leaflets of the coco-nut palm, the proximal ends of which are tied tightly together. 
A broom is about 67 cm. (26^ in.) in length. The htsakusa (W.), probably kauaakaiua, 
broom is apparently made of pandanus (kausa) leaves. 

Torehes, tu (W.), bei or ne (E.), are made of the dry leaflets of the coco-nut palm, the 
object and the material of which it is made having the same name, as is frequently the 
case. Torehes may be employed on the rare occasions when the natives walk by night, 
or when they spear fish by night. 

Tongs, komazer (E.), for putting things on or removing them from a fire, were made 
'by bending a strip of bamboo ; the rind may be retained or scraped off except at the bend 

Fia. 118. Bunboo tonga, Mer. 

(fig. 148). The length fiY>m the bend to the tips is 23 cm. in one old specimen, others are 
24 — 30 cm. long. 

The ashes of the fire may be prevented from scattering by a finmework of four low 

d by Google 


boards, a moveable fireplace of this kind with a wooden bottom, ur memeg (E.), is taken 
on board a canoe when going on a long journey. 

Stones, baa (W.), irmad (E.), usually three in number, were placed on the hearth to 
support the shell cooking-vessel (pi. XXXVII. fig. 3, where four stones are represented). 


Fire, mui (W.), ur (E.), was made by twirling with the hands a vertical stick upon 
one placed horizontally, the latter may be set on a stone to steady it and is held by 
another person at its ends. A notch is cut for the insertion of the upper stick, and 
a lateral groove is sometimes cut (fig. 149), but this does not appear to have been 


Fio. 149. Fire-Etick, Her; boriiontal stick Sl-Som., vertiebl 8ti«k 49011). The horiioDtal sUok has two burnt 
holea uid one out readf for dbs with & nick at euh side. In one old hole both mdea are »o burnt through 
tbftt one cannot tell whether there has been a aide uiek, and the other is bomt throng od one side, bnt 
unmiitakably baa a niok on the other. The tinder ased in this cau wae a pieoe of oooo-nnt huak. 

always made. The upper stick is held between the outstretched palms and fingers and 
is made to revolve rapidly by a backward and forward movement of the hands, which 
at the same time are brought down the stick with a strong downward pressure. On 
reaching the bottom they are brought rapidly to the top, great care being taken not 
to allow the stick to rise ftcm the notch. These movements continue until a quantity 
of wood dust, goigoi le, or pt (E.), accumulates in the hole, a good deal of smoke 
accompanying the process. The dust begins to glow from the friction, then some tinder, 
pea ur (R), is applied, and by gently blowing this a flame bursts forth. The tinder used 
is often the dried spadix of the coco-nut palm. The whole operation seemed to require 
great manual .dexterity, quickness of movement, and judgment. I have seen three natives 
engaged in the task, one holding the lower stick, while one of the others relieved his 
fellow when tired with the exertion of manipulating the upper stick, the change being 
effected with such deftness that the stick never ceased revolving. So &r aa I am 
aware the natives very rarely have recourse to this method of making fire, as matches 
are now so common, and consequently some of those who made fire for me had very 
little dexterity. I believe that fire could be produced in less than a minute by this 

Fire-sticks are called goigoi, and the process is goigoi aalgai (W.), goigoi drimli (E.). 
The Miriam call the vertical stick, werem, "child," and the horizontal stick, apu, "mother"; it 
is said of the latter, apu ur ikwar, "mother gives fire" (R). Sagai is the Western name 
for the horizontal stick, and aalgai is also the collective name for the two sticks. 

Wood for making fire is obtained from the following in Mer (c£ vi. 30): arq^rep, 
H. VoL IV. Ig 

d by Google 


argerarger (Callicarpa), ^ebt, ibi, kokokoko, kozb, marep (bamboo), sem (Hibiscus tiliaceus), 
si>be (Eugenia), urlagerlager^, ursekeraeker, warupwofup, viadewade, zih. 

The Tariants of the folk-tale narrating the origin of fire are given in Vole. v. p. 17, 
and vi. p. 29. 

Owing to the difficulty of making fire the natives were careful not to let fires 
go out, indeed in Mer there was a special fire-charm, bager, usually a small stone 
image of female form, which was designed for this purpose (vi. 202). When going out 
in a canoe, or into the hush, they habitually carry a glowing stick, which can be blown 
into a flame when required to ignite a fire, and they are carefiil that the brand is 
not accidentally extinguished. 

A sheath, iaka (W.), for protecting the ends of the sticks and keeping them dry 
was doubtless a necessary adjunct, as at Cape York and elsewhere in Australia. 


Pottery was entirely unknown, as was also the art of carving wooden vessels, recourse 
had therefore to be had to natural receptacles, such as shells, coco-nuts, gourds and 
lengths of bamboo. 

Although all the articles enumerated below are still used, some are employed but 
rarely at the present time owing to the introduction of European goods. 

Fio. ISO. Oiant Fubqb need as • water Tesul. This Fio. 151. Oassis oornuta used iu a wal«r TSBsel. 

Epecimen is 43 cm. long. This apeoinieii a 34 cm. long. 

The following Urge shells are employed as water vessels: the giant Fusus, bu (W.), 
maber (E.) (fig. 150; pi. XXI. fig. 3); Triton variegatus, which I believe received the 
same name as the foregoing; the helmet shell, Cassis comuta, as (E.), fig. 151; the 
bailer shell, Melo diadema, alti^ or galop (W.), ezer (E.) ; and a valve of the f{iant 
clam, Tridacna gigas, moAwa (W.), mi (E.). Shells used as drinking vessels are called 
SOT in Mer. 

The bailer shell is everywhere used as a cooking vessel (fig. 152), but the giant 
Fusus was also sometimes used (T. 41). These shells are used as receptacles for the 
blood, ovarian eggs and viscera of turtles, and for red ochre, etc. 

^ Ddelagerlager and laUaga-tager in TocabuUry, Tol. ni. 

d by Google 


Long pieces of the stem of the bamboo, in which all but the lowermost node are 
pierced, are used for storing water, and shorter sections are employed as water vessels, 
huttt morap (W.), mar^ (E.). In pi. XXIV. 
fig. 1, a man is seen pouring water from one 
bamboo tube into another (c£ Vol. vi. p. 3). 
A narrow tabular receptacle for holding small 
objects is called burar (E.), it is made from a 
small bamboo (fig. 153). 

The univeraal water vessels are coco-nut 

shells, kttsu, kusb (W.), m sor (E.), m, fresh 

water, sor, shell of a mollusc or nut; a small 

form is called idi sor, perhaps because oil was 

, ■ .. mi 1 1 !■ iL ^ ■ J Fio. 152. Shdl MQcepan. This (peauaen u 

kept iQ it. The husk of the nut is removed a9om Iodb 

and the shell scraped and polished ; afber having 

been in use for some time it acquires a beautiful dark brown or black polish. One 

"eye," gud (W.), is perforated, the other two are frequently painted red. These vessels 

are invariably carried in pairs, generally in one pair, but sometimes in two. The handle 

consista of simple or complex plaitwork (fig. 122), the ends being 

thinner than the centre. Each end is tied into a knot, this is inserted 

into the perforated "eye," and a stopper made of roiled-up leaves 

makes all secure and enables the coco- nuts to be carried by the 

handle (fig. 25; pis. III. fig. 1; XXL fig. 3). 

A. coco-nut shell cut in half is also used as a drinking vessel. 

Gourds, kabor (E.), were occasionally used as vessels, more especially 
as water vessels, ni kabor. 

Macgillivray (ii. 20) mentions that the Muralug natives make a 
"water basket from the sheath of the leaf of the Seafortbia palm," 
Ptychosperma elegans, lulko, which is also the name of the vessel. 
I was informed that a vessel, ubu, shaped like a whale-boat was made 
in Muralug of the bark of an Acacia, tihtL 

Saucepxms and kettles of European make are now in common 
use, as are cnps and saucers, dishes and plates. Instead of the last 
the natives employed banana leaves, and they frequently do so still. 

Pounding, Sceapino and Cutting Implements. 
Very tough food may be pounded with stones. Fig. 154 repre- 
sents a flattened cylindrical boulder of red granite, both ends of which P^. 168. Burar. A 
are bruised with use. It was employed for pounding dugong skin o«ptaole. Her. 1g6 
and for other purposes, and was obtained at Tutu but must have ^™-. ''*°^'„^ '^t 
been brought from Yam. A pebble, 80 mm. in diameter, is figured u engi«Ted in dot* 
in the Album, I. pi. 323, No. 6. Similar stones were carried in the rfS^^S'oSh^ 
hand by Miriam men when they walked about and old men crushed boats. Both end* 
their food with them. They were sometimes used as missiles. I obtained 
this particular idid baker at Mer and gave it to the British Museum. 


d by Google 


At Mer I obtained an implement for softening food by percussion (fig. 165). It 
conaiets of a mallet, par, 295 x 47 x 48 mm., of which the head is square and the handle 
circular in section, and a block on which the food is pounded. This is roughly triangular 
in form, measuring 285 v 105 mm. in its greatest diameters and 61 mm. in thickness. 
They are made of a close-grained heavy wood. This specimen was said to be used 
by toothless persona Other objects were employed for this purpose, for example, when 
I bought the old broken-off butt end of an mutiter or maid wap (see Section on 
Decorative Art) in Mer in 1889, it was being used for beating roasted bananas. 

Fio. 164. Btone potmder trom Tata. Fia. 165. Mallet and block tor ponoding food. 

The single valves of various bivalved molluscs, such as akul (W.), Cyrena divaricata 
or Batissa corbiculoides, id (W.), Tellina staurella, kaip (E.), Asaphis deflorata and other 
shells, gider (E.), Tellina diapar, are used for various purposes, such as cutting, scraping, 
ladling, etc. A supply of shells strung together and 
ready tor immediate use is occasionally found in some 
bouses. For example, I found in Mer that large 
numbers of unseparated valves of Asaphis deflorata 
and Tellina diapar were strung together by two strings, 
which were simply tied above the ligament of the 
hinge of each pair of shells. Sometimes shells are F 
carried behind the ear so as to be ready for an 
emergency (v. p. 89). 

When used as spoons, the shells are spoken of in Mer under the general name 
of aro4u, eating thing. A blade-like piece of turtle-shell trom Mer (fig. 156) was described 
as lewer ero karar, yam eatii^ turtle-shell, but it is strange that the Western and not 
the Extern word kaisu was given. 

Forks were never employed. 

A low trestle seat to one end of which is fastened a strip of iron with a toothed 
end is DOW generally employed for scraping out the kernel of ripe coco-nuta. This is 
A common Polynesian implement and has been introduced into Torres Straits. The 
operator sits on the seat, holds the broken nut in his Itand, and scrapes out the kernel. 
Formerly the natives used a strong valve of a mollusc for this purpose, a piece of 
pearl-shell or a turtle-shell scraper (figs. 157, 158). 

The shells of various bivalved molluscs were habitually employed as knives, such 
as id (W.), Tellina {Album, I. pi. 323, No. 9), akul (W.), Cyrena, the thin sharp us (K) 

d by Google 


shell was used for carving patterns, etc. (p. 15). The akul and not the upi was employed 
in cutting the tough skin of the dugong. 

Fragmente of quartz, va or mhw (W.), were used for scarification of the skin (p. 15) 
and for other cutting purposes, such as renewing the edge of an upi (v, p. 71)^. The 
islanders were very desirous of obtaining from the early navigators empty glass bottles, 
which the Miriam called " tarpoor " (tapor, box) ; these they broke up and used the 
fragments as knives. 

Boars' tusks, gi (W.), jrtr (R), were used as knives and scrapers (fig. 71). 

Fia. 1K8. Tnrtle-iheU unper, 12oni. (4jiii.) 
loDg. Mer. B. M. 

Knives, upi (W.), kwoier or koer (E.), were made of split bamboo, and were mainly 
employed in cutting up the flesh of turtle and dugong. I am not aware that any 
original knives exist, as they have long been replaced by European iron knives. Some 
may have had the form of the beheading knives (fig. 201), but probably knives were 
made as needed by simply splitting a bamboo, like the specimen which we collected 
at Eiwal The siliceous particles in the rind of the bamboo cause a thin edge to be 
very efficacious, and a new cutting surface is readily obtained by splitting ofiF the 
dulled edge. 

The only axe, aga (W.), panigob or inigob^ (E.) [tidik E. (p. 129) is now the 
general name, au net, for axes], so iar as I could learn consisted of a shell or occasionally, 
it is said, of a atone blade inserted in a stout wooden handle. I have never seen an 
original specimen and could obtain but two or three shell blades, and these only on 
Mer. In 1889 and again in 1898 I induced acme elderly men to fit blades into handles 
as they were tbrmerly made. On both occasions the handles were made alike, so that 
we can confidently assume that fig. 159 represents the old native axe. 

The shell blades (fig. 160) vary in length fi-om 83 to 113 mm., in breadth from 
43 to 49 mm., and in thickness from 24 to 28 mm. They are made from the shell 
of the giant clam, miskor. The handles, pes (E.), are about 60 cm. (23^ in.) long, and 

' Kageg (ti. p. 18) 'a said to have oat open a maD named Iriam Hori» with a stone knife, I do not know 
what it waa like, it ma; have been a qaaitz flake. 

> Another Miriam name tor axe is demner {dauintr) ipikeub lulik, pigeon tail iron. 

d by Google 


are made of 20m (Theapeaia popalnea) or kid {^) wood. The handles are straight or 
slightly curved, with a marked atop below the grip; the latter is circular in section 
but the handle becomes flattened laterally towards the other end; the flattened tip 
is painted red, and a red ring surrounds the insertion of the blade. The axes weigh 
nearly two pounds. 

Fio. 160. BheU axe bUdes, Mer ; the Fio. 161. Stone axe blade bom 

RDullMt BDd largest speoimeiia ; Kiwai ; 69 x E3 x 38 mm. 

84x13x34; 113x4gx38miii. 

Fia. IGQ. Bfiiiam ihell um. 

I add in fig. 161 an illustration of a smalt stone axe head, tapi, from Kiwai. 
If stone was once used for axe beads in Torres Straits, it would probably have been 
worked into a similar shape. 

Adzes are unknown. 

Finishing Tools. 

Wooden objects such as dugong harpoons, clubs, coco-nut shells, or other articles 
which required polishing, were smoothed down by means of boars' tusks, ngaingai, 
buruma gi, or simply gi (W.), gir (R) (fig. 71, p. 52). Sometimes, to judge from 
the wearing down of the tusk, these appear to have been held in the left hand. The 
action of scraping is directed away from the body. A groove is sometimes cat round 
the tusk near the point, probably for attaching a string. 

Rasping was done by means of a piece of skin from the toil of a ray, taimar 

d by Google 


upi (fig. 162). I did Dot see a apecimen of taimer (W.), taimar (E.), that was mounted 

on a piece of wood as is the case in New Guinea, 

so I suspect that my specimen was really a sample 

of the skin and not an actual tool. There is, 

however, in the British Museum, a rasp made 

of a neatly shaped piece of wood, 32 cm. long, 

to which a piece of shagreen is lashed by a strip F™. 162. Siin of the tail of a ray ased as 

of ratan ; it was obtained from Moa by H.M.S. " ""P' "'''"'^- * ™*- '"^ 

"Rattlesnake" {Album, I. pi. 291, No. 11). The taimar was employed in the making 

of turtle-shell and other implements. 

A smoother surface was acquired by friction with pumice, met (W.), zor (K) ; pieces of 
pumice are extremely abundant on the sandy shores and on the low islands. In Mabuiag 
we obtained an erect, flat, lobate sponge, gouga, which was also employed for polishing 
dugong harpoons. Zarzar or sarza leaves were used for a similar purpose in Mabuiag. 

Vabious Imflbhehts. 

A bandy tool, sok (fig. 163), is imported fr^m New Guinea. The lower leg bone 
(tihio-tarsus) of a cassowary is split longitudinally, and the upper portion cat away and 
brought to a flat rounded point, the lower articular end being left intact. It is mainly 
used for husking coco-nuts. Two specimens measure respectively 230 mm. and 295 mm. 
Occasionally they are decorated with simple patterns. Another specimen ia 
370 mm. in length, and has plaited cane fastened round the handle end. 
Sharply pointed specimens are said to he used as daggers in New Guinea 
and possibly they may have been so employed in Torres Straits, but of 
this there is no evidence. 

When a stick for husking coco-nuts is made out of kaa wood in Mer 
it may receive the same name, but the usual name is pat; digging-sticks 
are also used for this purpose. 

The slender awl-like bones that support the flying membrane of the 
fruit-eating bat or flying-fox (Pteropus) were used as borers, and were 
then called sapvr kimua (W.) (p. 10). 

Needles, saga (W.), atket lu (E.), sew thing, appear to have been 
made of bone, but I have never seen a specimen ; possibly these were 
really awls, and when European needles were introduced they received 
the same name, and the eyes were then termed sagau gud, needle's 
mouth. The bodkin, arem lu, used in thatching is described on p. 101, 

The employment of sharply-pointed awls of hard wood for piercing 
the nose and ears of in&nts has already been described. 

Skewers are called kep, or kas keg in iier. Fio. 163. Bone 

Long flattened bodkins or awla, ter or luper (E.), made of turtle- oocM-nut haak- 
shell, were used for piercing the septum of the nose and the ears of . - - 

in&nts, and for shredding the leaves of which petticoats were made ; on pi. XI. are 
shewn photographs (figs. 11 — 17) of seven examples which illustrate the usual variations 
in form of these objects; the upper broad end is frequently ornamented with simple 

d by Google 



pattems engraved in plain lines or in the characteristic zigzt^, and is usually per- 
forated. They vary in length from 13 — 27 cm. (5 — lOJ in.). Two ter from Mer, 258 
and 200 mm. long, are attached in the following manner to a three-ply plaited necklace 
composed of twisted native string, 68 cm. long: the ends of the cord are tied together 
in a knot and the free ends frayed out as a fringe. A delicate twisted string is passed 
through this knot and through a hole in each ter (the knot lying between them), then 
through a half Coix seed in which the string is knotted to prevent its slipping. One 
of the ter has a simple decoration with red colour rubbed in. 

One luper from Mer is a narrow, slightly crescentic tiurtle-shell implement, 92 mm. 
long, used for splitting leaves for petticoats and mats, the lower edge of one end 
being the cutting portion. The same name was applied to simple straight, sharply- 
pointed pieces of turtle-shell, 86 — 115 mm, long, which were used for a simitar purpose. 

I have not come across any drilling implement though the so-called pump-dnll is 
common in New Guinea, but it ia obvious that there must have been an instrument 
for drilling holes in wood, pearl- and turtle-shelL I was informed in Mahuiag that holes 
were pierced in pearl-shells by means of sharks' teeth. The anterior long, slender, simple 
teeth of Crossorhinus, im (W.), were used as drills when inserted in pieces of wood. 

There ia a pump-drill in the Kelvingrove Museum, Glasgow (89 — 67 ea.), which was given 
by R. Bruce, who stated that it came from Mer, but he gave as its name "good pegoo" 
(gvd pigti), which ia undoubtedly a Western and not a Miriam term (gud, mouth, bote ; pigv 
is probably a form of pagai, to pierce ; gvd pamai, is to enlarge a bole). The rod is 65 cro. 
loDg and has a atone point, the cross-bar is 247 mm. long, the two strings auapeud it at 
a distance of 24 cm. from the attachment, the fly-wheel disk is 107 x 90 mm. in diameter 
and is made of a light coarse-grained wood. It is decorated on both sides with a large central 
star-like figure and a marginal dog-tooth pattern, the relief is charred black and the alternate 
triangles are white and red. The technique of burnt relief, the peculiar pink ochre of the 
reddened parts, and the style of the decoration leave no doubt in my mind that the disk at 
all events was made in the Papuan Oulf district. 

When the fruit of the Pandanus is over-ripe and the succulent 
part quite dried up, the drupes readily become separated and the 
fibrous inner portion of each forms an excellent coarse brush (fig. 164) 
which is used for painting or for other purposes. 

Walking sticks, hogi (W.), koket (E.), are frequently carried by 
the isUnders, One of ratan that I gave to the British Museum is 
159 cm. (5 ft. 2J in.) long. 

A peg is called puidaua, pi. puidaizapul (W.), hang thing, Ngail 
(W.) is given in the Vocabulary as wooden hooks, but it may be the 
English "nail." 

, Iron. 

The earliest record which we have that the islanders knew of 
the value of iron is that of Matthew Flinders who served under 
Captains BUgh and Portlock in the ships " Providence" and "Assistant" 
in 1792. These navigators discovered Erub and Napean and most of 

d by Google 


tbe western islands of Torres Straits (Vol. iii. p. 1). The Erub (Damley) ialanders 
vehemently "asked for toore-towee ! hj which they meant iron" in exchange for "arrows 
and other weapons."^ When Flinders revisited the Murray Islands on October 29, 1802, 
the natives were "holding ap cocoanute, joints of bamboo filled with water, plantains, 
bows and arrows, and vociferating tooree] tooreel and mammooseel"^ On Aug. 5, 1843, 
H .M.S. "Fly" reached Mer, and the islanders were " clamorous for tooree (iron) and 
knives. For the latter they used the word 'knipa,' evidently got from passing vessels" 
(Jnkes, L 133). In his vocabulary (IL 293), Jukes gives the following names for 
"European articles" (his spelling is preserved). Iron: toorree, toolick (Enib and Mer), 
tooUck (Mer — Lewis*), toori, toodi, tooli (Masig, etc). Eiiife : naipo (E. and M.), queer, 
toolick (M. — L.), naipo (Massig, etc). Hatchet: sapd/ra (E. and M.), dauma oopy kew 
(M. — L.), aapSra (Massig, etc). Macgillivray, in 1849, found that the "Kowrarega" of 
Muralug called iron twr^ea, and the Gudang of Cape York called it gire ; the former called 
a knife gi-tarik (=iron tusk), and both called an aze aga (which was probably their 
pronunciation of "axe"). TurUe is now the Western name for iron and tulik the 
Eastern ; turika is also employed by the Bugi, Mawata and Kiwai natives of New Guinea, 
tbe latter also have kerere, and the Tugeri wokiriki. 

From the foregoing it is evident that before 1792 iron had become known to the 
islanders through passing ships; but the derivation of the words turi and turik or tviik 
is unknown. 

The natives at first readUy named European metal objects by combining the name 
of the equivalent native object with tulik. To quote again from Jukes (ll. p. 294), 
a spoon was " caip toolick," iron shell, a carpenter's saw was " teerick teerick toolick " 
(fvreg tereg ttUik, lit. tooth tooth iron or many iron teeth), a spike nail was " soa/ 
toolick" {sok p. 127, tulik). 

Main is the Western name for the metals they know and the E^astem for iron, an 
iron plate or sheet iron. Mr Bay (ill. 168) suggests that it comes fiY)m the Lifii mdele, 
thin. A chain is known in Mer as malil lager, and in the west as malU uru, both 
meaning an iron rope. 

1 Ustthew PliDden, A Voyage to Terra AuttralU, Loodon, 1814, i. p. xzii; in b footnote it ii ttatad tliat 
"tlie nune for Iron ftt Tfthei^ ia eurt-mre, or ooree, or ujoording to BooeuuTille, aouri." 

* Op. HL a. p. 100. 

* A Toe«biila(7 eolleet«d t? Cftpt 0. II. Lewie in 1886 and pnbUihed by Jnkea; of. Haul. Mag. n. 18S7, 
p. OM. 

d by Google 


On the whole the dietary ia sufficiently varied. Yams and sweet-potatoes are fiiiriy 
abundant on many of the islands, the former constitute the chief fEuinaceouB food. 
Nutritious food is generally very scarce at the end of the dry season and the beginning 
of the wet, at this time the natives often have to rely on anything they can iind in 
the bush that is edible. Some fruit or other, including the banana and coco-nut, is 
always obtainable on the more fertile ishmds. Mocgillivray, speaking of the Prince of 
Wales Islanders says, "The food of these blacks varies with the season of the year, 
and the supply is irregular and often precarious. Shell-fish and fish are alone obtainable 
all the year round" (II. p. 20). On pp. 1 — 3 I have drawn attention te the economic 
status of the different groups of Islanders. The Central Islanders are in a worse position 
than the Muralug natives. The natives of Badu and Mabuiag are in a better position 
as regards food, and the Eastern Islanders are the best off; but even in Mer times 
of scarcity are not anknown. Fish or shell-fish are eaten nearly every day, with 
occasional meals of turtle and dugong; the two latter are especially "rich" or oily, 
I believe that fruit is the only article of food which is habitually eaten raw, and in 
some cases this is cooked. Fruit or vegetables are never preserved with sugar or by 
pickling, nor is meat salted. 

Neither salt, spices nor any condiment are habitually mixed with their food, though 
the derh that was mixed with iniu (p. 135) was probably of this nature. I was informed 
that honey was used to sweeten biiu, but have no knowledge that it was used to 
improve the flavour of other food although it was greatly appreciated when obtainable. 
The same holds good for sugar-cane. No whets to the appetite are in use. 

As to the quantity of food eaten I should say from what I saw that it is on 
the whole about equal to that eaten by an average Englishman. 

Formerly the bther and his sons ate their meab before the mother and girls 
had theirs, the same applies to groups of adults ; but Mr Bruce says that in the 
Murray Islands the husband, wife and family ate together, the husband however reserved 
to himself the right of choosing certain tit-bits, and this is still frequently the case. 
In some of the western islands several families might occupy one house, and even 
now this sometimes occurs; when this was so each bmily had its own fireplace, and 
each provided its own food, but if one man had no fish or other food while another 
in the same house had some, the latter was bound to give some to him who lacked. 
The meals are very simple and unceremonious. Banana leaves are frequently used as 

d by Google 


plates. Feaata were made on all poBsible occaaions, especially b; the Miriam. QyriDg 
to the abeeDoe of intozicatiDg liquors there were no driDking festivals. 

The chief meal of the day is taken at night, soon after sun-down, the remains 
are eaten in the morning. The natives continually eat between meals anything that 
may be handy. 

In Mer the following food was considered necessary for the ttmeral feasts — as a 
matter of fact feasting in this island was mainly connected with these ceremonies: 
pea u, green young coco-nuts, gebgeb u or au «, old nuts, toai u, germinating nuts, 
and especially would they strive to make a good show of sopsop kaha, wrapped-up 
bananas, and luari lewer, a fine white yam (Vl. pp. 135, 138, 159). The food for the 
great funeral feast, bud lewer, was cooked either by roasting in the fire or by baking 
in an earth-oven, but never by boiling in shells. At one funeral feast (vi. p. 147) 
pieces of the kernels of ripe coco-nuts were strung on the mid-rib of a palm-leaf 
alternately with roasted bananas, this was termed zoffo leioer, sacred food. 

On the occasion of a feast in Mer the donors make a number of erections to 
contain the food about to be distributed (pi. XXIL fig. 3). Each consists of four 
or more bamboos stuck vertically in the ground to enclose a square; to these cross 
pieces are lashed four or five feet fixim the ground to strengthen them, the four posts 
are further connected by numerous transverse lashings of rope. The whole erection thus 
forms a kind of quadrangular crate, the sides of which are about two feet or more 
long and the height ten to twelve feet. In this uncooked food of all kinds is heaped 
np, there being great emulation to provide a large display (see vi. pi. SV. fig, 3). 

The Preparation of Food. 

Food is sometimes pounded to make it soft ; this is especially necessary with dugong 
skin (p. 137), or for the use of old toothless people Stones and wooden mallets used 
for this purpose are described in the section on Domestic Utensils. 

Meat, tubers and roots were formerly boiled, and still are to some extent, in large 
shells (p. 122); now iron pots are in general use. The shells are supported in the fire 
on stones (p. 121). Water placed in a laige shell was also brought to the boiling point 
by dropping heated stones into it. 

A mash of ketai and coco-nuts is called mabus in Mer ; see also Vol. iii. p. 1, footnote. 

Cooking is carried on either inside or outside the house, more generally inside, 
but the earth-oven was always outside. There were no separate kitchens. The cooking 
is done by the women only, excepting in the existing bachelors' quarters of the Western 
Islanders (or formerly in the kwod) and when the men go expeditions in their canoes 
or into the bush. The food for the men and women is cooked together. I never 
beard of any traditions as to the origin of the art of cooking, nor of any ceremonies 
connected with cooking. 

Kitchen-middens are not formed now, nor did I come across traces of ancient refuse 
heaps. Dugong and turtle skulls and bones were formerly, and often still are, massed 
in heaps or placed in rows by the Western Islanders; this was done for ceremonial 
purposes (T. pis. XIII. XV. XXI. and XXII.), or merely to keep count of the number 


d by Google 


of animals caught in any one season, in the latter case they were subsequently distributed 
and soon crumbled away. 

The earth-oven, amai (W.), ame, net«bu (K), is universal. It consists of a large 
shallow hole in the ground (I have only seen it in sandy soil), lined with stones, 
iarai (W.), on which a fire is lighted and kept burning until the stones are red hot. 
Many of these are removed with tongs, and " native food " of various kinds — ^yams, sweet 
potatoes, taro, etc. — wrapped in banana leaves, ie placed on the bottom layer of hot 
stones. Some of the hot atones are placed on the layer of food, and another layer of 
food placed on that, the number of layers depending upon the amoant of food to be 
cooked. Small pigs are put in whole. The food is then covered over with leaves and 
sometimes mats (pi. XXII. fig. 4), and eeirtb is heaped over all. A large stone, ned 
ame (E.), is sometimes placed on the top. In an hour or two the food is cooked to 
perfection. Macgillivray states that in Muralug "the meat is then laid upon the bottom 
layer [of stones, the size of the first] with some of the heated stones above it, a rim 
of tea-tree bark banked up with sand or earth is pat up all round, with a quantity of 
bark, leaves, or grass on top, to retain the steam, and the process of baking goes on. This 
is the favourite mode of cooking turtle and dugong throughout Torres Strait" (iL 25). 

The earth-oven is invariably called a kSpa mauri, "copper Maori," a term which ia a> 
widely spread over Oceania as kaikai, "food" or "eat." The word kopa is the Maori name 
for the ordinaiy earth^oven, or more correctly for the hole in the ground. The aimilari^ of 
sound between kopa and "copper" has led to the currant belief that, as the whalers in New 
Zealand used Urge coppers for boiling down the blubber, the native method of cooking was 
called therefrom " copper Maori," that is the copper of the Maoria. W. Churchill (Beaeh-la mar, 
Carnegie Inst of Washington, 1911) suggeata that the word maori is probably the widespread 
Polynesian term for "native, indigenous." 

Hollow trees, termite-hills, or snch like are not employed as ovens. 

Vbqetable Food. 

The vegetable food consisted of the edible portions of numerous wild and cultivated 
plants, the most important being coco-nuts, bananas and yams. 

The soft kernel of the green coco-nut was scraped out of the broken nut with a 
pearl-shell or turtle-shell scraper — now a knife is frequently employed. The hard kernel 
of the riper nut is also eaten. Oil is extracted by scraping the ripe kernel and squeezing 
or straining it through a cloth. The scrapers are described in the section on Domestic 
Utensils. Coco-nut oil was largely used by the Miriam for ceremonial purposes. A coco-nut 
shrine is described in VoL Vl. p. 206. 

Coco-nut palms are plentiful in the eastern islands and on Saibai and less so on 
a few of the western islands. They were formerly absent fix)m all the Prince of Wales group, 
probably a few occurred in Moa or Badu, they were not plentiful in Maboiag. As a 
rule they are absent on the small islands; some grow on Nagir, Aurid and Paremar 
(Cocoa-nut Is.). The distribution of the banana was very similar to that of the coco-nut. 

d by Google 


The palm and the nut are both called urab (W.), or u (E.). The importance of the 
out is shewn by the following names which are used to deacribe definite varieties or 
stages in the growth of the nut. 

Weitsm Iilandl : baribari, nut in the stage used for drinking ; gautonga (Muralug), 
green nut used for driakiDg; gi, old dry nut^ gi dub, kernel; muau, sprouting nut; mtUal, 
young nut with water but no kernel; maiolgi, nut when ripening, "little bit dry." Hmray 
Illandi: bd>e-b«b« aor u, b. var. with a deep yellow hnsk — lit. flame-coloured nut; beizam u, 
var.' ; gabegeb, old nut ; gad, green nut ; gamer u, var. ; gvriguri u, a var. with small fruit ; 
hirir, small unripe nut ; kujJeup tor u, a var. with a black shell ; kwab, a var. with a bitter 
husk ; med u, var. ; met aroaro u, a var. with edible husk ; pex u or pis u, unripe nut ; « «a&, 
a very young nut ; wai v, germinating nut ; WKunenoamer u or wamiwami lor, a var. with a 
brown husk ; xaxer u, a var. with a white skin. 

As the banana is so easily cultivated, &uita at all seasons, and the fruit is so 
highly nutritious, it fonns a very important food-stuff in those places where it can be 
grown. It is moat plentiful in the fertile eastern islands. In the western islands daum 
is the name of the tree and katama of the &uit, kaba is the eastern name for both. 
Rituals to increase the crop of bananas are mentioned in Vols. V. pp. 345 — 7, vi. p. 207. 

The Miriam cultivate the following 19 varieties: avx, borom, bubitam, buruem, idaid, itaele 
(or tuw), iud, markale katam, mauko, moar, tiemepi, ontar (in the vocabulary orwar is the sucker 
of a banana), pen, pekai, nukaJde (this var. is not cooked), tereg, wop, zarttem or zortMim (this 
var. has a sweet taste and is not cooked), seber»d>er (a lai^ var.). Kerea is an unripe banana, 
n«u a ripe banana and gwnuz kaba the wild, or lit. bush, banana. In addition to these indigenous 
varieties a Lifu and a Chinese ("Cavendish") variety have been introduced. 

Unripe bananas are roasted, as are some ripe varieties. A method of improving 
the flavour of bananas is given in the section on Horticulture. 

Macgillivrsy refers to the following fruits as being eaten : " The leif/ra\ a species 
of Anacaidium or cashew-nut (the l&rgcda of Fort Essington), which after being well 
roasted to destroy its acridity, has somewhat the taste of a filbert. The elari (a species 
of Wallrothia), the size of an apricot, soft and mealy, with a nearly insipid, but slightly 
mawkish taste. Wt^r [this is the Kauralaig pronunciation, it is elsewhere called vbar (W.), 
maia or enoa* (R)], the small, red, mealy fruit of Mimusops Kaukii [M. Browniana]. 
The ajnga (a species of Eugenia), a red, apple-like &uit, the pericarp of which has a 
pleasantly acid taste. [This is the Gudang (Cape York) name, in the west it is known 
as gabu, kuai (red fi-uit), kupar (white fruit), and in the east sbbe, and ero.] 

" The fruit of two species of pandanus yields a sweet mucilage when sucked, and 
imparts it to water in which it has been soaked, after which it is broken up between 
two stones, and the kernels are extracted and eaten " (ll. 27). The seeds of the pandanus 
are usually roasted before being eaten. Leichhardt {Overland Jovmey to Port EasivgUm, 

I Tbia Duj be the same u the red nnt, moir u (vi. p. 43). 

* Thij is Uie Hnnlng name, the OMbeii (Semecupni hetsrophjllaB) i« called dun in Habniag and iger In 
Her; the fruit ii Mten ttter being cooked. « 

> Thii is esUed the "date-plum" or "wild plnta" bj the whites, another name for it is imiii^ (W^, 
wo^i (E.). The sweet nntritioiu fruit "ia dried in the son, and Btraog tot u«e in seaBOO* of souoitr" (OlU, 
p. aOl). For Mtemoniea oonneoled with (his trait, see VOIb. v. pp. M7— 9, n. pp. 303—6. 

d by Google 


p. 406) says, "The Dativea [Australians], at this seaBon [Sept 16], seemed to live principally 
on the seeds of Pandaaus spiralis and Cycas, but they both evideatly require much 
preparation to destroy their deleterious properties.... Id preparing the [Fandanus] fruit, 
when ripe, for use, it is first baked in hot ashes, then soaked in water to obtain the 
sweet substance contained between its fibres, after which it is put on the coals and 
roasted to render it brittle, when it is broken to obtain the kernels." The Western 
names for the Fandanus odor&tissimus (P. spiralis) are aixd {cAal-dan, kernel of fruit), 
bom (cluster of fruit), and kaiaa for P. pedunculatus ; the Eastern are abal {ahai-kerem, 
fruit; ggrer, leaf of pandanus) and kapder. In the Murray Islands there are, according 
to Hr Bmce, two varieties of abal, the fruit of the mammam abai has a reddish tinge 
at its stem, while the kakekake abal has a whitish one. There is very little difference 
to a casual observer between them, bub the natives pay great attention to any alight 
difference in colour between varieties of the same species. The trees have the same habit, 
the fruit is edible and the leaves are employed in mat-making. Mr Bruce gives aolag as 
the name for the very prickly-leaved pandanus of the western islands. 

Other edible fruits are: Aid>au (W.), Morinda, perhaps M citrifolia. Kalapi, kolap, 
Queensland Bean, Entada scandens; in Muralug I was informed that the kolap has to 
be cooked twice; first it ia roasted in the fire, then soaked in water and finally cooked 
in an earth-oven; the beans are then broken np (probably pounded) and eaten with 
biiu; the kolap are eaten only in default of better food during the kuki season^. 
Kuman (W.), eaten in the rainy season, v. 325. £^uru&u (W.), a yellow pungent fruit. 
Meke (^-)' mikir (E.), native almond, Terminalia Catappa. Tamad, breadfruit, Artocarpus 
inciaa. Waim (W.), waiwai (E.), wild mango, Mangifera indica, the old Western name 
was komaka (v. 103); in Brit New Guinea it is called vnxVtcai at Kiwai, veivei at 
Nala and Mekeo, vaivai by the Motu, waiwai at Hula and Sariba; waiwai at San 
Cristoval, Solomon Is.; and probably vaivai in Fiji (Vocab. p. 168). Wibu (W.), Farinarium 
Nonda, "the esculent drupes resemble in size and appearance a yellow egg-plum, when 
ripe they taste somewhat like a mealy potato, with a slight astringency" (Maiden). 

The papaw, Carica Papaya, has been introduced, as has probably the water melon, 
Cucurbita citrallus, waitain or vmitin (W,). Capt. C. M. Lewis (Naut. Mag. vl 1837, 
p. 760) states that in 1836, a piece of ground on Erub "was dug and sown with 
-culinary seeds; which Mam-moose appeared much pleased with, and promised to cultivate. 
Among them was the rock-melon, and maize, also pumpkin seeds, potatoes and peaches, 
all of which may be of essential service; as the Indiana seem at this period of the year 
[June — July] to depend principally on fhiit for their subsistence." 

> Tbe tollowing aooonnt ii taken bom Maiden'a Ditfvl Native Plantt of Auitralia (1889): "Tbate largs 
bMUi* are oaten by the aboriginals. The; are pat into tbe stone oven and h«ated in the same way and for 
the aame time m those of JvUtnnia tomtnioia (q.v.) for abont two hoon; the; are then pounded fine and 
pat into a di1l7-bag, and left tor ten or twelve hoars in water, when the; are fit for ose" {Slurrell'i tettiiaony). 
•• The natires of India also eat them after roasting and soakioR in water," p. 31. Mosele; [tioUt by a 
NaturalUI, 1BT9) safB tbe staple article of food of the Qadangs of Gape York is afforded b; then beani. 
■■Their onlf stone implements are a ronnd fiat-topped stone -.nd another long Aonioal one, saitable to be grasped 
in the hands. This U need as a pestle with which to poilid these beans on the fiat stone. Both stones are 
mere); selected, and not shaped in any way" {pp. SST-S*;. 

d by Google 


Mangrove ahoota. Hacgillivray (ll. p. 26) thua describes the preparation of biiu : 
" When the rains set in the inyu becomes the principal support of the Cape York and 
Muralug people. This is a grey slimy paste procured from a species of mangrove (Oandelia?), 
the sprouts of which, three or four inches long, are first made to undergo a process of 
baking and steaming — a large heap being laid upon heated stones, and covered over with 
bark, wet leaves, and sand — after which they are beaten between two stones, and the 
pulp is scraped out fit for use. It does not seem to be a &vourite food, and is probably 
eaten from sheer necessity. Mixed up with the biyu to reader it more palatable they 
sometimes add large quantities of a leguminous seed, the size of a chestnut, which has 
previously been soaked for a night in water, and the husk removed, or the tuber of 
a wild yam (Dioscorea bulbifera) cut into small pieces, and well steeped in water to 
remove its bitter taste." I was informed in Mabuiag that fish, turtle or dugong were 
eaten with bixu; it was said to "last long in belly" and therefore they used to eat 
it before they went out to fight. Also I was told that it made the "belly strong and 
last long," like malu damu. The latter is the sea-grass, or grass-wrack (Cymodocea), 
various species of which are eaten by dugong, but this is the only record which 1 have 
that it may have been employed as human food. Biiu was formerly eaten wherever the 
mangrove grew (cf. v. p. 98). 

Dr Seligmann obtained the following information on Mabuiag. " The fruits, uru 
(they are the sprouts or shoots of the mangrove), are usually collected by old women, 
and are cooked in an earth-oven, when done they are spread out to cool on a mat, 
beside which a woman squats. She works with her body slightly bent forwards, one 
knee being raised and bent at about right angles, so that a basket can be introduced 
between it and the ground. The woman splits the shoots longitudinally with her fingers, 
each half is then held with its skin resting against the upper part of her leg, she 
scrapes the spht surface with a shell, the scrapings falling into the basket. Later the 
basket with its contents is immersed in fresh water for twenty-four hours^ The root 
of the dtfrt plant is often cooked with the biiu; its outer sur&ce is removed, and the 
remainder is scraped fine, and made into a mash with the biiu." The derb or more 
properly dlktbu is a wild "yam" (p. 136) which was described to me as "colour like 
curry, he bite too," apparently therefore it was added to render the biiu less insipid; 
this is probably the wild yam to which MacgilUvray refers. 

Sago, bin, was occasionally imported from New Guinea, and I have heard that in 
some western islands an inferior kind of sago was made occasionally from the pith of 
a local c^cad. The sago palm "is occasionally carried by the winds and currents [from 
the Fly River district] as far south as the Prince of Wales Islands, where the natives 
scoop out the soft spongy inner wood, wash it well with fresh water, beat it up into 
a pulp, separate the brinaceous substance which &Ils to the bottom of the vessel, and 
bake it aa bread" (Uacgillivray, li. p. 62). Sago is imported into the Murray Islands 
wrapped up in long, oval bundles (vi. 261). 

> The Wftter in whieh biiu hks been washed ie oalled idiiri (Vol. ii. p. 60). 

d by Google 


Sugar cane, geru (W.), neru (&,), is grown in several islands ; Mr Bruce states that 
it is indigenons to Mer, where the following varieties are grown: aspeaaapea neru, eats 
neru, golegole neru, govgu, koair, mammam, mowat, aermairmair, zemkep. The kaiaragam 
is a Mabuiag variety, the stem of which is easily broken. When ripe the pith ie chewed ; 
the cane is not prepared or cooked in any way. 

Wargon (K) is an aroid with a leaf resembliog that of taro, but the stem only is 
eaten in Mer; it is veiy pungent. 

Tntwn and Kooti. 

There are numerous varieties of yam (Dioecorea), but in a few instances I aaa not 
sure that the particular plant is strictly speaking a yam. 

The wild yams, or yam-like tubers, of the western islands are : hoa or }ma (v. 91), bttd, 
tiaibau (Vol. v. p. 156), cfib6u or dw* {p. 136). The cultivated varieties are: gi^xm (the general 
name, koi nd, for yams), bizar (purple), ketai or kutai (a climbing form), ktiiJedai (red), naguai 
(a Tutu name), lagu (purple), aouur (a kind of yam eaten during the dry season, uxiur), 
lege (a long yam in Tutu). The Miriam names are : Imoer (the comprehensive name, an n«t, 
for yams and for vegetable food in general), bad fewer', borom maiai leioer (a flat aud sweet 
variety), bug« usare, daibar, dob, eUt, gagaba, go's*, ipigaba, tworiuur, Icalngaba (perennial, white), 
kepiabeg, kimiar, kuirkwr leuxr, kuttbager, tamar leuxr, madupenatt*, mamTnam tuare, mapis*, 
mutmrU v*are, penau or peneu (pink variety), r^, aap*, segei, aorbe hep letoer, tap*, tuan, 
utarip*, wdbed, tpahvna, uKntnatoaima *, uMuer. The ketai is a perennial, climbing kind of 
yam, one or two varieties occur, such ao the borom kelai, *ru kelai. Yams are cooked by 
roasting or boiling; sometimee yams and taro after having been roasted are boiled in coco-nut 
"milk," the maeb so produced is called papai la Mabuiag. 

Several varieties of sweet-potato (Iponuea Batatas) are known: the genera! Western 
name is urugahau or wegabau, with the varieties nunnuri, tapan. The three indigenous 
forms in Mer are nuri, orgagah and mamimam orgagab, but several other varieties have 
been introduced. At the present time the name kumcUa is widely employed, having been 
borrowed from South Sea men (Lifu, kamala; Samoa, umala; Banks Is., Fiji, New 
Zealand, kumara, eta). 

Taro (Colocasia macrorhiza) is called in the west ginn or guin, the white, black, 
and green varieties are respectively known as kaHtai, kubikubi, and 'un6a ; other names 
in the Vocabulary (Vol. III.) are gvamakia/m, and Jdma. The Eastern name is aneg. 
I am not sure whether the indigenous, edible gaine is the same species as the above. 

"Arrowroot" (?Maranta) is called gaai (W.), and kep aahez in Mer. 

Formerly the tuberous rhizomes of the aroids bode or hadi and ager (E.) were eaten 
(pi. XIL fig. 5), but I have not been able to identify them. These are now very rarely 
eaten, and I have suggested (VI. p. 2) that the Miriam tales in which the use of 
this food is mentioned may date back to a time before the cultivation of yams and 
sweet-potatoes, when the islanders were merely collectors of food. In common with many 
aroids these rhizomes probably contain a poisonous latex which is dispelled by heat, 

' All these names were sent to me bj Ui J. Bnioe, eioepl those marked *, vhioh were obtained b; the 
Eipedition ; we also obtained Eeveral of the former names. 

d by Google 


consequently they had to be cooked in an earth^oven. The phraae osed in this connection 
was "too much be fight, when we roast him be no fight"; in jargon English the word 
"fight" also includes the idea of something pungent which burns the month. The rhizome 
of the ager often attained the size of a man's head, Beferences to this food occur in 
Vol. VL pp. 6, 9, 11. 

The following roots were eaten in Mer : gauda, a kind of vine, weskep, Pueraria 
pbaseoloides (root eaten raw or cooked), mar, a grass-like scented plant, the leaves and 
roots of which are edible (Vol. ii. p. 183), and vmgao. 

The following were described as edible plants in the western islands, but there is 
no information as to what parts were eaten : gobeg(Ae, gitgabe, igaru, iku/r, kuaJc, vzu. 

Animal Food. 

As there were no indigenous land mammals, milk and ordinary fiesh food were 
unknown. The dingo was domesticated but, so &ir as I could make out, never eaten. 
I think it doubtful whether the pig was introduced into and allowed to run. wild on 
any of the islands before the coming of the European. Dr Seligmann was informed 
that it was once domesticated on Mabuiag, but was given up as being a nuisance, for, 
as tbey said, " too much he humbug," probably meaning by this that the pigs danu^ed 
the gardens too much. Dr Seligmann found a few old men in Mabuiag who had recently 
leamt in New Guinea to eat the aapar, flying-fox (Pteropus), but this was considered 
a somewhat objectionable practice ; Waria informed me that " big men " eat it in Mabuiag. 
In the stoiy of Mutuk it will be remembered that Mutuk and his crew were transformed 
into flying-foxes (v. p. 90). The porpoise or dolphin, bid, is not eaten in roost of the 
islands; Waria informed me that the Eulkalaig (natives of Muralug, etc), eat porpoises, 
bid, but the people of Mabuiag and Saibai do not. On the other hand, a Muralug man 
said to me, " Me fellow no kaikai bim, he too &t ; Masig, Pourma (Parema) and mainland 
(Australia) man kaikai him, 'cause be no savvy spear dungal (dugong)." Marrow is 
unknown as neither the dugong nor tturtle have any in their bones. 

The dugong is a very important article of food, more especially in Mabuiag owing 
to its contiguity to the great ree& where that sirenian abounds, but it is not much 
eaten in Mer. All the soft parts are eaten except the gall bladder and probably the 
brain as the latter is not easy to extract. Pieces of dugong meat are generally roasted 
over the fire, and small pieces are often eaten half cooked ; sometimes the meat is boiled 
in shells. Even the blood and oSsX are carefully collected. Strips of dugong skin with 
the blubber attached are smoked, making a fairly good bacon. I have seen such strips in 
Mabuiag on a cord hanging out of doors, they were prepared in the dry season for use during 
the north-west monsoon; in Mer it is called kaiger, before it is eaten it is roasted and 
beaten between stones to make it tender. Macgillivray says, " The blubber is esteemed 
the most delicate part; but even the skin is eaten [in Muralug], although it requires 
much cooking in the [earth] oven" (iL p. 26). Waria informed me that dugong meat 
was preserved by first cooking it in an earth-oven, after which it was smDked,*or it was 
niaBted after being cooked in an earth-oven, then scraped, dipped in salt water, and 
finally dried in the sun. 

H. Vol. IV. 18 

d by Google 


One method of carving a dugong is shewn in pi. XXII. fig. 2. An annular cut 
ia made at the end of the :abdominal cavity, and a cheek piece including the eye is 
cut, which again may be longitudinally divided, and the upper border of which extends 
beside the median dorsal line to the posterior cut, three lateral vertical cuts are made 
in the region of the shoulder, and a number of longitudinal cuts along the flanks. 

Various birds are shot and eaten, the most important being the Torres Straits 
pigeon, gainau (W.), daumer or gaino (E.), Carpophaga luctuosa. Birds' eggs were eaten, 
more especially those of sea-birds, which were collected on the sand-banks where they 
breed. According to a folk-tale (v. p. 12), the mound bird, mrka (Megapodius), occurred 
on Boigu, its eggs were eaten. 

1 was informed by Waria that while the natives of Mabuiag do not eat karum, 
"iguana" {the monitor lizard, Varanus), those of Muralug, Moa, Saibai, Dauan, Boigu 
and New Guinea do. The Madvb eogo te of Mer (vi. p. 232), according to Mr Bruce, 
both men and women, eat snakes, which are roasted first. Snakes are not eaten in 
Mabuit^; we were told that on the coast of Daudai "snake-eater" is a term of reproach. 

The various species of turtle and their eggs form the most important meat diet 
of the islanders, and the turtle has the advantage over the dugong that the catching of 
it does not injure it, and as it la perfectly helpless when turned over on its back it 
can be kept alive for a long time. 

All the soft portions of the green turtle, including the blood, are eaten with the 
exception of the gall bladder; but on certain occasions women are not allowed to eat 
turtle or their eggs (v. p. 196). 

It is not customary for the Miriam to eat the hawksbill turtle, though they say 
that the other islanders to the west do eat it. Gill says, "The flesh is eaten by the 
Straits islanders and by the whites engaged in shelling" (l.c. p. 291), but he refera 
to the Western Islanders only. Macgillivray states that " the hawksbill turtle and its 
eggs are forbidden to [Muralug] women suckling" (ll. p. 10). Waria, of Mabuiag, informed 
me that fathers do not let small boys put their bands in the blood of his turtle, unoa, 
since if the blood got below the nails they would have sore places and headache. In 
Vol VI. p. 227 I refer to the death of three infants beii^ attributed to their 
mothers having previously eaten some boiled flesh of this turtle, and the Miriam 
have a special form of nefarious magic which causes people to die after eating the 
kesur (vi. 227). 

According to a folk-tale (VI. 48), a turtle was placed on a framework over a fire, 
and it was boiled with sea water in its own shell ; to do this the plastron must have 
been removed and the animal placed on its back, the carapace would then form a 
convenient saucepan with the meat in situ. Turtle are cooked in the same manner aa 
dugong, Macgillivray says, "The Torres Straits Islanders are accustomed to dry the 
flesh to supply them with food during their voyages. The meat is cut into thin slices, 
boiled in a melon shell, stuck upon skewers, and dried in the sun. Prepared in this 
manner it will keep for some weeks, but requires a second cooking before being used, 
on account of its hardness and toughness. The fat which rises to the sur&ce during 
boiling is skimmed off and kept in joints of bamboo and turtles' bladders, being much 
prized as food; I have even seen the natives drink it off in this hot fluid state with 

d by Google 


as much. gusto as ever aldermiui enjoyed his elaborately prepared turtle soup" (ii, p. 23). 
In Mer turtle are preserved for several days by keeping them above a slow fire. 

FrogB were eaten by all the Miriam, who used to collect them by basketfuls; they 
were gutted before roasting. The akin of the head, mai gegur, was removed as it was 
very pungent, au kapkap, to the palate. The Uabui^ people apeak conbemptuously of 
the Miriam and Ej-nb natives as foreigners who eat frogs. 

Numerous kinds of fish are eaten. The larger fish are gutted before being cooked, 
and may be boiled in freeb water in a shell saucepan, alup or bu, roasted over the 
fire, or cooked in an earth-oven. Fish are often wrapped in pandanus leaves when being 
roasted (vi. pp. 16, 25). When not required for immediate use the fish are dried in 
the sun, dried and smoked, or slightly roasted on a bamboo &ame hung over a fire 
or on a light wooden framework, noai, noi (W.), taAor' (E.), under which a fire has been 
lit; in Mer fish ao prepared was termed takar iar. The natives of Maaig, Waraber, 
and other central islands fi-equently brought takar Iar, along with turtle-shell and other 
spoil from the sea, to Mer to trade with. The small fish known as tup are generally 
boiled in a shell 

Dr Sehgmami was informed that sharks, except the carpet shark, tm (Crossorhinus 
dasypogon), were not eaten at Mabuiag, but were eaten at Saibai and Boigu. The young 
shovel-nosed skate, katgas (Rhinobatis), was also eaten in the three islands. He was 
also informed that the U2t (Synancidium horridum), was employed in magic in Mabui^, 
ajid that young men might not eat it (" he too cold ") although old men and 
women might 

Numerous kinds of molluscs and crustaceans were eaten. Holothurians (tripang or 
b^be-de-mer) were and are not eaten, although in the western and central islands large 
numbers are prepared for traders to be exported to China. 

The larvie and pupie of a Longicom beetle are considered delicacies in the western 
islands, they are eaten raw or roasted. The Madvh zogo men of Mer ate locnsts, hut 
only male ones, jre&o pern, ; they were eaten raw and wrapped up in the leaves of a shrub 
named paiwer {J. Bruce), in the Vocabulary, paiwa is the " chili " plant. 

I never heard of any unusual substances, such as clay, being used as food in times 
of scarcity. Pregnant women in Mer frequently eat small lumps of a greasy chocolate-like 
earth to make their babe light-coloured. Sometimes it is eaten raw, but, in order to 
ensure the best results, it is wrapped in a banana leaf and roasted (vi. 105). Children 
eat the same kind of earth till about five to seven years of age to make them strong, 
brave and hardy (vi. 111). So &r as I am aware earth is not eaten elsewhere, I was 
definitely informed that it was not eaten in Mabuiag. 

Anthropophagy purely for the sake of eating human flesh never occurred so far as I could 
discover. Among the Western Islanders certain portions, generally the eyes and cheeks, 
1 nw pour was mors p&rticnlarlj the framework on whioh corpie* were dried {-n. pp. 18C, U8>.. 

d by Google 


of enemies killed in battle were eatea in a raw or partially cooked condition (v. p. 800). 
Similar portions were given to lada on the occasion of their firat fight, the avowed 
object being to make them brave and fearless. I was informed that a Mabui^ warrior 
would sometimes hold up the head of a man whom he had killed and beheaded and 
let the dripping blood &11 into his mouth, and he might give some to a yonng man 
who had accompanied him but who had not yet killed a man, sajring, "Tou do not 
know how to fight. Tou drink it and it will give you a strong heart." Tutu men 
also drank the sweat of renowned warriors, and all the scrapings from their gfxcy 
finger-nails were mixed with their food in order "to make strong and like atone; no 
afraid." A Tutu warrior would tear out the tongue of a man whom he had just killed 
and eat it oa the spot (t. p. 301). Part of the training of a sorcerer in Mabuiag was 
said to consist in eating the decompoeing flesh of a dead man (t. p. 321). 

It was formerly the custom in Mer for men to drink the jnices which exuded 
from a desiccating corpse. This "grease belong dead man" would also be mixed with 
food and eaten. The juices of dead women were never drunk or eaten (vi. p. 159). 
There is some evidence that portions of the body were occasionally eaten. It is probable 
that the juices of dead relatives only were consumed, in which case the practice must 
have had a different significance from that of the Western Islanders. Only two cases 
of cannibalism among the Miriam are narrated in the folk-tales (vi. pp. 17, 53). 

Food Restbictionb. 

N'o member of any clan might kill or eat the totem of that clan among the Western 
Islanders; two exceptions to this rule are noted in Vol. T. p. 186. Owing to the absence 
of totemism this rule does not apply to the Miriam (Tl. p. 250) despite Hunt's statement 
to the contrary {Joum. ArUk. Intt. xxviil. p. IS). 

Among the Western Islanders certain food was tabooed to lads during the period 
of initiation (v. pp. 210, 212, 216, 269, 270), but we could not discover that this 
prohibition obtained among the Miriam. 

At certain periods women were debarred from eating particular kinds of food 
(y. pp. 196, 202 — i; vi. p. 105). Women may not eat any kind of bird in Mabuiag, 
as birds are believed to be aphrodisiacs; for as pigeons fly from tree to tree so the 
woman would desire one man after another. The women of Muralug, Moa, and Saibai are 
permitted however to eat any kind of bird. Examples are given in Vol. VI. pp. 105, 106 
of the effects which are supposed to be produced on the offspring when a pregnant 
mother eats certain kinds of food. 

Taboos were occasionally imposed on gardens and garden produce in Mer (Tl. 
248, 249). 


Fresh .water is drunk and a great deal of coco-nut " milk " in those islands where 
the palm abounds. As fresh water is generally scarce coco-nut " milk " is fi:«quently 
substituted for it for purposes of boiling and imparts a delicate flavour to the food. 

No fermented liquor has ever been made and till the European arrived none was 

d by Google 


drunk ; nnfortaiiately many natives have acquired a liking for " gro;;," and although 
it is illegal to sell beer or spirits to them, they sometimes manage to get some. Eava 
was unknown^. 


Tobacco is the only narcotic employed, and it was formerly invariably used for 
inhalation from the characteristic bamboo pipe (fig. 165, pis. III. fig. 3, XXIV. fig. 2). 

The Papuan pipe is made from a piece of bamboo frt)m over a foot (30 cm.) to 
between two and three or even four feet in length. The natural partition at one end 
(and the intermediate one, if such occurs) 
is perforated. At one end of the pipe 
there is always a complete partition, and 
near this a small hole is bored into 
which a narrow wooden or cane tube 
a few inches in length is inserted. The 
tobacco, previously cut and crumbled, is 
put in this and the open end of the 
pipe applied to the mouth, and by suction 
the pipe is filled with tobacco smoke ; 
sometimes the bowl will be inserted in 

Fis. 166. TobatMO pipes (from Jnkas, i. p. 160). 

the mouth and the smoke blown down through it into the pipe^. As soon as the 
pipe is filled with smoke, the right hand is placed over the open end and the bowl 
is removed. The small hole is applied to the mouth and the smoke sucked through 
it afler the withdrawal of the hand from the open end (pi. XXL fig. S). The length 
of the pipe causes such a draught that the smoke is violently inhaled. When a man 
has had a suck he will put bis right hand to the open end to prevent the further 
escape of smoke and psss it on to another, who receives and maybe transfers it to 
a third in the same manner after taking a suck. The women usually prepare the pipe 
and pass it on to their men. 

> Ouffioda or lurmata, as kaTB U called, » drunk on oerUin oooaaionB in boom places on the ooaat of 
Dandai, u at the iaitiatian teoat of lads (cf. E. Beaidmoie, Journ. Ant\. Init. xn. p. 460). HaoForlane found 
that kara wai drunk bj natives near the FI7 'Biytt, " Here It Is the boTB who obew the root" {Anions t>^ Catmibak 
of Nt» Oviiua, 1888, p. 136). IXAlbertia [Nns Oainta, 11. p. 197) aa.ja that he was given at Mawata some roola 
of a plant which the DativeB chew for its nsrootiD and intoiieating propertiea. Malna explained that to eiperienoe 
ita intoxicating eflecta perfeotl; tobaooo should be smoked after obewing. Apparently it indooes pleasant dreams 
or viaiona. Sir WiUiam HaoQr(«or (Ann. Rep. of Brit. New Ou<n<a, C.A.I. 1693) in a deapatoh dated April 37, 1891, 
givei details of the obewing of the rootlets and stems of Fiper methystioam (kava) b; the Uasingara of Dandai ; 
41 thia has been reprinted in the Journ. AMh. Intt, iii. p. 201, I need not give farther partioolars. 

* Hosele; {NoM by a Naturalut, p. 8G6) fonad that the most prized possessions of the Qudangs of Gape 
York w«r« tobaooo pipea which were "proonred b; barter from the Momj' Islanders. No donbt the Anstraliana 
have learned to smoke from the Hnrray Islanders." They employed a small oone ot green leaf, instead of a 
small bamboo tabe, as a bowl. "A man, or oltener a woman, then opening her month wide oovers the oone 
tad lighted tobacco with it and applies her lips to the bamboo all around it, baring the leaf cone and homing 
tobacco tboB entirely within her month. She then blows aod foroes the smoke into the cavity of the bamboo, 
keeping her hand over the hole at the other end and olosiiig the apertnre at toon as the bamboo la fall." He 
oomparet this method with the one by Jokes observed in Damat (1. p. 166), in which the month wai applied only 
to the open end ot the bamboo ; bat as stated above both nethods are In vogae in Torrea Straits. 

d by Google 


The effect of this kind of smoking appears to be very severe. The men always 
seem quite dazed for a second or two or even longer after a single inhalation, but 
they enjoy it greatly and prize tobacco very highly. I have seen an old man reel 
and stagger from the effects of one pull at the pipe. Jukes says of the Erub people 
(I. p. 187) : " In smoking their own tobacco [which is of a light brown colour], they 
break off a piece from the plait' into which the leaves are twisted, and wrap it in 
a green leaf to prevent its setting fire to the wooden bowl. A woman is then deputed 
to fill the bamboo with smoke, and on its being passed round, each person takes a 
long draught of smoke, which he awallows, apparently with considerable effort, and stands 
motionless a few seconds, as if convulsed, with the tears in his eyes; be then respires 
deeply, and seems to recover. They call it 'eree oora' [eri ur] (to drink heat or fire), 
and, patting their stomachs, seem much comforted after it. I tried their tobacco, but 
found it intolerably hot and strong." Macgillivray offers similar testimony : " On several 
occasions at Cape York I have seen a native so affected by a single inhahttion as to 
be rendered nearly senseless, with the perspiration bursting out at eveiy pore, and 
require a draught of water to restore him; and, although myself a smoker, yet on the 
only occasion when I tried this method of using tobacco, the sensations of nausea and 
&intne88 were produced" (i. p. 126). 

A white acquaintance of mine who at one tioie took to smoking tbe Papnau pipe gave 
me the following account of his experienoes. The inhaled smoke ie retained for as long as 
poMible and let out through the mouth and nose. There is a very strong draught through 
the pipe which drives the smoke right into the lungs. On the first occasion this nearly chokes 
a person and this experience generally satisfies all curiosity. After a single Inhalation the 
confirmed smoker feels happy and sleepy ; the effect is much the same as with opium but 
with none of the illusions; all the aeusee are deadened, and alter a whiff or two, the smoker 
goes off into a deep), heavy, but not refreshing sleep. The smoke is quite cool. My informant 
smoked in this manner for about six months, but had to leave it off as his heart became 
affected, but not his lungs. The heart's action was weakened, and be had a dry barking 
cough. The smoking made him generally lazy and indolent, but extremely nervous. He always 
took a pull when the effect of the last wore oS, and bad a great hankering after it. 

At the present time most of the natives have adopted a short; clay or wooden 
pipe and roll up cigarettes in bits of newspaper or banana leaf as the case may be. 

The pipe is called sukub morap (W.), 2u6 (£.), and the bowl turku (W.), tarkok (B.), 
tobacco being «ujtu6a or tugvb (W.), aokop (R). The diy leaf of the banana used as a 
wrapper for a cigarette is called taugoi (W.) and that of the Mimusops, vharau ris. The 
Mabuiag terms for smoking are: gamu loidai, light it; ngalkai, blow or suck (smoke 
into the marap); sukvha wani, swallow tobacco (i.e. smoke); wax, exhale (smoke). The 
decoration of the pipes is dealt with in the section on Decorative Art. 

The bowl is generally made of a piece of narrow bamboo, the lower end of which 
is cut into a cone so as to fit into the hole, gud (W.). The length varies in our 
specimens from 10 to 22-5 cm. Sometimes it is decorated in the manner characteristic 

> Brookett {Vcyaft, ISS6, p. 22) Ba;a, " After the tobaooo it dry thej plait it like a three jara eennet." In 
the Naut, Mag. (vi. 1887, p. 764) it ia stated: "They alao cnltiTftle the tobacco plant, wbioli the; prepsre tor 
srookiiig, hj dtjing the Iv&vea, and twietlng it op into 'fige.'" 

d by Google 


of pipes; Brockett, Narrative, 1836, pi. 4, illustrates threfl decorated Bpecimens and one 
is shewn in the Album, i. pi. 318, No. 3. A large quadrangular heavy wooden bowl 
from the mouth of the Fly River is figured in the Album, II. pi. 188, No. 1, it tapers 
gradually helow, is 134 mm. long, and is carved on two sides with zigzag patterns and 
on the others with a design which probably represents a nose (cf. p. 26). Mr Robert 
Bruce presented a similar bowl to the Glasgow Museum which is carved with various 
common patterns ; it is stated to come from Mer but is probably an importation. It 
is 184 mm. long and 57 mm. square above and 44 mm. . below. I believe that many 
of the Daudai bowls are characterised by a fillet or raised ridges round their middle. 

MacFarlaue says (Among the Cannibah, p. 125): "When it [smoking] was introduced we 
cannot say. In 1871 we found the natives of Saibai and Eat&u smoking from bamboo pipes, 
and on our voyages up the Baxter and Fly rivers found tobacco plantations far in the interior. 
On the south-east peninsula however it is a recently acquired habit. They did not know the 
use <rf tobacco when we, first met them.... On the peninsula, in the vicinity of Port Moresby, 
the tobacco is rolled in a leaf, and the smoke inhaled from the end of the bamboo." 

It seems probable that tobacco was introduced to Torres Straits from New Guinea, and, 
further, that it came from the north down the Fly river. From the mouth of that river 
as far west as the Dutch boundary it is known by the same name, avJcuha, Eiwai and Mawata; 
takaha, Dabu; takupa, Bugi; tukrtha, Dungerwab; tokttva, Bangu. According to La wes (grammar 
and Voeabvlary of Jfotu, 1668), the word itufeu is emplo}^ by the Motumotu, Maiva, Kabadi, 
Uoto, Eerepnnu, Aroma and South Cape natives. The name fiau&ou for the bamboo pipe 
extends, according to Lawee, from South Cape to the Motu districts inclusive, but, passing 
up the coast, it is kemtma at Eabadi, trnn at Maiva, and jtifet at Motumotu. Bau is the 
Motu name for bamboo, and maraip, marep, morap, etc., over the whole district west of the 
Fly river. The Papuan tobacco is referred to on p. 150. 

Betel-chewing was not practised in Torres Straits, though I believe it is done to some 
extent in neighbouring parts of New Guinea; the areca nut is called tmu by the Western 
Islanders. The people of the adjacent coast of New Quiuea are said to chew the pungent 
fruit of the m«idu palm, Nipa fruticans (see Tol. v. p 15). la bis account of the natives 
of Kiwai, MacOr^or, in a despatch dated Dec. 9, 1869, says: "They know betel-nut, time, 
and pepper, but they do not use them, just as they know of the Pijier methytticwn (kava] 
which also it is not customary to employ, and they do not cultivate it. I have not seen 
a person with stained teeth on the island I do not know of any stimulant they possess, 
unless we may call by such name their hom&growa tobacco, which they cultivate in considerable 
qnantities" {Ann. Rep. Brit. Nno Guinea. C.A. 105—1890, p. 41). 

d by Google 


Except when kept under cultivation the land ia covered with " bush." The nature and 
density of the vegetation vary somewhat with the soil; thus it is on the whole of a more 
luxuriant character on the volcanic eastern islands than on the western high islands or the 
central coral banks. In any case vegetatioD is apt to grow rapidly when not kept under 
cbnstant control, which circumstance renders horticulture rather h^td work. 

The lighter work in the gardens is done by women. The earth is turned over with 
a digging-stick, pat or potur (W.), wet (K) ; the Western verb for digging a garden is gowa 
pc^. The diving-sticks are simple staves, with one end pointed or more characteristically 
slightly turned up at the digging end and cut to form a flat sur&ce (fig. 166). The end may 
be hardened by being burnt. They are made of various kinds of wood, thus argerarger wet 
is a digging-stick made of that wood. The longest digging-stick which we collected in 

Via. 106. Front vid side vieira of tba end of e. digging-Btink, Mei. 
Uer is 178 cnu long (67 in.), and is crooked and ill made; two stout ones are 83 and 
94 cm., four stout argerarger wet are 75 cm. (29J in.), and two thin ones, which have the 
bark on, are 61 and 69 cm. — these are also used for husking coco-nuts, as probably are 
all digging-sticks. 

Although many natives of Torres Straits must have been familiar with the shell hoe which 
is employed in soft ground in Dandai, they do not appear to have adopted the implement. The 
Daudai hoe is made of a piece of melon ahell inserted into a hole cut through a club-Iike stick 
and wedged in position by pieces of wood. This interesting implement has been figured by 
D'Albertis (il. p. 378, fig. 11), by myself (Sead-ffunterg, 1891, p. 110), by B. Etheridge, jnn. 
{Proe. Lirm. Soo. N.S. W. IX. 2nd Ser. 1894, pi. vi. p. 109), and in the Albums l. pL 346, No. 6. 

Qardens are frequently fenced, mainly for the purpose of keeping out pigs. Fences, 
pa (W.), kar (E.), are made of various substances, but mainly of saplings and bamboo; 
thus «J kar is made of mangrove wood, beriberi kar is a rope fence, and kegar kar a 
stone fence. 

In the Murray Islands each group, le (vi. p. 172), has its territory with defined 
boundaries which are well known to ite own members as well as to neighbonring groups. 
Mr Bruce informs me that these - territories are called pur god, which he translates as 

d by Google 


tribal laud. Each &mily has ite own portion marked (or supposed to be) by recognisable 
boundaries, which are generally large standing trees or a ridge of earth ; the latter is made 
by each &mily, when cleaning their land, throwing weeds and refuse on the boundary. Id 
some cases the ridge thus formed is well-defined, but in others the boundary is very vague 
and frequently gives rise to land disputes. Some &milies have larger areas than others 
through members of the family dying, the land thus falling into fewer hands. Land is 
sometimes owned in several pur ged ; a man may bold bis father's land, perhaps a portion 
of his mother's land, as well as bis wife's share of her pur ged, aad the same holds good with 
allotment lands on the beach, on which the houses are built. Every man is a landowner. 
When a &mily has been very prolific for several generations, the portion allotted to each 
member ia very small, and a network of boundary lines results. This may lead to 
litigation in consequence of encroachment. The natives are in such a case very eager to 
enlarge their gardens at the expense of their relatives or neighbours, much bitter feeling 
arises amongst them from this cause and each maintains that his boundary mark is the 
original one. 

The time, gedub eged or gedvb iami, for cutting down the timber and clearing the 
undeigrowth in preparation for planting new gardens begins about the end of August, and 
ccmtinues till the wet season sets in about the end of the year. In fact as long as the wood 
can be burned off. Some begin clearing earlier than others. There are various signs to 
indicate the beginning of the clearing season, such as the flowering of the sdbe^, waiwi^, 
meaur and hid trees, and the ascension in the north-east horizon of the stars Usiam (the 
Pleiades) and Seg (Belt of Orion). Usiam appears first, Seg a httle later, as when it 
appears Usiam will be about nine degrees above the horizon ; they consider it time to 
prepare their gardens when Seg is first seen, but they term it " Usiam time." Another 
sign is when Tagai (Crux)* is in its declension and seen in the southern horizon after 

The work of clearing the land, itara, and preparing it is divided between the two sexes. 
The women clear the undergrowth and cut down the small bushes ; the men cut down the big 
timber if necessary and do all the axe-work. Grass land is generally cleaned by the women 
only, but Mr Bruce notices that the men now begin to take a share in this work. The minor 
clearing is done by means of large scrub-knives purchased from Europeans. Trees are now 
felled with trade iron axes (tomahawks), but formerly shell axes were employed (p. 126); 
when I asked whether it was not hard work to use such an axe, my informant replied 
laconically, " Plenty sweat." 

When the wood, branches and bushes are thoroughly dry, they are set on fire to clean 
the ground. Those who have got their timber burnt oflF early begin to plant ketai, yams and 
bananas in October, paying attention to the altitude of Usiam. They also plant when they 
see the kaJcigaba and ketai plants (p. 136) beginning to shoot out their vines — these two 
are perenniids. At the beginning of the planting season the yams are planted whole, so 
as better to resist the drought, but in November, when they generally get rains, they cut 
the tubers into pieces, as we do potatoes. They are planted at fairly regular intervals. 

1 p. 188. ' p. 154. 

' Tagai ia a very Isi^e constellfttion vhich inolades several of ourg {see Aatronocny) ; «8 ideatified the Southern 
Cron M the left h&nd of Tagai. 

H. Vol. IV. 19 

d by Google 


When a ketai is planted it ia provided with a small pole for the plants to climb ap 
(c£ VI. pi. IV. fig. 4). 

The whole of the planting should be finished in November, but just as elsewhere there 
are some men more industrious than their fellows, who pride themselves on having early 
crops, while others delay and even run the time so fine that the rains set in before the 
scrub has been burnt ofT. The plants have then to be dibbled in anyhow without due 
preparation, with the result that the crops are poor and late. The latter class, however, 
is the exception. 

In seau ged, that ia coco-nut groves, or close-timbered land with a rich soil and good 
shade, they plant yama, k^ai, kakigaba and bananas. The yams are planted some distance 
from the trunk of a tree, and bamboo poles up which the yam vine climbs are placed in 
the ground beside the plant, the other ends resting on the tree. The same plants are 
also grown in open timber land, kebe aeau ged. 

In wargor ged, that ia land with no large trees on it or very spataely timbered, or land 
;. with bushes or with grass only, they plant sugar-cane, bananas, sweet-potatoes, papaws, and 
some varieties of yams, but for the last they always prefer teau ged. All the sweet-potatoes 
are planted in this land. The women generally clear the ground of grass and the men dig 
up the mounds in which the cuttings of the vine are planted, for the tubers are never 
planted; the women cut up the vines and assist in planting them. Sweet-potatoes are 
planted when the heavy raina commence in January and February, as they have to wait 
until the vines of the old crop have grown strong enough for taking cuttings. If the vines 
have not sufficient mobture they die ofT; thus sweet-potatoes are nearly always planted 
whilst it is raining or immediately after a big downpour. They are never planted in drills, 
but always in hillocks or mounds. 

The roots of the ketai arc called t^ and the clusters of tubers mot (E.). When digging 
the tubeiB the original or parent tuber is never disturbed, but the new tubers are removed 
from it annually. It is believed in Mer that the parent tuber, apu ketai (mother ketai), is 
everlasting and will keep on indefinitely producing new tubers, werem ketai (children of the 
ketai), if not injured in any way (cf Vi. p. 51). 

In former days all land outside coco-nut groves was called wagor ged, as the old people 
say that there was no gross land, the only grass being that large coarse kind used for 
thatching the housea. The old wagor ged was like open forest country, and the hills were 
also well timbered where now there is nothing but grass. Mr Bruce informs me that be can 
still trace the stumps of trees on the hill-sides. I remember hearing fi^m Mr Robert Bruce 
in 1889 of the enormous quantity of timber that was cut on Mer by the missionaries to 
construct houses, but more especially to build boats. It looks as if this ruthless exploitation 
had permanently afiected the producibitity of the island. 

A new garden patch is called kerkar gedub (new garden), after it is one season old and 
has been cropped it is called keas gedub, and after two yeara, gazag gedub. They do not 
replant keaa or gaea^ gardens, but always select a new patch that has been fallow, where the 
• undergrowth is four or five yeara old. Thus every year a man clears new ground and lets 
another portion lie fallow; this chiefly applies to yam and banana patches. They believe 
that in order to get good yama the planting must be made on fresh soil. They recognise 
that the banana crop exhausts the soil rapidly, so when a banana patch is planted they crop 

d by Google 


it for Bome years, until the fruit gradually diminishes. Ab a rule they do not root out the 
old stocks but allow the place to be overgrown and the bananas to die off. Sugar-cane they 
crop for two or three years ; as the canes are cut the stools throw out fresh canes, after the 
third year they are much deteriorated and the land is left to lie &Uow. Sweet-potatoes are 
cropped for two aeasons ; they never dig the vines out when taking the tubers, but grope for 
the latter with their bands, leaving the vine whole. At the second season the crop is of an 
inferior quality, the tubers being very stringy. 

There is no regular rotation of crops, as it is generally the same kind of plant that is 
cultivated on the old Uud, but, as we have seen, tubers usually are grown for only one season 
<Hi the same patch, which is then allowed to lie &llow for several years. This necessitates 
the clearing of fresh ground every year. 

The practice of leasing or loaning land in the Murray Islands has been described by 
Mr Wilkin in Vol. vi. (pp. 165 — 7), but Mr Bruce has recently given me the following fresh 
information. The natives, especially relatives, are very fond of getting loans of land and 
giving others in return, as they like a change. In loaning land no time is stipulated nor 
payment made ; the tenant gives a present of food from the first crop, but this is not a 
compulsory payment, though it is customary. Generally this is the only payment, and the 
owner expects nothing from the keas or gazag crops; it is purely optional on the part of 
the tenant whether he shall give anything to the owner from them. Formerly when the 
tenant left the land he could take away whatever he chose in the way of plants or even 
destroy what he did not take. The lease was generally dissolved on account of the parties 
quarrelling or because of some dispute among the families, in which case the tenant would 
be immediately ordered to quit the land; he therefore tried to do as much damage as 
possible to that portion of the crop that be could not carry avray. The law now is that the 
tenant must receive due notice to quit, and that too at a seasonable time so that he may 
be able to replant what plants he removes, but nothing may be destroyed. He takes suckers 
of bananas, yams and so forth, but must leave the parent plants intact. The people 
aj^reciate the new regulations, although at times, if there has been sore feeling between 
the parties, the tenant would prefer the old feshion so that be might vent his spleen on the 
owner's property. 

The banana is easily cultivated, and requires no care except the weeding of the plantation. 

In the Murray Islands a bunch of green bananas while still banging on the tree is 
packed with the leaves and small branches, lialia, of certain trees and shrubs, and wrapped 
round with ewa, the cloth-like spathe of the leaf of the coco-nut palm ; the whole is neatly 
tied round diagonally in opposite directions with ked (pi. XXI. fig. 1). The ked that I saw 
employed for this purpose consisted of strips of coco-nut husk knotted together (p. 89). 
The bananas are allowed to mature and ripen under this covering; eventually their skins are 
beautifully and richly coloured in shades of brown, maroon and red. The bananas are very 
sweet, and the leaves impart a sweet odour to them. They are very " rich " eating, and the 
wrapping process certainly greatly improves the fruit. They are known as sopsop kaha, 
topem ita6a, avaneraumer kaha, all of which mean " wrapped-up banana " ; another name is 
araar kaba, the adjective in this case being derived fi^m ur, fire. A funeral feast in Mer 
gains in reputation in proportion to the number of sopaop kaba provided (cf Vol. vi. pp. 127, 
135, 138, 159X 


d by Google 


The same practice of wrapping is adopted in the western islands. I^ndtman says that 
in Badu banana leaves are wrapped round the bunches of green fruit " to keep him nice " (as 
the natives say) and as a protection from birds. 

Mr Bruce informs me that the horticulture on Erub and Uga was perfectly similar to 
that on the Murray Islands, and that the Saibai-laig were horticulturalists and uianaged their 
gardens in very much the same manner. 

Commander C. M. Lewis saw on Erub " a spot of ground containing about eight acres, 
cleared and planted with yams, banana, and a species of [sweet] potatoes.. ..An Indian and 
his wife were industriously at work cutting down branches of trees to shelter the plants 
from the sun.. ..This plantation displayed do small skill and even taste." The natives 
would not barter away yams or fruit; "on one occasion some iron tools were offered in 
exchange for sweet potatoes, which, valuable as they esteemed them, they refused " (NatU. 
Mag. VI. 1837, p. 759). 

The following notes on horticulture in Badu have recently been sent to me by Dr G. 
Landtman. After the end of the rainy season all trees and bushes in the selected ground are 
felled and cut to pieces, " smashed up " as hia informant described it, and left to dry ; later 
they are burnt. The ground ia then dug — " stone he out, root he out, clean him good, 
make him nice." The natives claim that the following have been cultivated from time 
immemorial : coco-nuts (which are said to have been introduced from New Guinea), bananas, 
sugar-cane, yams, sweet-potatoes and tare. 

The coco-nut in its husk is placed vertically in a shallow hole so that half of it is 
exposed. The planting is done at any time of the year, " no time more better," and the 
surrounding ground is carefully kept clear of grass or bushes. 

Concerning the banana the natives say, "piccanniny tree come out mother tree, 
take off, plant Christmas time, make hole in ground seven inches deep." The old trees 
are cut down and sometimes the suckers are allowed to grow up in the same spot, 
^like father and mother them dead, and piccanniny live"; sometimes they are planted 

Old sugar-canes are burnt, and " new suckers he come up nice same place, plant 
him new place, take bead and root altogether, cannot grow good old place, must take 
away." This may take place at* any time of the year. 

At the end of the dry season shallow holes are made in the ground, in each of 
which one or two small pieces of the tuber of a yam are buried, care being taken 
that they are only juat covered over with earth, if buried deep "he be dead." In 
the rainy season " all root he go over garden now," and later on " root he grow big " 
(thick). In April, " yam, sweet-potato all come dry, start eat." At the present day 
the tubers are dug up with an iron rod about one yard (1 m.) long, pointed at one 
end; this has evidently replaced the former digging-stick. 

Apparently the method of cultivating sweet-potatoes is similar to that of the Miriam. 
A fairly long portion of the tuber is removed, and earth heaped over the root so as 
to form a small mound. " End he stand up, he grow, root [tuber] he come that end, 
sweet-potato he grow like yam, same time finish, he grow out of the ground, he come 
«ut himself, you pull him off." 

For taro a bed is made in the ground with a piece of wood, and a hole dug with 

d by Google 


a stick. The head is cut off from the tuber and inserted in the hole ; it takes about 
six months to produce a new tuber, after which the process is repeated. The head is 
sometimes kept in water for perhaps a month before being replanted, which may take 
place at any time of the year. 

Many plants have been introduced by the white man, as for instuice, maize (known 
as koun or vtit), water-melon, waitain or waitin, pumpkin, manioc (Manihot sp.), and 
papaw* (Carica Papaya); see p. 134. The last three are called by their E^lieh names. 

Maize is planted at the same time of the year as yams, one or two grains being 
put in holes and covered over. The plant grows to about six feet in height and bears 
three or four branches. " He finished [ripe], break down branches with hand." Five 
or six branches are tied together and hung up to dry over a fire. I do not know 
how it is prepared or eaten as it has been introduced since the time of the Expedition. 

Water-melons and pumpkins are preferably planted on stony ground, whereas good 
ground is generally selected for yams, etc Holes are made with a' stick in which one 
or two seeds are inserted and covered over. " He start grow, you take little earth, 
cover him over; he grow big like yam and sweet-potato, break off end of plant to 
make fruit quick, if not break plant be grow long, cannot make fruit. Pumpkin any 
time be make fruit, eat him, take seed, plant bim." These are planted at any time 
of the year. 

The root of the old manioc plant is taken from the ground and a piece cut off, 
which is either planted in the same place or in another specially prepared bed. 

The seeds of the papaw are scattered on the ground. " He grow, you take little 
bit earth, cover over." 

Mr Wilkin has informed us (v. pp. 284p — 291), that in Mabuiag, "the gardens which, 
though by no means so important as those of Mer, were once second only to the sea 
as a source of subsistence are now little more than objects of a more or less sentimental 
regard... .Sweet-potatoes, tare, yams, bananas, sugar cane, [coco-nuts], and water-melons 
are now cultivated... .There is no attempt to divide the crops — as in Murray Island — 
among the gardens, or to plant yams in one and bananas in another according to the 
nature of the soil or convenience in working... .Gardens are cleared of undergrowth by 
members of the family assisted by such friends as care to help. In such cases feasts 
are usually given.. ..If a man has too many gardens, bis friends and relations help him 
to clear and plant and a feast is made. He addresses the assembled company, thanks 
them for their assistance, and often adds a dugong or turtle to the &re.... Gardens are 
often lent on the understanding that the firstfriiits are paid to the owner." Possibly 
the reason for the customs with regard to land being less strict in Mabuiag than in 
Mer " lies in the comparative unimportance of gardens (compared with canoes) as a means 
of subsistence among a people so much addicted to fighting, fishing and trading as the 

During the period of prosperity of the pearl-shelling and b§che-de-mer industries, 
the Western Islanders found it more profitable to fish for the foreigner than to till 
their gardens; consequently the gardens were largely neglected, and the natives bought 

1 This is aoiverMll;, bat enoDeoaalj, o&Uecl "m«rom7 apple" in Tonea Straits; the tFoe mammM apple 
<Uaintaea ameticana) ia a ver; different fniU-tree. 

d by Google 


for their own consumption when at home the flour, rice and biscuits to which they 
were accustomed when on the boats or attached to a fishing station. The comparatively 
good pay which they obtained enabled them to live well, but when these industries 
especially the former, began to decline, the natives were often hard pressed for food, 
and since the practical abandonment of the fisheries much distress and indeed starvation 
has occurred, aa the art of gardening has been neglected for so long a time. 

The remaining western islands were in much the same condition as Mabuiag or 
Badu, but in most of these there was very little cultivation, in some cases this was 
partly owing to the rocky soil, while other islands were merely sandbanks. MacgiUivray 
(ii. p. 36), says that on Nagir, "where, on the slope of a hill in good soil, we found 
many patches of rude cultivation. The chief plant is a broad-leaved species of yam, 
trained upon tall poles kept in position by cross bamboos, forming a firamework divided 
into little squares, each of which contains a plant. A species of Oaliadium with an 
esculent root is also much cultivated; it is planted in regular rows with the earth 
heaped up in ridges." He also states (ii. pp. 25, 26), that "at the Prince of Wales 
Islfuids the cleared spots are few in number, and of small extent, — nor does the latter 
group naturally produce either the cocoa-nut or bamboo, or is the culture of the banana 
attempted. On the main land [N. Queensland] again I never saw the slightest attempt 
at gardening. The principal yam, or that known by the names of keitai and kOM, 
is the most important article of vegetable food, as it lasts nearly throughout the diy 
season." The process of making a garden is simihir to that already described, " cut 
pieces of yam are planted at irregular intervals, each with a small pole for the plant to 
climb up. These operations are completed just before the commencement of the wet season, 
or in the month of October." Forty years later the Eauralaig had practically no gardens. 

Tobacco (p. 142) was cultivated to a small extent in a few islands, and in Mer 
and probably elsewhere charms were employed to make it grow quickly (VI. p. 20V). 
Macgillivray noticed on Nagir "some small plots of ground prepared with more than 
usual care for the growth of what Gi'om told me was an herb used as tobacco; the 
young plants were protected from the sun with pieces of matting" (ii. p. 36). 

Mr J. H. Maiden in his "Notes on some Indigenous Sago and Tobacoo from N'ew Ouinea" 
{Proe. Linn. Soc. N.S.W. for 1887, ii. {2nd ser.) 1888, p. 457) says, "The species of the 
genus Nicotiana are all indigenous in America, except our suavwlms, which is to be found 
all over Australia." He refers to specimens collected by Mr T. Bevan at the village of Tumli, 
fifty miles north of Cape Blackwood, Papuan Gulf, where it is plentiful. "It is wrapped in 
a portion of a apathe of a sago palm, is eun-cured, and was prepared for local use or tribal 
barter by natives who, in all human probability, had never seen a white man. It consists 
of the leaves and petioles but of no other portions of the plant." Mr Hugh Disaon reports 
it of "the same species aa the tobacco of commerce, if it has been at all crossed by an 
indigenous species it is to an imperceptible extent., .it is essentially a cigar tobacoo in contra^ 
distinction to a manufacturing tobacco, having a very decided cigar tobacco flavour; the strength 
of the flavour is remarkable." Mr Maiden adds, "The presence of a longbh petiole at once 
excludes this tobacco from N. Tahactim, and of all the species described by Asa Qray it certainly 
comes nearest to N, ru*tica. It is not very remote (I apeak of the foliage alone) from our 
N. maveolena with its spathulato leaves" (p. 463). 

d by Google 


Practices of a magico-religioua character were universally employed to ensure the 
fertility of crops and the productivity of fruit-trees, and even now they are not entirely 
discontinued. For the ceremonies and practices connected with gardening the reader is 
referred to Vol. T. pp. 345 — 347, and Vol. VI. pp. 207 — 213, and for those connected 
with fruit-trees to Vol. v. pp. 347—349, and Vol. vr. pp. 202—207 ; but it is evident 
there were many others about which we were not informed, and unfortunately several 
of the examples are extremely imperfectly recorded. 

There is only one folk -tale that deals with horticulture, and I have termed it "The 
sad end of Yawar the gardener" (v. p. 36). The tales of Sida account merely for the 
abundance or absence of various food plants on certain islands (v. pp. 28, 31, 35 ; 
VI. p. 19). Instruction in gardening was given to lads during the period of initiation 
(VI. p. 310). 

The ownership, loaning and inheritance of land and crops are dealt with in Vol. V. 
pp. 284—291. and Vol. vi. pp. 164—168. 

Methods of obtaininq Water. 

Shallow wells are to be found in some islands, but we have no means of knowing 
whether they are aboriginal. Mr J. Bruce informed me that in former days the Miriam 
had no wells. Behind the village of Las, in Mer, there are two shallow sur&ce holes, 
which are always dry except in the rainy season when the ground is saturated with 
water, they then hold a little brackish water for a very short time ; a mythical origin 
was attributed to even these unsatisfactory water-holes (vi. pp. 7, 283). 

There are very few springs of fresh water even in the most favoured islands. A 
famous one in Mabui^ was caused by the hero Kwoiam driving his spear into the 
rock (v. 82). The rains of the north-west monsoon cause streams which rapidly flow 
away to the sea. In Mer the pot-holes in the beds of these temporary streams retain 
water for some time; but water is scarce on all the islands except during the rains, 
and in many islands there is none to be obtained, except by digging holes in the 
sand into which a little brackish water may percolate. This for example is the case 
in Tutu, and the natives when they are living there have to import fresh water from 
Yam, a distance of over twelve miles, as the brackish water obtained from the one 
or two water-holes in the centre of the island is fit only for cooking and washing 
purposes. Jukes describes the water-holes in Damut as "large irregular excavations in 
the sand, fiilly ten feet deep, and near the middle of the island. At the bottom of each 
excavation was a little hole containing a few inches of fresh water, carefully covered from 
the sun by sticks and lumps of wood " (i. p. 162). 

Bain-water is collected in the valves of the giant clam (pi. III. fig. 3). A wisp of 
leaves or bark is often tied to the moat convex portion of the trunk of a curved coco- 
nut palm or round the stem of a pandanua. The rain-water which trickles down the 
trunk is thus drawn into a stream which is caught in a large clam shell (at the preseat 
time a bucket is often used) supported on a framework. A similar device is employed 
in many other places in Oceania and elsewhere. Dumont D'Urville (Voy. au Pol. Stid. 
IS. 1846, p. 235) says of the Tutu Islanders. "To collect fresh water they place large 
[clam] shells under the broad depending leaver of pandanus trees"; these he figures in 
AUas pUt. pi. 188. 

d by Google 


Hunting and Snaring. 

There was no true hunting in the islands, owing to the absence of land mammals. 
The pig, burum (W.), borom (E.), has been introduced, and has run wild in one or 
two islands, where it is hunted every now and again. I do not know whether this 
pig is the New Guinea hog. Sua papaensis, or the more destructive, recently introduced 
European hog. The former, I was told, does not root up gardens like the latter. The 
arrow, named sUkdri or auJcuri (fig. 188), the head of which is made of a narrow split 
bamboo, was used for shooting wild pigs, and it is still used for that purpose in Daudai. 
At the present time the natives employ guns in hunting wild pigs, and it is the only 
esciting amusement left to the Murray Islanders. 

Birds were formerly shot with bows and arrows, but now they are mostly shot with 
guns. The archers carefully concealed themselves behind trees or rocks, and often made 
a booth of branches and leaves within which to hide themselves (v. p. 38 ; VI. p. 23). 

Mr Bruce says that the Miriam shot small birds when perched on trees with a 
small bow, efewr aarik, and arrow, ebur gdub^; there were two kinds of arrow: the 
sakeiaaJ^ gelub was pointed at the end like an ordinary arrow, while the teter gdvb 
(fig. 187) was provided with two to four bamboo prongs like the ieter baw (p. 157), 

The tble, a small bird like a snipe, was shot with the teter gelub. A small screen 
of coco-nut leaves was erected on the beach above high-water mark, behind which the 
archer hid himself. The idle ran along the water's edge, and accomplices drove the 
bird opposite the screen, mud. The tble is also killed by men with bamboos, nagjuig, 
within the fish-weirs (pi. XXII. fig. 1). About sundown the men lie down behind a stone 
fence at low water ; the birds feed inside the fences and towards sunset they get on 
the wing to go to their roosting grounds, when they are knocked down while flying low. 

With the exception of the Torres Straits pigeon, Carpophaga luctuosa, gcanau (W.), 
dawner or gaino (E.), there are very few birds of any size in the Torres Straits 
islands. These handsome white and black pigeons migrate &om Daudai at the close 
of the south-east monsoon, about the beginning of November, to breed in certain of 
the western islands of the Straits and in North Queensland. As the pigeons feed 
exclusively on nutmegs, or more correctly on mace, they only stay in those localities, 
such as the island of Bauan, whero the wild nutmeg-tree abounds. The birds rotum 
to New Guinea as soon as the north-west monsoon breaks. Jukes (i. p. 157) says, "In 

' The gelub a erroneoasly described as a bamboo apear in Vol. ti. p. S3, ftod in the Tooabalai? (ni. p. 143). 

d by Google 


September [the pigeons] were comiDg thickly from the northward to Endeavour Strait, 
and they seem to return in March. March all the pigeons left in the islands were 
young ones." Gill sajB {lAfe in the Southern Isles, p. 208), "The natives of Tut go 
out to sea and kill numbers of these birds with aticks and stones, while flying over 
the Straits on their way to Australia. Even birds leara from experience, for of late 
years these pigeons, having become wary, avoid crossing that island." Dr Seligmann 
was informed that this bird was sometimes domesticated in Mabuiag. In 1888 I obtained 
in Masig a large coral (Maeandrina) figure of this bird (pL XII. fig. 6), which probably 
was employed in sympathetic magic I also saw a somewhat similar one in Mer. 

Wild duck are plentiful in the marshes of Saibai, and various shore birds occur 
on the less frequented beaches and sand-banks. The pelican, fish et^le, and other large 
birds were occasionally shot for the sake of their long feathers, but of all indigenous 
birds the white form of the reef heron, Demiegretta sacra, karbai (W.), sir (R), was 
the most valued for its plumage, as the brilliant white feathers were used in making 
the characteristic head-dress (p. 37), as well as for other purposes. It is caught in 
Mer by a line and bait at low water, when the bird is walking about picking up its 
feed. One end of a line, pom lager, is tied to the middle of a splinter of bamboo 
about 75 mm. (3 in.) long and sharpened at both ends, and the other is fastened to 
a stone which serves as an anchor, par. The splinter ia inserted in the mouth and 
passed lengthwise through the body of a small fish, tup, which is .laid down on the 
birds' feeding ground. The bird swallows the bait and is caught by the skewer sticking 
in its throat. When the splinter or skewer is made of bamboo it is called kus kek, but 
if made from an arrow it is sarik kep. 

The frigate bird or man-o'-war hawk, Fregata minor, womer (W.), waumer, waomer 
or omer (E.), which comes to Torres Straits in heavy weather during the north-west 
monsoon, is either brought down by throwing a stone which hits it on the breast, or 
caught in Mer by a bait. A tup is &stened on to a piece of wood to which a string 
is attached ; the fowler carries a supple bamboo about l'S5 m. (6 ft.) in length and walks 
into the sea till only his head is above water, dr^gii^ the bait after him. He lets 
the bait float away to a convenient distance, and when the bird pounces at the bait 
the man strikes it with his bamboo and cripples it, for being heavy it cannot recover 
itself quickly enough to rise before being struck. The man throws the bird ashore, and 
is ready for another bird. The fowlers are kept pretty busy as the birds are plentiful 
at these times and continually swoop down on the bait. 

A method of catching pelicans, awai (W.), according to a folk-tale ("Amipuru," 
V. 99) was to hold a bunch of leaves in front of the head when swimming to where 
the pelicans were floating on the sea. The pelicans, not knowing that a man's head 
was hidden by the leaves, permitted the man to approach sufficiently close for him to 
seize one by the legs. 

Dr Walter E. Roth (UTorth Queensland Ethnography, BvUetin No. 3, 1901, pp. 26, 27) 
states that the natives of the BouUa District sometimes catch pelicans with their hands when 
in ambush under oover of overhanging folif^e. Ducks are noosed of speared by the same natives 
standing or swimming in the water, their beads being covered with leaves, sometimes "holding 
with one hand a bunch of leafy switches in front." 

H. Vol. IV. 20 

d by Google 


The tern. Sterna Bergii, sera (W.), airar or aerar (E.), is caught by the Miriam 
by meane of a snara^, pom (K) (fig. 167). A atake about half an inch (12 cm.) in diameter 

or a piece of bamboo is firmly inserted _„-_- 

in a vertical poeition on the coral surbce 

of the fringing reef; a deep notch is 

cut into the upper end. A nooee is 

made &om a strip of the outer skin 

of the green branch of a coco-nut palm, 

bei lid; the strip is light and tough 

and about 1*85 m. (6 ft.) in lei^h. 

One end is bent uid tied to form a 

small loop, the other end is passed 

through this, a fair-sized running loop 

being thus formed, bei gogob ; this is 

placed in a horizontal position in the 

split or notch at the end of the vertical ^'°- '"■ «'*" '" '^^^"^ '^"'' ""■ 

stick, pom, care being taken to put the small loop in the split so that it will be held 

sufficiently tight to prevent the movements of the water from making the larger loop 

run open or close too much. A stone is tied to the other end of the strip, bei lid, 

to serve as an anchor, par, and is so placed that the strip has a good slant. A small 

piece of wood is fastened horizontally to the stick, about 305 mm. (1 ft.) from its upper 

end, and on to this a tup is so fastened that it lies directly below the centre of the 

large loop (gogob). The whole apparatus is erected under water. When the tern sees 

the tup, it dives right through the centre of the loop to it, and when it rises with 

the bait it takes the loop with it, pulling it out of the split stick. The weight of the 

stone causes the loop to run, and the bird is snared. 

During the breeding season natives went to the sand-banks to collect terns' eggs 
(VI. p. 219). 

Birds are also caught by smearing branches of trees bearing fruit with the sap of 
the enav tree (Mimusops Browniana) which acts as bird-lime; the birds perch on these 
branches, and when they try to fly topple over and hang by their feet, A large tree, 
«raperap, has small fruit shaped like seed-pods about 26 mm. (1 in.) long and filled with 
a, glutinous substance which exudes irova them. Birds that feed on the fruit get their 
feathers so smeared with this substance that they cannot fly, and are picked off the 
trees. Birds are also caught in the same manner by the small berries of the warup- 
tonrup, which are full of a gummy substance. 


Shell-fish and fish are obtainable all the year round. Collecting the former for food 
is exclusively the work of women and children, but diving for crayfish, pearl-shells, the 
large conus, or other shells of value is performed by men, though some women are also 

' The Wefltsm word Iot ft sntre ia niu, but I do Dot know vbethet it was (he nme fts eithei of the Eaatem 
pom; niui-amai is "to entangle, oatch." 

d by Google 



excellent diven. Fishing is practised by both sexes. Fish are either killed with a plain 
pointed spear, often merely a stick sharpened at the end, or by a many-pronged fish 
spear, or are taken with the hook and line' (fig. 168). A humorons 
story of a fisherman is given in Vol. V. p. 104. 

Line-fishing was practised from canoes (v. 18), off rocks (v. 89), 
or standing in the water on the reef at high tide (fig. 171). The 
fishing lines are deacribed elsewhere (pp. 89, 90). The bait was tied 
on to the large, barbless, turtle-shell hooks (fig. 168 ; pi. XI. fig. 1). 
Hooks vary in length irom about 38 to 100 mm. (1^ to 4 in.). 
The recurved portion of the hook varies in our specimens from 
50 to 75 mm. (2 to 3 in.) in length, and the loop has an average 
breadth of 38 mm. (1^ in.) ; the breadth of the flat hooks averages 
9 mm. (|ths of an inch). This will give some idea as to how 
clumsy the native book is. It is never used now, the natives 
employing European hooks when they can get them ; failing these 
they make neat barblesa hooka out of wire or anything else that 
will suit. Those wbicb.we collected are tied two on to one line made 
of coco-Dut fibre, with a small piece of thin twine fastened on to 
each hook for the purpose of tying on the bait. Fig. 169 illustrates a Miriam sinker, 
mekek par, which fivm its name was used with a fisbing-line, but may have been used 
for snaring the white heron (p. 153); it measures 100x94x65 mm. and is made of 
volcanic ash. 

In the Murray Islands at various times of the year large shoals of small fish, tup, 
resembling sardines or sprats come very close to the shore, usually forming dense masses 
which look like a dark shadow in the water; fi«quently they are 
chased by other fish. The Miriam catch them in this wise : two 
men each provide themselves with a long light pole or piece of 
thin bamboo about 1-5—2 m. (5 or 6 feet) in length, the head of 
which is wrapped round with string, etc., so as to form a &ir-sized 
knob, this stick is called wenr (fig. 170 a). Another man takes 
a weret or conical basket or scoop (fig. 170b; pis. VII. fig. 6, 
XXIV, fig. I). These three men walk along the beach on the look- 
out for tup, followed by several children with baskets. When a shoal 
of tup is seen in a suitable spot, the two men cautiously advance 
and fiing themselves into the sea, holding the werir with extended arms ; they so manage 
this movement that the shoal lies between them, and the splashing of the werir hinders 
the fish from fleeing sea-ward, especially as the werir are so held that their knobbed ends 
nearly meet in the water. In an instant the fish are huddled together within the triangle 
formed by the men and the poles, and this critical moment is seized by the third 
man, who dives into the sea and scoops up as many fish as he can. The fish are 
then emptied into the baskets, which the children cany with them into the water. 

There are three kinds of tup, called koa, areare and merdud. They are in full roe 
r tvdi (W.), kek (E,)[ flghing-Iine, ariga (W.), ariag, mektk gtm (B.); hook and line, tiuktk 

Tia. 169. Sinker for 
Bsltiiig-liiie, Met. 

I Hook, tvd a 
M (E.). 


d by Google 


in October and November. They seem to have no particular season as they are found 
all the year round. Sometimes they are scarce on the south-east side of Mer, but 
plentiful on the north-west side, or at Dauar. 

ifup fishing is referred to in the story of Nageg and Qeigi (vi. 16, 17), and a painted 
top illustrates two werir and a iveres (vi. pi. VI. fig. 3). 

Fio. 170. Wtrir uid verti, Her. With detaile of IwhiogB. 

The weres made for us (fig. 170b) is of small size (length 79cm.), and is constructed 
from a piece of bamboo which is split below a node into eight strips. The strips or 
bare are kept apart by seven cane rings of which the lowermost is 22 cm. in diameter. 
Eight to ten midribs of palm-leaves run longitudinally in the spaces between these bars 
and like them are securely lashed to the rings, the thicker ends of the midribs being 
at the larger end so that they naturally diminish with the decreasing diameter of the 
weres. A lashing binds the top of the cone. One specimen measures 122 m. (48 in.) in 
length and 345 mm. (13 J in.) in diameter ; it is made in the same manner as the above. 
Weres are made by men. 

The two werir are narrow bamboos about 183 cm. (6 ft.) long. At the head end strips 
of the skin of the stem of a banana are bound crosswise to form a bulb, probably around 
a central wad. 

Several kinds of fish-spears are made, which are either a simple pointed stick or 
a pronged instrument. The best speara in Mer are made fix>m a hard wood, tol, the 
Western tulu (Polanisia viscosa), very like rosewood, which is traded to the Miriam by 
Western Islanders, principally the Waraber le. 

The largest of the simple spears, fx^er (the Miriam kus bager is made from kus 
wood, VI. pp. 24, 40), has no prongs and is used for spearing large fish or turtle. The 
Miriam kolap pespes and dab are similar spears used for the same purposes. The burar 
nudil (E.) has a piece of half-inch iron, about 61 cm. (2 ft.) long, fiited into a bamboo 
(burar) shaft (pi. XXI. fig. 3). The Western rod or rada is also a long straight stick 
with a sharp point. The takai (W.) is a pointed stick about 76 cm. (30 in.) long which 
is used (mainly by women) for catching fish. A small sharply pointed stick, pat (W.), 
was employed by women for catching the octopus (v. p. 23). 

Baur is the general name of the Miriam for a fish-spear with several prongs. 
The walek baur has a bamboo shaft and three prongs, about 61 cm. (2 ft.) long, made 

d by Google 


of the hard tol wood ; it is used for large fish. The tet«r baur is made of bamboo, one 

end of which is split and whittled to make four prongs; this end is called teter, feet; 
the au t^er baur is now usually provided with four iron prongs about 23 cm. (9 in.) long. 

The Aw6e teter baur vaa made of small bamboo with four bamboo prongs, 16 cm. (6 io.) long, 

fixed on to the shaft, which is now fitted with several iron wires; it was used for 
spearing tup. 

Fio. 171. Dnwing by Pasar of Habniag of a man flihisg with hook and liue for a tcad flsh in a bole 
in tbe ooial rock, and of another man spearing a poaifi fish. 

The Western taku or takul (fig. 171) had several prongs with a shaft made of iaer 
wood. This spear is now made by lashing several wires to the end of a long shaft 
so that they diverge slightly from one another; formerly splints of wood were employed. 
The dagulal (W.) is a fish-spear of bamboo with several points. The barugut (W.) had 
two prongs, bat possibly this may have been a two-pronged arrow. The tul (W.) had 
a shaft of tvlu wood and its prongs consisted of the spines of the sting-ray (the auai (W.) 
was a similar spear of smaller size that was employed by the maidelaig, or sorcerer, for 
malevolent magic). 

Spearing fish is the most common method of catching them; it is employed either 
while walking along the shore or on the reef at low tide, or from caooes. Speaking of 
the Murray Islanders, Brockett (1836, p. 23) says, "Their manner of fishing is much 
the same as that practised by the natives of New South Wales. One party is engaged 

d by Google 


in driving the fiah towards the shore, while another walks on the beach ; and, as soon as 
the fish are near enough to the shore, the latter party dart into the water, and bring 
ihetn up on the end of a spear." 

Women go on the fringing reef at low water with spears to fish for all kiada of 
small fish and collect shell -fish. 

At Jiaiger time (November) and on calm days in the north-west monsoon, Miriam 
women go to the detached ree& near the island in a piece of a broken canoe, pau, which 
has a piece of wood nailed across each end to keep out the sea. They wade about on 
the top of the reefe spearing garom and other small fishes. 

The right method of spearing saw-fish, according to a folk-tale (vi. p. 55), is to spear 
the nearest one ; should the one in the middle be speared, when the man dived to catch 
it the other fish would cut him up with their saws and kill him. 

Toreh-light fishing, ne igi (E.), is carried on nearly throughout the year when the tide 
is suitable. When there are tow tides at night men and women go on the fringing 
reef with small torches searching the lagoons and small pools and turning over stones. 
One person generally holds the torch, while the other uses the spear. They get small 
fish, octopus, crayfish and crabs in this way. Sometimes they go out at half-tides. 

In Mer during the south-east monsoon the lowest tides, miskep, are in the day-time, 
but after November and during the north-west monsoon the night tides are the lowest; 
these are termed tugei^. 

In the months of February and Uarch great shoals of gar-fish, zub (Hemirhamphus 
&r), come into the deep water passages between the ree& round the Murray Islands 
and are speared at night by torch light. In the day-time, when the season sets in, 
women and children collect dry coco-nut palm leaves, bet, tie them into bundles in the 
form of a large torch to hold in the hand, and place them beside the canoes on the 
beach. When it gets dark the canoes are launched and the torches piled in them. A 
certain number of men in a crew are told off for paddling, others as torch-bearers and 
the remainder as spear-men ; for this purpose the av tOer baur is employed. When they 
get out to the passages the torches are lighted and the work of spearing begins. The 
water is Uterally alive with gar-fish, and the spear-men have only to keep dabbing at 
the sur&ce of the water to catoh them. They get large catohes of these fish in the 
season. When they come ashore the fish are divided among those who have helped, the 
canoe-owner getting his share. At the present time, owing to the scarcity of canoes, they 
go in dinghys, but they do not get so many fish as formerly, as there are fewer 

On the eastern aspect of the large fringing ree& of a few of the western islands 
(e.g. Mabuif^), but notably in the eastern islands, long low walls of unworked boulders 
of stone were built, or rather piled up, to a height of three or four feet. These walls 
enclosed large irregular areas of the reef, graz (W.), sat (E.) (pi. XXII. fig. 1), During 
the north-west monsoon, when the sea is calm on the lee-side of the island, large numbers 
of fish swim over the walls at high tide, they are prevented by the low night tide (tugei) 

1 This word (and Eome ot tha information) ha« reoentl; been sent to ine bj Ur Broce, it eridentl; li oonneoted 
with the dogai referred to in Vol. vi. p. 271, bud h probabl; the Mme word. Perbkpa aome future iixreBtigator 
will olear op the matter. 

d by Google 


from escaping aad so fall an easy prey. According to Mr Bruce, no native within the 
memory of man ever made a sat*, all they can do now is to repair them, and the 
natives argue that if Abob and Kos (the reputed inventors and builders) had not been 
endued with power exceeding that of mere mortals, the fish-weirs would never have been 
built. The walls are partly made of blocks of lava, which the natives say Abob and Kos 
brought down from the bush, as there are no stones of this kind on the reef, only 
lumps of coral-rock (cf. vi. 26 — 28, 218). During the south-east season, as they are then 
on the weather side, the walls are broken down by the seas rolling in on them and 
thus no fish can then be taken. 

No fish-traps with pockets or labyrinths are built of stone, piles or screens. 

Fishing nets or landing nets were unknown. 

The practice of shooting fish with a bow and arrow is widely distributed in New 
Guinea, and occurs along the Papuan Gulf. In the legend of Sesere of Bada (v. 40) 
this method of fishing is mentioned, he was in the habit of shooting fish in the pools 
on the reef at low tide ; also the boy Geigi of Waier (Tl. 15) shot fish. These are the 
only records for Torres Straits. 

Farts of certain plants are scraped and pounded or bruised and placed in pools or 
lagoons on the reef at low water to stupify fish, which are easily caught as they float 
on the surface. In Mer the process is called kublam and sad: kuh is a fruit which 
is crushed and thrown into the water which turns red; sad (E.), eozi or saei (W.) is 
a vine, Derris uliginosa Bentb., that turns the water a milky white. "The leaves [of 
the scandent shrub Derris] are pounded and thrown into water for the purpose of 
stupefying fish by the natives of many tropical countries " (Maiden, Useful Native Plants, 
pp. 168, 416). Ib(dm (W.) was described as a plant " like milk inside " which will 
kill fish and eeb; itamar (W.), Indigofera austratis, was employed for a similar purpose. 

For magico-religiouB practices connected with fishing, see Vols. V. pp. 342 — 5 ; 
VI. pp. 217, 218. 

Tdbtlb Fishing. 

There are two periods for turtle fishing, the one during October and November, 
which is the pairing season^ when turtle are easily speared owing to their floating on 
the sur&ce of the water ; the dugong harpoon is generally used for this purpose, or 
they are captured by hand. The other turtle season extends throughout the remaining 
months of the year, during which time the turtle firequent the deeper water and the 
channels between the reefs. It is then that the sucker-fish is utilised. 

The following species of turtle occur in Torres Straite: — 

The green turtle, Chelone mydas-C. viridis, tcoru (W.), nam or nam kar* (£.), which 

> Captain Leirii howsTer ststet; "Dnritig; Iha fiBb leoaon, the iDdians live priDaJpall; on fish. For the 

pDrpoM of taking them, exteniive itone weira are made on the soath-eaat side ol the iiland [Enib]. The; were seen 

lYifiVing oD« of Tei; large dimensions. This ia also practised by the Murray Islanders." Want. Mag. ti. 1837, p. 760. 

* W. W. Oill eaye; "We happened to be in the Straits in the pairing seaBon....Not a da; paB»ed without 
oar teeing single turtle asleep, idl; drifting with the ourrenL At other times pairs floated past, amoronsl; 
eUtUring the edges of their abellB against each other" (I.e. p. 391). 

* In Her a tall-growD turtle is nam ; robi kar means true turtle, one nearly fall grown is korkor, whUe a small or 
medinm-tized one is mergai ; the term lirvar is employed when the word nam is tabooed (p. 106), it refers to a tniUe 
of either sex. 

d by Google 


often weighs 360 pounds. This is the onl7 species which is at all common at the Murray 

The hawksbiU turtle, Chelone (Caretta) imbricata, unoa, unatoa, unouxi, unutm, vxmawa (W.), 
kttur, baug (E.). This is the species that produces most of the tortoise-shell' of commerce, as 
much as four pounds of which may be obtained from a full-grown specimen which sometimes 
weighs 260 pounds. According to Macgillivray it "resorts to the shores in the aeighbonrhood 
of Cape York later in the season than the green species, and is comparatively scarce. It is 
only taken at night when depositing it« eggs in the saud, as the sharpness of the margin of 
its shell renders it dangerous to attempt to turn it in the water" (ii. p. 23). It is scarce at 
the Murray Islands. 

The loggerhead turtle, Thalassocbelys (Oaouana) caretta, urea (W.), tirit (E.). This species 
grows to nearly the same size as the hawksbill, and is prized as food by the Western Islanders, 
but not so by the Miriam who do not care about eating it. 

Mr Bruce also refers to a kind of turtle named maiu which has a head like baug, but 
the shell is thin like that of the green turtle and has no market value. As a rule the Miriam 
did not eat it, but the Western Islanders did. 

The mud turtle. I do not know this foriu, but Qill describes it thus: "This species is 
particularly plentiful at Tauan [Dauan], and all along the muddy south-weetem coast of New 
Guinea. ..[it] never weighs more than fifty or sixty pounds, and generally much less" (I.e. pp. 291, 
292). It may be a Sphargis or Oarettochelys insculpto. 

The importance of the breedii^ season of the green turtle is so great that in tbe 
west (and perhaps in the east) the name for the turtle at this period ia applied also to the 
season — surlal or surlangi (in Muralag, according to Macgillivray, it is sulangi, from mlur). 

Macgillivray says, tiiat this season "at Cape York usually extends from about the middle 
of October until the end of November, but the Umite are not constant. During the season they 
are to be seen floating about on the surface of the water, often in pairs, male and female 
together. A few are caught at night on the sandy beaches, but the greater number are captured 
in the water. The canoes engaged in turtling, besides going about in the day, are often sent 
out on calm moonlight nighte. When a turtle is perceived, it b af^rooched from behind as 
noiselessly as possible, — when within reach, a man in the bow carrying the end of a small 
rope jumps out, and getting upon the animal's back, with a hand on each shoulder, generally 
contrives to turn it before it has got far and secure it with the rope. This operation requires 
considerable strength and courage, in addition to the remarkable dexterity in diving and 
swimming possessed by all the blacks of the north-east coast and Torres Straits." "Even the 
green turtle, with a comparatively rounded margin to the carapace, occasionally, in sti-u^ling 
to escape, inflicte deep cuts on the inner side of the leg of ite captor, of which I myself have 
seen an instance." "There are some favourito look-out stations for turtle where the tide runs 
strongly off a high rocky point. At many such places, distinguished by lai^ cairns of stones, 
bones of turtle, dugongs, Ac., watoh is kept during the season, and, when a turtle is perceived 
drifting past with the tide, the canoe is manned and sent in chase" (ii. pp. 21 — 23). 

I was informed at Mabuiag that during surlal there were two ways of catching 

■ TortoiBe-shell, or as it Bhould be called tartle-shell, ia used for making maaki, fieh-hoolu, onwments, ale. 
The Western name ii kdrar or krar uid the Eastern ia kaita, bat the Miriam have a Main word for it, baagem 
(nL p. 61) and the iogo tui or aaored name ia otai. Recently Hr Brace has informed me that this tattle is 
called baug and the Bhell keiiur or Icarar, 

d by Google 



1. The canoe warily approached the turtle when they were floating on the top 
of the water, whilst they were " fast " (copulating). One was speared by the dugong 
harpoon, and whea it swam below the surtace a man holding a rope dived after it and 
tied the rope round a fore flapper. The turtle was thus fastened by two ropes, the 
other being the am attached to the harpoon dart, and thus it could be hauled cloee 
to the canoe, when the rest of the crew would jump into the water and roll the turtle 
into the canoe. 

2. Another method was for a man to tie a rope round his arm near the shoulder, 
and when he came close to the unsuspecting turtle he would leap on to it and catch 
hold of the carapace back and front with his 
hands. The man, still holding the turtle, would 
be dragged to the canoe (flg. 172). 

Gill says, " Turtle are very quick-sighted, 
diving to the bottom at the first intimation of 
danger.... As soon as a sleeping turtle is seen 
they stealthily paddle close to it, when one of 
their number, with a rope tied under his armpits, 
leapa upon the back of the unconscious victim. 
Of course the man goes down to the bottom with 
the reptile, but immediately twists the fore- 
flappers over the back, and holds fast by them. 
The man and the turtle are then hauled up into 
the canoe" {l.c. p. 292). 

Turtle are also caught by being turned over 
on their backs when they come on to the sand 
beach to lay their eggs. Evidently this is not 
considered a "sporting" method as no tallies are 
kept, at all events in Mer, of turtle so caught, 
whereas a kupe (fig. 224) may be made for those 
speared or caught in deep water. 

Very shortly after mating the female turtle 
lay their eggs in sandy beaches, usually selecting for this purpose an uninhabited or 
sparsely inhabited island. There are in Torres Straits numerous low islets without a 
permanent population, as well as sand-banks raised just above the sur&ce of the water, 
and mainly about the end of October and during November the natives make expeditions 
to them to hunt for turtles' eggs. As is well known, the female turtle scuttles in the 
night time up a sand beach, makes a large hole in the sand in which she lays her 
eggs, and then smooths over the surface for a considerable space round the nest and 
finally betakes hereelf to the sea by a different route from that taken on leaving the 
water. The way to find the nest is to take a bearing along each track of the turtle, 
and to go to the spot where they converge. A pointed stick is procured and this is 
thrust into the sand, when it comes up clean a blank is drawn, but should an egg 
be pierced the glair of the egg causes sand to adhere to the stick (cf. Vol, v. p. 19). 
The overlying sand is speedily removed by the hands, and the mass of nearly spherical 

H. Vol. IV. 21 

Fio. 173. Drawing b; Wuia ot Uabniag of 
a man eatobing b tnrUe. 

d by Google 


eggs with their parchment-like shells is exposed to view. A nest may contain from about 
one hundred to two hundred eggs. The eggs are carried away in an ordinaiy basket, or 
in one extemporized out of neighbouring grass or leaves. 

The most interesting method of catching turtle is that in which the sucker-fish is 
employed. The sucker-fish, often wrongly termed " aucking-fish " (Echeneis naucrates*), 
gapu (W.), glp (K), is a member of the Scombridse or "Mackerel" &mily, the dorsal 
fin of which is transformed into a sucking organ. By means of this disk they are enabled 
to attach themselves to any flat surj^e, the adhesion is so strong that the fish can only 
be dislodged with difficulty, unless it is pushed forward by a sliding motion. The 
"suckers," as they are sometimes called, attach themselves to sharks, turtles, ships, or 
other object. They cannot be regarded as parasites, as they obtain their food independently 
of their host, but beir^ bad swimmers they allow themselves to be carried about by other 

Macgillivray states that Qi'om informed him that the natives of Muralug catch a 
small form of turtle, which he never saw, in the following manner: "A live sucking 
fish (Echeneis Remora) [the only sucker-fish I saw was E. naucrates], having previously 
been secured by a line passed round the tail, is thrown into the water in certain places 
known to be suitable for the purpose; the fish while swimming about makes &st by its 
sucker to any turtle of this small kind which it may chance to encounter, and both are 
hauled in together ! " (ii. p. 21). 

Qill states that "several captive sucking-fishes are kept swimming after the canoe 
until a turtle is seen, when three or four of them are thrown as near the sleeper as 
possible. These sucking-fishes at once attach themselves to the turtle. The cords are 
now cautiously hauled in.... Sucking-fishes are sometimes kept two or three days in a 
lagoon, or in a boat half-filled with sea-water, until turtle are seen" (^c. p. 292). 

The sucker-fish is not used to haul in the large green turtle ; I was repeatedly 
assured that it would be pulled off, as the turtle was too heavy; but small ones are 
caught in this manner^. 

■ The sncker.flBh from its peculiar habits and sliuotute baa titi>a;B attracted attentioa. Dr Q. Shaw in his 
Oenerai Zoology, Vol. iv. p(. 3, 1808, p. 201, qaot«s Pliny's aoooDnt of tbis fieb, and Dr A. Oiinther in the 
^nn. ami Mag, o/ Nat. Hut. Vol. v. 9rd ser. 1860, p. S86, give* a bUtor; of Echeneis and a revision of the 
species. The spedes falls into two groups : (a) with a stout and rather short bod^, {fi) with a slender body, 
the most oommon and representative species respectively being E. Bemora aod E. naucrates. The Torres Straita 
species belongs to the slender group, and the number of the lamina of the disk (22—35) d 

' Shaw etaleg (p. 309) that "The Count de Cepede informs us, from the manuscripts of Commerson, that it 
Is also very eommon about the coasts of Mozambique, where it is sometimes made use of for the following veij 
siugalar method of catching turtles. A ring ie fastened round the tail ot the fish in such a manner as to prevent 
its etoape, and a long oord fastened to the ring. When thai prepared, the fish is carried in a vessel of sea-water, 
ADd when the boatmen observe a tattle sleeping, as ie the frequent ouslom of those animals, on the enr&oe ot 
the water, tbej approach as near as possible vithoat disturbing it; and then, throwing the Remora into tho 
sea, and giving it the proper length ot cord, it soon attaches itself to the breast of the sieeping tartle, which is 
thus easily drawn to the boat by the fishermen." Dr Qflnthei (I.e. p. 396, and The Study of FUhei, p. 461) 
suggests that this "appears to have originated rather from an experiment than from regular practice." 

d by Google 


According to one of th^ folk-tales (v. p. 44), there wae a, time when the people of Badu 
did not know how to ostch turtle hy means of the sucker-finh, and they used to employ a black 
toothless "dog-fish," hanuar, when they went for turtle. The atory goes on to tell how Bia 
taught his fellow islanders how to employ the sucker-fish. In the Bomai-Malu legend of the 
Miriam (vi. p. 41) it is stated that Barat of Moa, according to the fashion of olden times, 
tied a r»pe round the tail of a kamomr, then he made a sucker-fish, and instructed the Western 
Islaoders who were with him how to catch turtle with it. I do not understand how turtle 
could be caught by a "dog-fish," but as the identity of this fish, which is said to live in 
crevices of the rock in deep water, is unknown nothing further can be said, except to hazard 
the suggeetion that it may be an unidentified kind of lamprey ; but against this it must be 
stAted that no member id the Cyclostomata is known from Queensland waters, though Mordacia 
mordax occurs in Tasmania and species of Oeotria are found in southern Australian waters. 

I was informed that in leashing a sucker-fish, a hole is made at the base of the tail-fin 
by means of a turtle-bone, and one end of a very long piece of string inserted through th« 
hole and made fast to the tail, the other end being permanently retained. A short piece of 
string is passed through the mouth and out at the gills, thus securing the bead end. By 
means of these two strings the fish is retained, while slung over the aides of the canoe, in the 
water. The short piece is pulled out of the mouth of the fish when the turtle is sighted and 
the gapa is free to attach itself to the turtle. 

When starting on a trip to fish for turtle by means of the eucker-fish, the owner 
(or captain) of the canoe gives the order where to go and when to let go the anchor, 
having arrived at their destination. 

The bvai-garka (see section on Canoes) makes a fire on which he places some 
turtle-bone which the owner has brought with him. When the bone is charred the 
buai-garka breaks it up and throws it into the water so as to attract the sucker-fish. 
When one is caught it is the duty of the buai-garka to attach to the fish the leashing 
which he bad previously made. 

The direction of affairs is now assumed by the haai-garka, who gives the word 
to move to another place, and the directions where to go. When he gives the order to 
stop, the mat sail is rolled up by the other men (or at the present time the sail is 
lowered), he not taking any part. He gives the order to paddle till he sees the turtle, 
then gives the word to stop, and the anchor is let go by the owner, having been 
previously shifted to the stem of the canoe. When the buai-garka sights a turtle 
swimming deep down in the water, be removes the mouth-string &om the sucker-fish 
and throws the fish overboard with the tall-line attached and plenty of slack is thrown 
with it, he then hauls in the superfluous slack and as &r as possible indicates the 
direction of the turtle by pressure on the line. The sucker-fish on perceiving the turtle 
immediately swima towards it, and attaches itself to the reptile's carapace. When this 
is accomplished, the buai-garka gives the order to heave up the anchor and move the 
boat up to the poBiti<m of the turtle. 

One of the crew (but not the buai-garka), with a long rope attached to the right 
npper arm, dives into the water, and ia guided to the turtle by the line fastened to 
the fish's tail. On reaching the turtle, the man gets on to its back and passes his 
arms behind and below the fore flappers and his legs in front of and below the hind 


d by Google 


flappers, or secures it in some other way. The man is then rapidly drawn up to the 
sur&ce of the water bearing the turtle with him. On the arrival of the diver the 
sucker-fish nsually shifts its position jrom the upper to the under sur&ce of the turtle. 
As soon as enough turtle have been obtained, the owner of the canoe gives the order 
to go home, and the buai-garhi resumes his subordinate functions, and resigns into the 
hands of his brother-in-law the direction of affairs which had been his part during the 
actual process of fishing^. At the end of the day's fishing the sucker-fish is eaten. This 
method of fishing is described in a folk-tale (v. 92). 

The huai-garka knows whether the fish has attached itself to a turtle or to a shark 
by the nature of the motion of the string. If the pull is intermittent it means that 
the fish has adher^ to a shark, but if it is steady, then a turtle has been secured. 

In order to understand the method of leashing a sucker-fish, I induced a native to make 
a model of a giipu for me. Fig. 173 indicates diagrammatically the arrangement. A loop is 
inserted by means of a wooden arrow point through the gills and out at the mouth, the ends 
are passed through the loop, and one of the strands is threaded through the other, the two 
are then twisted into a string. The mouth-string is called yudaa and is made of the Inner 

Fio. 17S. niiutrating the method ot leaahing a Backer-fish. 

bark of the root of the vxdi tree. The other end of the gtidaz is tied into a slip knot, kaxa 
wiaikah, the end of a long piece of twine is simply bent twice round the string at the knot; 
when the free end of the gudaa is pulled the knot runs out, and the twine (one end of which 
is still held by the fisherman) slips off the gudaz. The main fishing string is a very long and 
strong cord of twisted coco-nut fibre, igal; this is fastened to a braid or plaited string, dan\ 
the other end of which is bent round on itself so as to form a loop; the end of the dan and 
the loop are bound round with wait. The loop is furnished with two strings of mali ; it looks 
as if these were threaded through the tail of the fish above and below the vertebral column and 
tied together on the other side. Another lashing binds the cord close to the side of the narrow 
portion of the tail. 

.Vote on the mcker-Jtsh lea$h by W. I. Poeoek, One can only conjecture the rationale of all 
this. Possibly the double string, besides being easier to pass through the fish's gills, is less 
likely to tear them and thu may be the object of the threading which dbpensee with a knot. 
The strings are perhaps twisted to avoid tangling. The effect is to convert a lark's head to 
a running noose made with an eye of string, and it is not quite clear why a single string widi 
an eye could not be used. 

■ A ooDiiderable portion of this intormatioD was collected by Dr Rivera (cf. v. p 
* From its nttiae this shonlil be made of Ficoi bark, but in our specimen the braid w 

d by Google 


The twieting of the twine round the gttdaz is of course equivalent to twisting the gudea 
round the twine. Thus we have a third example in this small collection of a knot of which 
a simple twist, a turn and a half, is an essential part The efficacy of the twist depends 
principally on tiie knot being drawn, when the strands of the twist tend to nip, something 
like a double half hitch. With the gudaz and twine, if both ends of the latter are fast, the 
Dip is automatic. One would think the reef knot bow would have done 
here, as there is no great strain, though it is bad in principle to unite 
two different kinds of strings with it. Better would be a bow-line finished 
by passing a bight instead of the end. The diagram shews this knot as 
tied by a sailor who commenced, as in a sheet bend, by making an eye 
of the standing part. A gentleman shewed me a very different method 
with the same result This is not the only way the bow-line could be Pio. 174. A slip knot, 
applied to the purpose, but the sailor assured me he had seen no other. 

•Hie bow-lme is the knot used in the widely distributed art of netting, hence its absence from 
this collection, is remarkable. 

The natives have a great respect for the sucker-fish and firmly believe it to possess 
ominous powers. For example: when the fish does not take a good hold on the turtle 
and then swims ofT it indicates that some part of the canoe is not secure; when there 
is something the matter with the bow of the canoe, the fish is said to attach itself to 
the neck of the turtle, but should the stem of the canoe be weak, the fish adheres 
to the extreme hinder end of the carapace ; when it fixes itself firmly to the front 
part of the carapace, the canoe is strong; when it goes to one side of the carapace or 
keeps OD moving about, it shews that the lashings of the float to the outrigger on that 
particular side are insecure. More than oace I was told, " Qapu, savvy all same man, 
I think him half devil (t.e. spirit)." 

Qapu is a subsidiary totem in some of the western islands (v. p. 181), and at one 
time may have been a chief totem of a clan which has now become extinct. 

When a man catches a turtle with another man's gear, the turtle belongs to the 
owner of the gear, with the exception of the lower half of the turtle and the lai^ intestine, 
kuriamar or Icutamal, which go to the owner of the canoe. The actual catcher of the 
turtle receives only a small portion. 

There were many ceremonies connected with turtle fishing, more particularly during 
the aurlcd season, some of which are described in Vol. v. pp. 183, 330—336; vi. 213—216, 
and it is probable that each island had its peculiar ceremonies, though these are now 
irrecoverable. Gill says, " In Torres Straits are numerous turtle-giving gods, whose assistance 
is invoked and to whom offerings are made. These gods are merely round painted 
atones" (tc. p. 293). 

The green turtle, Surlal or Warn, was a totem in several of the western islands 
(v. pp. 164, 155, 167), and in addition the turtle-shell turtle Unawa was a totem in 

The importance of the turtle as an article of food is well seen in the taboos which 
are enforced to prevent any possible detriment to fishing being occasioned by menstruous, 
pregnant, or recently delivered women (v. pp. 196, 207). Continence was imposed during 
the surlal season (v. p. 271). This information refers to the Western Islanders only. 

d by Google 


Both the turtle and the sucker-fish are represented in decorative art, the sucker of 
the latter giving rise to a decorative motive. 

Turtle are not often referred to in the folk-tales and then but incidentally, with 
the exception of one important tale which describes the legendary origin of the nam toffo 
of Mer (VI. p. 46). 

Nam is a family name for girts of the Meaurera le and K6met U (v. p. 102), 
therefore their naubet and relatives by marriage, awim, have to call the green turtle 
siruar (vj. p, 100), and address the person named Nam as Siruar. The Heaurem le are 
the principal Nam togo le or Nam boat (vi. pp. 51, 236), and the K6met le are associated 
with them in the zogo. 

DuQONO Fishing. 

The dugong (Halicore australis) (pi. XXIII. fig. 4) is a very fovourite article of 
food, and its capture is an exciting occupation. This bulky marine mammal attains 
a length of eight or nine feet or more, and is a perfectly harmless vegetable feeder, 
its food consisting of one or two species of submarine flowering plants (Cymodocea)*. 
Although dugong occur eveiywhere in the district they are abundant only on Orman's 
Beef immediately to the north of Mabuiag, and over the unsurveyed expanse of reefs 
between Mabuiag and New Guinea. The former island may be regarded as the head- 
quarters of the fishery of this sirenian. 

Dugong were speared either &om the bow of a canoe or from a bamboo platform, 
noat, nad or n^t (W.), narat (K). The implement employed is the dugong harpoon 
or wap (pi. XXIII.). A barbed head is loosely inserted in a terminal hole at the butt 
end of the wap. It is lashed on to a long rope or running line, nearly an inch in 
thickness, and some 40 or 50 fathoms in length. The native-made rope, am (W.) (p. 00), 
is preferred for this purpose to European rope on account of its greater buoyancy. The 
other end of the rope is made fast to the canoe or to the platform (fig. 176 ; pi. XXIIL 

fig. 1). . 

When the canoe is close enough to the dugong the man bearing the harpoon jumps 
into the water, at the same time striking the dugong as it is in the act of breathing. 
The creature immediately dives down, and runs out the rope which is fastened to the 
harpoon-head which now becomes separated &om the shaft. The man has to be careful 
not to get his head entangled in the loops of rope, as deaths have occurred through 

^ Cfmodooeft Is oue of the genera of the ZoBteraoue of wbiob Zoatera nurina (the " Eel" or " Sneet Gnu" 
of ODi fiBhertnen, more oommonly known aa "Seo^wraok") is a familiar example. The ZoBteraoen are among 
the Tei7 few true flowering plants that live in the «ea. The flowers are inoonspicuons and the leaves an 
of a bright gTM«-green oolonr. The dugong appeal to sabsut on the three species (Head Hunlen, Blaclc, 
White, and Brtncn, 1901, fig. IG), bat two other species oconi in certain looalitiet in Torres Sttaite. W. Baville 
Kent (Tfu Oreal Barrier Reef of Auelralia, 1893, p. 927) states that its food "consists almost eiclimivel7 of 
PotUoida auftmlif." 

It Is tempting to suppose that in (he sarlj Tertiary Period the anaestor of these Sirsnia was a land 
animftl whioh fed on tho ancestral land Cjmodooea. Being aocastomed to this fodder, the animal followed 
the plant when it became paltistral, and as the plant became more and more aqnatic, so did the aniinal, 
until at last both became modified for a porel; marine habHat and incapable of even temporary terrestrial 

d by Google 


this accident^; I believe he usually throws himself backwards on reaching the water 
in order to avoid the rope. ' The man returns with the harpoon to the canoe. Other 
men immediately dive into the water, and when the dugong once more riaes to breathe 
they tie a second rope, galai gaipapi (p. 91), round its tail ; then, whenever it attempts 
to rise, the men by diving at the same time pull it down with the rope, frequently 
also they cling on to the animal so as to hamper its movements, and m a very short 
time the unwieldy dugong is suffocated. So iar as I know death always occurs through 
asphyxia. Owing to the thickness of the skin and blubber, and the shortness of its 
point, the harpoon head can never penetrate to a vital organ, unless it should happen 
to pierce the spinal cord, which is unlikely. At the present time the dugong is almost 
invariably speared from luggers, as these vessels are so much more convenient to handle 
than canoes. 

The platform was only employed on moonlight nights. A man would walk on the 
reef at low tide in the daytime to watch for traces of the dugong. When he found 
a patch of " dugong grass" which had been partially browsed, he would erect the staging 
there, knowing that the dugong would repair nightly to the same spot until the fodder 
was exhausted. The platform* was constructed of six bamboo poles lashed together, 
surmounted by the centre-board, walurtga, of a canoe. On this the rope was coiled, 
one end being tied to the two oblique poles where they cross each other, and the 
harpoon put in readiness, the other end of the rope having been lashed to its head 
(fig. 176; pi. XXIII. fig. 1). All night the man would perch on this board, awaiting 
the arrival of the dugong. When it approached sufficiently near it was speared from 
above. Usually a wooden or stone image of a dugong would be slung on to the 

> Death by itnnsnlation in (bi« manner wlb olwajTB acaredited to a sorcerer, ntaidelaig (W.). The 
aooompaDyiDg flgnre U reproduced Iroin VoL v. p. S89, and is taken from a mantuoript by Waiia of Maboitg. 
The writer sayi, "Him Haiak hither (at) Aladaoba strangled when (at) Aladnoba here. Abatm, Aba, ttntd, b«, 
noat, dugong-platlorm, moidadin, bnilt, lieda, thai, keda, in this manner, nuin, him, dangalan, a dogong, kato- 

Fio. 175. Drawing by Waria illnitrating the death at Haiak by itnuigalation when dagong-flahing. 

InuismiAin, itrangled." Hr Ray undentsndi Irom the leit that Aba bnilt the platform from which Haiak wai 
atrangled and that afterwards Aba married Tigt, Maiak's widow. The "him" in the figure evidently refers to 
Haiak; in the aeconnt in Vol. t. I onderstood that it was he who bnllt the platform. The two men w^ 
takoiah to eaoh other (v. Tables 6, Sa). 

■ A writer in TeiermoKnt Qeagr. Mitt. (xvni. 1873, p. 26i) states that the platforms for spearing dngong on 
Um island of Tnta were one (o two (wt above the water on the reef. 

d by Google 


platform to serve as a charm to eneure the approach of the animal, thifi ia illustrated 
in fig. 176 and pi. XXIU. fig. 1. (See Vols. v. p. 338, vi. p. 217.) 

Fid. 176. Uethod of hBrpooning dagong from s neit, dnwn hy Habniag tutSTea. To the right a ft nun 
atuiding OD the nsB nadj to spekr, below tfaa boaid U a ohana. To the left, one man has harpooned 
the dngong vai htt« flaog himself biokwRrda eo aa not to be entangled in the rope, the harpoon ii not 
drawn; another man has tied a spare rope ronnd the dagong'a tail. A drawing of a dugong hj Qua 
ii ftdded. 

The platform was always erected with its long axis in the direction of the wind, 
and the harpoon was held in the same manner. If the platform was built across the 
wind, it would blow through it and make a noise which alarmed the dugong. I was 
informed that the platform was used only at night, for it is only then that the dugong 
approach the shore ; in the daytime they keep to the open or over the large isolated reefe. 

Dr W. Wjatt Oill Btates that "the dugong is caught at the new and full of the moon, 
because the high tides then cover the reefs. At other times the water ia too shallow for 
the creature to pass over." After describing the platform he goes on to say, "The instant 
the spearman succeeds in striking one in the head — the only vulnerable part — he leaps down 
upon the animal, with one hand holding on to the line, and with the other hurling the spear- 
shaft back to his friends. A canoe in waiting follows to throw a rope to the adventurous 
spearman for him to slip round the head or tail, as may be most convenient. The struggle 
is soon over, and the prize towed ashore. A second wound ia never required. It is remarkable 
that BO large an animal should be so easily killed." {Li/e in the Soulh^n Idea (18761), 
pp. 301 — 2. The figure given on p. 196 is somewhat fanciful.) 

The Uiriam only occasionally erected platforms over a patch of dugong grass, and 
as the turtle also feed on the same grass they might spear the latter also, but the 

d by Google 


noro^ was erected solely for the dugong. As a matter of fact dugong were rarely caught 
at the Marray Islands though they are ahundant to the south and west. 

The dugOQg harpoon, wap (fig. 177, pi. XXIII. figs. 1, 4), is the most charactoriBtic 
implement of TorreB Straits, and it may have been invented by the Islanders. It is mora 
common in the western than in the eastern islands owing to the fact that dugong are more 
plentiful on the western and central reefs, but the harpoon is also used for catching turtle. 
It is made of hard wood and varies in length, excellence of workmanship and details o£ 
ornamentation. The length of the finer specimens ranges from about 3 to 4-6 m. (10 — IS ft.). 


Fill. 177. Wap, Bketohed at Uoialng i» 1886; length 4-6 m. (16 ft. 1 in.). 

It must have been no easy task to hack a harpoon out of a tree trunk, and it is 
surprising how straight and well-ehaped they are. According to my experience those fiom 
the Murray Islands are of somewhat crude workmanship, indeed this implement was not made 
in these islands but was imported from the western islands. The Miriam valued them more 
as ornaments or works of art, and like the imported spears tliey indicated the wealth of 
the owner; they wera exchanged or given as presents at marriages. Muralug was said to 
be the chief manufactory for the harpoons, though they were occasionally made elsewhere. 
The Mabuiag people pride themselves on their harpoons, but personally I thought those of 
Mnralug were finer, being beautifully finished and polished with oil, the butt end was also 
larger and well-shaped. But the Mabuiag men said that the Muralug implement was too 
heavy, for, when they jumped into the water with it, the spear had a tendency to fall vertically 
and thus miss the dugong altogether; therefore when they purchase a Muralug harpoon they 
pare off some of the superfluous wood. In less finished implements the butt end has an elongated 
ovoid form (figs. 367 — 370), but in better examples it is more barrel-shaped (figs. 366 a, 371), 
and in the finrat the form is more elegant with a distinct but gentle meridional swelling and 
a very slight ridge extending along the median line of the upper surface (pi. XXIII. fig. 3). 

The neck of the spear is frequently marked by several incised Unes or beads. The shaft 
is of fairly uniform thickness, being about 30 — 40 mm. (1^ — IJ in.) in diameter. 

The treatment of the posterior or upper end is subject to a good deal of variation. As 
a role about two feet from the end the shaft is square in section for about a foot in length 
and has on each face a row of holes into which tufts of cassowary feathers are inserted, 
each end of this region is sometimes decorated with a white cowry shell, a bunch of seed- 
rattlee, goa, or perhaps tassels of rags. The succeeding portion is plain, sometimes it is 
perforated by a long slit. Close by the end the shaft is square in section and bears four 
tnfta of cassowary feathers. A iMip is figured in the Albu/m, i. pi. 326, No. 6. 

Macgillivray (ii. p. 34) states that the head, kwiur^i, kwoidro, ktoiuro (W.), kwir (E.), 
kitffwra (Eiwai), was made "of bone, four inches long and barbed all round." The specimen 
collected by the RattUmake now in the British Museum, is of a pale brown, close^rained 
wood. With the exception of the modern iron heads mentioned abov^ all the beads I have 
seen are made of hard, heavy, clos&^rained wood. The more common form is cut out of a 
piece of wood that is triangular in section, the barbs being cut out of the angles; there 
are thus three barbs in a whorl, and usually there are four whorU. The heading, which 

H. VoL IV. 22 

d by Google 


serves as a atop to prevent the rope from slipping off the head, is aiao Erequentij triangalar 
in section; the terminal portion which is inserted in the harpoon is circuhur in sectjoo. A 
htrge specimen may be as much as 20 cm. (nearly 8 in.) in length. One figured on pL XXIII. 
fig. 2 is 146 mm. (6 j in.) long, it is triangular in section ; the other is 222 mm. (8j in.) long, 
it is quadrangular in section at the pointed end and has nine whorls, each consisting of four 
barbs. The ordinary type with three harba in a whorl was called pulil (W.) (see Arrows). 
I also obtained one with two rows and another with one raw of barbs, which were called 
respectively kopiiai (W.) and uxuaia (W.), they were quite new and appear to be too slender 
to be very efficient. I gave to the British Museum (93) a very strong specimen from Uer 
232 mm. long, with two rows of barbs, and there are two examples in the Dreeden Museum 
(4354, 4355). At the present time the head is invariably made from a file (triangular in 
section) which has been softened by heating and then allowed bo cool slowly, next the bubs 
are cut with another file, and finally it is re-tempered. 

Very few Miriam men are proficient in the uae of the wap, and they seldom cany 
it in their boats or canoes, the large hurar mcdil spear being their favourite implement. 
They speak of only two men who formerly could use the wap with dexterity, these 
were two K6met men, Obra (6)^ and Sager, an uncle of Jimmy Dei (4 b). Some of 
the young men now profess to be able to use it, and say that they were taught by 
the Wfetera Islanders. 

According to a Miriam folk-tale (Vol. Vl. p. 41) Barat of Moa taught the Western 
Islanders as follows: "Put a Hrmu (arrow-point) in the nose of the dugong and it 
will die, then you can lift it up and put it in your canoe." I did not get any con- 
firmation of this method of asphyxiating a dugong in Torres Straits, possibly it alluded 
to the Queensland method of fishing. The Moa and Muralug people intermarried, and 
the latter were in constant touch with Cape York natives. 

Hr W. Saville-Kent says: "The capture of the dugong is conducted on distinct principles, 
in difierent parte of the Queensland coast-line. In Moreton and Wide Bays, nets of great 
strength, having a mesh of a yard's width, when measured diagonally from knot to knot, or 
eighteen inches on the square, are stretched, at night, aoross the tracks the herds are wont 
to follow to their pasture- grounds. A little further north, at Repulse Bay, just above Mackay 
...the natives pursue the animals, moonlight nights being most favourable, in their frail haik 
oanoes, with heavy dugong harpoons, to which long lines are attached. Two men are included 
in one canoe, the business of one being to keep a look out for dugong, while the other bails 
the cranky boat. The endeavour, in the first instance, is to spear the animal through its 
fleshy tail, whereupon it is apt to twist itself up, and get entangled in the line. A second 
spear is then thrust through ita muisle, which stops its breathing, and thus the animal is 
speedily sufibcAted and dispatched" {The Oreal Barrier Reef of AuHralia, 1893, p. 328). 

Dr W. E. Both writes: "Dugong are either harpooned or speared, more generally the 
former. As pointed out to me by the natives on the Cardwell coast-Iine, very good indications 
of their presence in the neighbourhood are the short lengths of grass, which have been bitten 
off by the dugong grazing below, floating on the water-surface. I am informed that on the 
south-west portion of Bentinck Island are to be found alley-ways in the form of bush fences, 
built in the shallow water, into which these animals are driven" {yorlh QuMudand Bt&nograpky, 
Bulletin No. 3, Sept 1901, p. 30). 

■ Ct. Oenealogiea, Vol. ti. 

d by Google 


1 was informed at Mabuiag that when a man on a ne?£ apeared a dugong, he 
called out the name of hie sod, or if he had none, that of his brother, although neither 
might be in the attendant canoe. The men in the canoe, which was stationed some 
distance o£F, by this means knew who it was that had speared the dugong. When asked 
the reason for this, and why he did not call out his own name or that of someone 
in the canoe, the only answer I could get was, "That foshion belong we fellow." 

In Mabuiag when a dugong is caught half of it is given to the man who owns 
the gear and half to the owner of the canoe. A man who has borrowed both gear 
and canoe only gets a small portion of any dugong be may have caught. 

The dugong was a totem in most of the western islands (v. pp. 154, 155, 162 — 164). 
Various practices were associated with it in this capacity (v. pp. 182, 183, 339—341), 
and there was a totem taboo (v. p. 186). Charms connected with the catching of 
dugong are described in Vol. v. pp. 337 — 342, and in Vol. vi. p. 217; see also Album, 
II. pi. 203. 

The foIlowiDg statements are made by Dr W. Wyatt Oill : — " In passing through a grove 
of palms on the Island of l^uan [Dauan], I came upon a dugong-giving god. It was simply 
a large perfectly round stone, painted red. A white streak encircled it. Some sacred stones 
have two white streaks intersecting each other. The etone itself is intended to symbolise 
the dugong; the bands or streaks, the ropes which will, it is hoped, make it a prisoner. 
The head-man who has resolved on a dugong-huot presents an offering of fish and cocoa-nut. 
In approaching the stone he mimics the paddling of a canoe. On getting tolerably near, he 
rushes at the stone and &nnl; grasps it in his arms, all the while uttering a prayer for 
success. The firmer the grip of the worshipper, the surer will be his success. As these stones 
are often of considerable weight, they occasionally slip down — an evil omen in the estimation 
of the dngong-himt«r [the illustrative figure on p. 322 is not very convincing]. 

" On Warrior Island [Tatu] stood, until lately, a stately banyan-tree, c(»npletely ornamented 
with dugong bones, the supposed shrine of a spirit possessing the power of giving or with- 
holding success in dugong-hunting. Under a remarkable tree of tbe same kind on Jervis 
Island [MabuiagJ dugong-f easting still goes on, tbe bones being piled up round the trunk. 

"In some of tbe Straits Islands, when a dugong is caught, tbe skulls of parents and 
other relatives are brought out, talked to, wept over, and presented with a portion. When 
this worship is concluded, feasting begins, followed by dancing. They believe that the spirits 
of their deceased friends ud them in chasing dugong, turtle, etc. Hence tbe importance of 
securing their good-will" {L^e in the Southern Idea, 1876, p. 302). 

As might be expected, the dugong is referred to in folk-talea Sesere of Badu 
(v. p. 40) is the reputed discoverer of the edibility of the dugong, this feet, and the 
method of capturing it and of making a platform were revealed to him by the skulls 
of his parents, which he employed in the usual manner for divinatory purposes. All 
the detaib relative to the capturing of a dugong, even to the method of cutting it 
Dp when caught (p. 138), are still carried out according to the plan thus revealed to 
Sesere. Gelam (t. p. 38, vi. p. 23) made a model of a dugong with which he fooled 
his mother and in which he swam from Moa to Mer. 

Digitized by VjOOQ IC 



As the tactics employed in wariare and the method of fighting are sufficiently 
described in Vols. v. pp. 298 — 319 and vi. pp. 189 — 191, I need only refer here to the 
weapons and other objects employed when Sghting. 

Vakious Weapons. 

The weapons, ares lu (E.), of the Torres Straits Islanders were mainly the bow 
and arrow and various kinds of clubs, to which were added the javelin and spear-thrower 
in some of the western islands. There was no special kind of thrusting spear in general 
use, neither were swords, axes nor similar weapons employed. A sharply pointed aok 
(p. 127) is frequently used as a dagger in New Guinea and it may have been so 
employed in Torres Straits, but the only instance known to me of a bone dagger is 
in a folk-tale (VI. p. 21), where Sida, who used it, came from Daudai. It was the 
same hero who sharpened a piece of kaka wood to use as a sword (v. p. 33); this is 
a very strong bard wood with which, made in the shape of a sword, the women of the 
Fly River at the present day help the men when hard pressed in their fights. 

When the former marriage customs of Mer were described to me I was informed 
that in the marrit^ fight (VI. p. 114) a weapon made of shark's teeth fastened on 
to a stick was employed ; the same weapon is referred to in the Bomai-Malu legend 
(VI. p. 40). As none of these exist I got a native to make one for me (fig. 178), 
This specimen is made very carelessly, but there is no reason to doubt that a weapon 
of this kind was once in vogue. The shark's jaw is 50 cm. long, and the total length 
of the stick is 136 cm. I do not know whether this weapon was known to the Western 

The only record we have of slings occurs in the narrative told by Mr Wilkin in 
Vol. V. p. 311, where two boys on Moa are described as "slinging stones with grass 
slings at the tree-tops and shouting afler the manner of boys when they play." I have 
looked up Mr Wilkin's original note which says "playing with grass apparently slinging"; 
presumably the fuller account was supplied from additional information. Ut Ray obtained 
the Miriam phrase ma baker dau, you stone sling, I do not know bow he obtained 
this translation, which he modified in the Vocabulary to " throw " ; but the ordinary 
phrase would be: to throw a stone, baJcer hataueredi; to throw stones, baker itimeda. 
.Stones would naturally be used as missiles, and I collected at Mer in 1888 a stone. 

d by Google 


idid baker, which was carried in the hand when walking to use as a miasile and was 
also employed as a food-pounder, it is 88 mm. in greatest diameter {Album, i. pL 323, No. 6). 

The beheading knife and head-carrier may be regarded as post- 
mortem weapons. 

Shields or armour, save the groin-shell, were anknown. 

Bow AND Arrows. 

Bows and arrows were the only missile weapons employed in Torres 
Straits with the exception of the javelin, the use of which was confined 
to certain of the western islands. 

It is doubtful whether any of the natives of North Queenalaod really 
employed the bow and arrow ; apecimens have been seen at Cape York, but 
it must be remembered that certain Western Islanders were in the habit of 
visiting that district, and as soon as Europeans began to fish and trade in 
Torres Straits tbey made Somerset in Albany Pass their head-quarters, and 
brought in their train numerous islandeni. The bows seen by travellers may 
have been in the hands of visiting islanders only, and even if they belonged 
to Australians they were imported from the islands. So far as I am aware 
there is no evidence that this can be r^;arded as an Australian weapon in 
any sense of the term. 

On the other band bows and arrows are of universal occurrence along 
the neighbouring coast of Kew Guinea, indeed, except for a considerable part 
of the south-east peninsula, they occur everywhere in that island. Bows and 
arrows, especially the latter, formed an important article of export trade from 
New Guinea to the islands (cf. v. p. 295, vi. p. 186), as the reeds of which 
the arrow shafto were made do not grow in the islands. 

The bows, gagai (W.), aarUc (E.), are usually of large size and 
invariably made of bamboo. "The bows are made," as Jukes pointed 
out (Voy. of Fly, i. p. 179), "of the upper part of a stout bamboo, 
partly split in half, flattened and bent over the fire. The string \3agai 
urw (W.), let (E.)] is a broad strip of the tough outer rind of a bamboo, 
and the fastenings are very ingeniously and firmly made. The bows are 
large, and very powerful, some being more than seven feet [2*134 m.] 
long, and in the centre more than three inches [76 mm.] wide, and an 
inch [25 mm.] thick. The bows which we collected varied from 1'64 m. 
<64i in.) to 1"88 m. (74 in.), but longer apecimens occur in other ^ weapon uinad 
Museums." with a shark's 

Captain Flinders says {Voy. 1. 1814, p. xxiii) that the bows at Erub t^*^. Mer. 
" are made of split bamboo ; and so strong that no man in the ship 
could bend one of them. The string is a broad slip of cane, fixed to one end of the 
bow ; and fitted with a nooee, to go over the other end, when strung." 

In all the islands the bow is held vertically, that end of the bow being held 
uppermost which in the living bamboo grew nearer the ground. Jukes (1. p. 209) 
gathered at Erub that the opposite was the case, and his informant "could give no 

d by Google 



reasob for the cuBtom." In striDging and unstriDging the bow the etaae end is placed 
agaimt.the ^rquod, as it is the stronger. 

When shooting the bow is held in a vertical position in the left hand, and the 
butt of the arrow is held between the thumb and flexed forefinger of 
the right hand, the string being drawn back either by the second and 
third fingers or by the three remaining fingers. The arrow is steadied 
and shot between the forefinger and the second finger of the left hand . 
and to the left side of the bow (pi. XXIX. figs. 1, 2). 

This is the "secondary release" of Morse (E. S. Morse, "Ancient 
and Modem methods of arrow-release," BtUL Essem Inst. Mass. U.S.A. 
XVII, 1885), but that author does not record the use of the little finger 
in assisting to pull back the string. The secondary release is the 
universal method in Torres Straits, and appears to be characteristic of 
British New Guinea, but according to Van der Sande the "archaic 
release" ia that most common in Netherlands New Guinea; he discusses 
this point. 

The Torres Straits Islanders even in 1888 had for so long given 
up the practical use of this weapon that no reliance could be placed 
upon the comparison between the exhibitions of archery that I saw and 
their former proficiency. Their extreme f^ge was probably about 200 
3ranlB, and I was not impressed with the accuracy of aim, but owing to 
lack of practice that was readily explicable. 

Dr Seligmann {Geograph. Joum. 1906, p. 228) experimented with the Toro, 
on the Bensboch river at the Anglo-Netherlands boundary, he found th&t the 
time an arrow took to traverse forty paces was between 1 and \\ seconds. 
Van der Sande also discusaea Papuan archery, and states that "the power 
of the shot as a rule exceeds the expectation of Europeans" {Xova Guinea, 
p. 261). Sir W. Macgregor tells of an arrow shot through a man {Ann. Hep. 
C.AI. 1893, p. 34). 

Jukes (l.c. p. 179) says, "They shoot their light long arrows to 
great distances, but not, I think, with very accurate aim." Dr Wilson 
(J,.c. 1835, p. 311) however refers to the "amazing feats" (which he does 
not specify) of " these athletic savages" in Mer. Captain Flinders writes, 
"The depth to which the arrows penetrated into the decks and sides of 
the brig was represented to be truly astonishing" {l.c. i. p. xziii). 

A spare bow string, tapal (W.), was usually carried doubled up 
in the bracer (pi. XIV, figs. 9, 10), the modifications which this has under- 
gone are described on pp. 57, S8. 

Extra arrows are held in the bow hand. A quiver is not employed, but I have 
seen bundles of arrows, kaunU or konil (W.), tied in two places with a piece of string, 
the intermediate portion of which formed a handle for carrying the bundle. A plait^ 
strap 81'5 cm. (32 in.) long, used for carrying a bundle of arrows in this manner when 
goii^ out to fight, is sketched in the Album, i. pi. 343, No. 4; it is the same kind 
of strap or band, hcUa, which the Daudai women use for carrjring firewood. 

Fio. 179. Bow, 

d by Google 


The collective name for arrowB in the western islands is tatak or taiMk (pi. tcakel), 
tarek in Muralug, and in the eastern islands l^p or sank, bnt the latter includes the 
bow as well ; I also obtained aarik oker as an au nei or general name for arrow in 
Mer. Various types of arrows were named from the wood of which the head was made, 
SDch as biei kep, kua kep, etc., or from the carving of the head, such aa U op, maa 
fece, while others were called by similar names to those which they receive in the 
district of New Quinea where they are made. 

The Rev. J. Chalmers states in a MS. that the collective name for arrows at 
Tureture and Mawata in Daudai is teere; iana, boboku, garagaro, beromamu, parti iatia, 
koteretuti, dupamutu and aokeri are so named from their markings, carvings, or shapes. 
They are used indiscriminately ; the sokeri are used for long distances. The arrow that 
has killed an enemy is named in song. The bow is called gagari, and the bow 
string ivada. 

As all the arrows are imported into Torres Straits from New Quinea, there is no 
need to attempt to discriminate between the different kinds that were obtained from 
various islands, especially since they have not been of any practical importance for the 
last fifty years, and certainly the stock has not been Renewed to any extent for many 
years. The following account will therefore comprise a brief description of all the ma 
varieties of arrows known to me to have been obtained in Torres Straits, and it w 
serve at the same time as an enumeration of most of the types of arrows made 
the neighbouring district of New Guinea. 

None of the arrows in New Quinea are feathered, and the end of the shafb is 
cat square, there being no nock. The arrows usually consist of a head composed of 
wood, and a reed shaft. In one class of arrow there is added a bone point. Another 
class possesses a bamboo bead which may be fastened directly into the shaft, or into 
a piece of hard wood which in its turn is fastened into the shaft; this intermediate 
piece of wood is best termed an aftershaft. Some might like to term the bard wood 
portion of a bone-tipped arrow an aftershaft, but it seems preferable to me to regard 
the bone tip as an accessory to an original wooden head. 

A. Arrowi with wooden pointt. 

(a) L<mg arrows with plain heads. 

The head consists of a cylindrical, gently tapering piece of bard wood which is 
generally blackened, occasionally there is a slight swelling near the point (fig. 180 a). 
In some the lower port of the head is varnished with a red gum which is worked 
up into three groups of encircling ridges. The average length above the lashing is 
38 cm. (15 in.). 

The junction of the head with the shaft is very neatly and closely bound round 
with a thin lashing of vegetable fibre, which is covered over with gum. 

The red varnishing of the lashing is generally extended over a small portion of 
the upper end of the shaft. Below this, usually for one intemode, the yellow skin of 
tiie reed is left entire. As a rule the rind has been scraped off the remainder of the 

d by Google 


blackened shaft so that the colour may adhere more readily to the rougher and more 
porous surface. 

The length of the Cambridge specimens varies between 1'232 m. (48J in.) and 1435 m. 
(56J in.), other specimens average in length from about 1*445 to 1'626 m. (57 to 60 in.). 

They are known as kaigob or kaigup in the western islands. Maine of Tutu called 
them Wigilkau kaigob, probably from their place of origin. 

■ TTrr ii i ii V- ' •'■■' ■" " '^- - - 

Fio. 180. LoDg aad short arrows with plain heads, Her. 

(6) Short arrows vnth plain heads. 

One very clearly marked group consists of small, slender arrows which are imported 
from the Fly River. The heads taper gradually to a very sharp point, usually they are 
quite plain but occasionally ma^ be carved. The heads are sometimes simply inserted 
into the end of the shaft which is cnt off abruptly (Sg. 180 c), but in most cases 
the latter is tapered and a plaited cane lashing placed over the junction (fig. 180 b). 
The shortest head in our collection measures 275 mm. (10{ in.) and the longest 405 mm. 
(16 in.). The whole, or the greater part, of the uppermost intemode of the shaft is left 
with the rind intact, and so appears of a bright shining yellow colour. The rind is 
scraped off the rest of the shaft, either completely or longitudinal lines of the rind 
may be left, the pared surfiu^e being of a dull pale brown colour. The average length 
of these arrows is about 1'14 m. (nearly 45 in.). The shortest collected by me in Mer 
is 1*044 m. (41 in.) and the longest 1*165 m. (nearly 46 in.), but in 1889 I collected one 
1-245 m. (49 in.) long. 

Some arrows, taraiat, of this type collected by the Rev. J. Chalmers in Waboda 
or Wabuda (an island in the delta of the Fly River) are however larger than the 
foregoing, but they fall into a longer and a shorter group, the former, with red palm 
wood heads, run to 1*425 m. (56 ia) and of the latter the shortest is 1*022 m. (40^ in.> 
All the beads are plain. The shaft is bevelled above with a plaited cane lashing. As 
a rule the rind of the upper portion of the shaft is intact, but in some it is removed 
in longitudinal linea The remainder of the shaft is usually entirely scraped. These 
are called (*W by the Western Islanders, bum is also their word for a small arrow, 

I obtained one variety of this group, the bead of which was made of a wood that 
readily splinters, tmd 1 was informed that before using them in war&re the combatants 
split the point with a thin shell so that splinters might become detached in the enemy's 
body. Maine of Tutu called them bop, and said they were made by the Wogatfiri men. 

Another variety of (a) and (b) were called bisi kep in Mer, as the head was said 
to be made of sago-palm wood. 

d by Google 


(c) Arrows with carved heads. 

There is a great variety of wooden pointed arrows with carved heads. 

One group possesses a gradually tapering head (fig. 181 A, b) which terminates in 
a longer or shorter conical point, but sometimes the head itself is slightly swollen so 
as to give it an extremely attenuated spindle-shape (f, o, h). The accompanying illus- 
trations (fig. 181) give a better idea of the characteristic variations than any description 
can do. Sometimes (a) a bead is carved below the swollen end. In c the bead has 
an elongated narrow sagittate point. In one case (e) a seriea of cone-in-cone is carved 

' MfT=^ 1 I III 1 , 1 

Fio. IBl. AnowB with dmpla oured haaia. Her. 

below the point, and in another (a) the same area is decorated with four series of 
roughened nicks. In H the slightly swollen portion near the point ie indistinctly 
rectanguUr in section. The shaft is either blackened all over, or more or less of the 
bare rind may be left. The arrows average from about 1-385 m. to 1-485 m. (54J to 
5^ in.) in length ; the head averages 380 to 510 mm. (15 to 20 in.) in length, but 
some are distinctly longer and others shorter. 

The name given to these arrows in Daudai ie susuome and the Wabuda name is 
boboku; the Western Islanders call them baul'Uaig, or baulilug, one with a very conical 
point was called bok in Muralug. 

A well-marked type with a conical or fiisiform point (fig. 182 A, B) is decorated with 
a zigzag beading about the middle of the head. The fibre lashing is usually gummed. 
The greater portion of the shaft is generally scraped and blackened, but part may be 
varnished with red gum; the upper intemode may be left intact. These arrows average 
1-380 m. in length. The arrow shewn in fig. 181 d is somewhat similar but the beading 
ia not a zigzag. 

: ^a^ 

" JW " I ' " — 

Fro. 183. C»rved arrows with brnding, Mm. 

Another variety consists of arrows with a long conical point, the head being carved 
with beads (fig. 182 c, d). The shaft ia scraped with the rind left at the anterior end. 
The average length of these arrows is 1'250 m. 

I was informed that the necks of the conical points, or other constrictions on the 
H. VoL IV. 28 

d by Google 



head, were nicked before use in order that they might readily break off and remain 
within the body of the enemy, and Chalmers has a MS. note that "bead goes inside 
and breaks off." 

A selection of wooden pointed arrows with carved heads is given in fig. 183, and 
several more will be found on pi. 324 of the Album, Vol I. They are so varied in 
form that no useful purpose would be served here in describing or classifying them. 
They vary in length from 1-335 to 1-565 m. (52^^ to 61J in.). The longest (c) is tipped 
with a cassowary's claw. One somewhat similar to B from Wabuda is called sittmasapate. 

Fia. 18>. Anom with TfeHed oarriBg, Mer, 

{d) Arrows with a single or double row of barbs. 

Arrows with a single or double row of barbs (fig. 184) usually have a fine gummed 
lashing, and some have the base of the head varnished with a red gum which is 

ITia. 181. AiTowi with k aingla or double row ot wooden barbi. Her, 

worked up into three groups of encircling ridges as in fig, 180 A. In moat cases the 
upper end of the shaft has the rind entire, but the rest is scraped and blackened. 
In one (c) the head has an indistinct zigzag beading. These arrows vary from 1-280 
to 1-460 m. in length (50^ to 67^ in.). A typical specimen is figured in the AUtum, L 
pi 324, No. 1, length 52 in,, head 11 in., and two specimens, Nos. 9, 10, with a cane 
band at the junction of the head with the shaft, the former is only 39 in. long, with 
a head of 10( in. The B type is called oto-oto at Wabuds, and Dr Seligmann obtained 
the name caraaawa for one from Sumai, Kiwai, which was very similar bo c. One was 
called kus kXp in Mer, probably from the wood of which the .head was made ; it was 
also called Saibaunde kip or arrows of the Saibai people. 

(e) Bird hoUs. 

Only two arrows were obtained with heavy blunt ends, the purpose of which evidently 
is to stun plumage birds, such as birds-of-paradise. Probably they were imported as 
curiosities, since there would be no use for them on Mer where they were collected,- 
iroleas it were for shooting the white reef heron. Both have slightly carved conical 

d by Google 



heads of hard wood. One (fig. 186 a) has a total length of 1135 m. <4i} in.), the head 
is 102 mm. (4 in.) long, the base of the cone measures 81 x 38 mm. There ie a small 
string lashing in addition to a long lashing of fine fibre. The whole of the shaft is 
scraped and blackened. The other (b) has a total length of about 1-32 m. (52 in.), the 
head is about 75 mm. (3 in.) long, the base measures 26 x 29 mm. The conical end 
and the central corolla have been painted red and the other two corollas blue. There 
is a neat fibre lashing, and the shaft has the rind entire. 

Fio. 18S. Bird bolts, Mer. 

A third specimen (fig. 186) is also perhaps a bird arrow. The head is carved to 
represent a fish's head with the mouth open. The meaning of the loop-shaped decoration 
which occurs on the top of the neck is discussed in the Section on Decorative Art. 
The head up to the lashing is 18o mm. in length, and is smeared with red ochre; 

Fto. 186. Bird bolt, 

it is simply stuck into an arrow shaft. It was called gir kep and tu kep, and originally 
came from New Guinea. A bird arrow with a squared end in the Brit. Mus. is figured 
in the Album, i. pi. 324, No. 6. 

(/) Arrows with more than one point. 

The tete baur (K) are arrows with several points, which are employed in shooting 
birds*. Three of our specimens have four points while a couple have two, but one of 
these has bone barbs and is probably not a bird arrow. Tete is the Kiwai word for 
the pronged fish-spear. 

One with four points (fig. 187 b) has a total length of 1*21 m. (47| in.), the points 
being 216 mm. (8^ in.) long. These are made of splints of bamboo tied firmly about 
the middle to one another by a lashing ; the proximal portions of the points are cemented 

1 The Uiriam atata that these urows were need foi birds, thoie with barbed points or aeveral points 
■aggest Bsh urows, bnt, bb pointed oat in Vol. t. p. 40, footnote, the only reeoid we have of this praatioe is 
in a lolk.lale. 

d by Google 



together with a white subatance which I believe to be saga They are fastened to 
the abaft by a neat laahiog of v^etable fibre which is coated with gum. 

A very similar specimen is M75 m. (46^ in.) long, the points measuring 23 cm. (9 in.). 
A third specimen has palm wood points and is IM m. (62} in.) long with points 25 cm. 
long. The Former only has a cement of sago. 

One two-pointed arrow (fig. 187 a) has a length of 1-232 m. (48^ in.), the pahn 
wood points being 295 mm. (llj in.) long, irregularly serrated, and lashed together by 
a cane about one-third up. The points are fastened to the shaft by a cane lashing. 
The rind of the shaft is entire. 

The other specimen (fig. 187 c) is 1-395 m. (55 in.) long, the head being over 
385 mm. (15J in.). The head is made of a red wood, and the points are provided with 
bone barbs. There is in the Museum another arrow of this type which was collected 
by the Rev. J. Chalmers in Daudai and is called paru iana ; it is 1'4 m. long. In the 
Vocabulaty a two-pronged spear is called barvgid (W.) (p. 157). 

B. Arrows ^th bamboo headi. 

As frequently as not the bamboo heads are inserted directly into the shaft (fig. 188 a), 
otherwise an aftershaft is present. The shortest of these arrows in the Cambridge 
Collection (fig. 188 b) is 136 cm. (53^ in.) with a head of 23 cm. (9 in.), and the longest 
166 cm. (65^ in.) with a blade of 38 cm. (15 in.), thus the shortest arrow has the shortest 
head, and vice versa. The longest I have measured is 1*83 m. (72 in.), and I have one 
which combines this greatest length with the least width of the head. The average total 
length is 156 cm. (61} in.), and the average length of the head is 28 cm. (11 in.). The 
broadest head is 30 mm. and the narrowest 13'5 mm. 

Sometimes the head is a piece of narrow bamboo split in half and sharply pointed, 
or it may be a lanceolate piece of a large bamboo also with a sharp point. Thus in 
section the head may be deeply concave (a) or nearly flat (b, c). The latter type leads 
to those arrows of which the head is made of a sharp, flattened, lanceolate piece of 
palm wood ; but there is no record of any of this type having been collected in Torres 
Straits, it probably comes from the east of the Fly River. The head is sometimes 
notched at the base (c), but the notches do not appear to be tally-marks ; rarely there 
may be a little simple decoration. 

The aftershaft may be plain or carved. The head is usually lashed to the aftershaft 
by a strong fastening of stout string, which is much stronger than the lashing of other 
arrows ; it is often coated with gum. The aftershaft is fastened to the shaft in the same 
manner as the heads of other arrows to their shafts. In one variety the lower lashing 

d by Google 



consists of a plaited cane binding, it may have a flat bamboo or palm wood head, but 
this variety apparently does not occur in Torres Straits. 

The shafts vary in the same way as those of the bone-pointed arrows. 

In a few specimens a lock of hamau hair is tied round the base of the head (a), 
or a strand of vegetable flbre, and in one instance a strip of (Cuscus ?) skin. There is 
a specimen from Erub in the Brit. Mus. about 127 cm. (52 in.) in length, with a lock 
of human hair tied round the base of the head; Mr Kennett who collected it made the 
following note: "When an arrow ia recovered by which a man has been slain a lock of 
his hair is fastened to it and the weapon kept with great care." 

Fio. 188. Arrowi with buDboo heads, Uer. A. H«ad deeply conoave, five Dioka at eaob side ol the base, 
a twisted atraud of hnmau hair tied lonact the lashing, tbe opper part of wtuoh is string and the lower 
•inootb strips of v^etable fibre. Ths npper end of the shaft has the rind entire which is scored b; 
fine intersecting lines, the rest ot the shaft is black. B. Head flattened, tbe lashing of the head on 
the aftershaft is thiokly coated with gam, tbe short sftersfaaft is fastened by means of a neat gammed 
laabing. C. Head flattened, notched at base on hollow side and there is a simple herring-bone pattern 
In this T^on an the other ride. There is a long strong string lashing followed by one of delicate fibre. 
The skin of the upper end ot the shaft is entire and is deooratad with intersecting scratched lines, the rest 
is scraped and blackened. 

These arrows are used for shooting wild pigs or for wariare. I was told that when 
employed for the latter purpose they were aimed at the abdomen of the enemy, in order 
to rip it open (V, p. 13). There is no doubt that tbe broad, sharp bead would cause an 
iigly wound, and the groove would facilitate copious beemorrhage. 

In the western islands these arrows are called sukuri or sUkSri, aUkri, buru-sitkUri 
(Mab.), buru-tisal (Tutu) (burum, pig) or hoi kaigob; in the eastern uwere. The Daudai 
and Kiwai name is uwere; were or wen is also the name for a bamboo beheading knife. 
A<xordinf{ to Chalmera, sokeri is the name of an arrow that is used for long distances. 
They are made in Kiwai and probably elsewhere. 

C. Amwi with bone polnti. 

The arrows of this group consist of three portions, a shaft of reed, a head of hard 
wood, and a bone point; they agree in all essentials except in the treatment of the 
head (or, as some might prefer to term it, tbe aftershaft). This may be (a) left quite 
plain, (6) carved with simple devices, (c) carved with bracts, (d) carved to represent 
a man, or (e) carved to represent a crocodile. 

The extreme lengths of those that I have measured are 1*28 m. (50^ in.) and 1*62 m. 
<63| in.); the bulk of them fall within 1-42 to 1*58 m. (56—62 in.). The sbafl averages 
1 m. in length, but frequently varies by 5 cm. longer or shorter ; the lashing is about 
5 — 8 cm. wide. The bone point, with it« barb, varies fix>m about 120 — 265 mm., the 
usual length , being about 196 mm. (7 j in.) ; the free portion of the barb varies from 
2— 3 cm. 

d by Google 


The point ia composed of a narrow flattened bone, the lower end of which is 
neatly lashed on to the upper end of the head, the junction being thickly coated 
with gum. 

A barb is always present, which consists of a sharpened splint of bone, often 
artificially rounded in section; sometimes the bone is cut bo form a hook, probably in 
imitation of the wing claw of a cassowary, which is also occasionally employed instead. 
The barb is lashed to the head, and smeared with gum. 

These points are reputed to be poisoned by being stuck into decomposing human 
corpses, but of this there is no corroborative evidence*. Macgregor says (Ann. Bep. C. A. I. 
1892, p. 48) : " There is nothing to lead one to suppose that these tribes [in New 
Guinea] prepare poisoned arrows, but the bone-tipped arrows often used probably contain 
septic germs lodged in the bone." At all events the Torrea Straits natives always take 
the greatest care not bo get pricked by the points of these arrows. 

I suspect that the points are made of human bone, in which case the dread of being 
wounded by one would find a ready explanation if there be any idea of a power like 
mana being associated with the bone. 

In former days, before going to fight it was customary in Mabuiag to touch the 
white band of one of the Kwoiam head-dresses (t. p. 371) with the points of arrows 
in order that they might not miss their mark. Mr J. Cowling informed me that a 
man made the white paint poisonous by thinking hard while pounding the shell; the 
arrows were then dipped into this and that made them poisonous. It thus appeais 
that for making their arrows deadly some at all events of the islanders relied on the 
transference of power irom the relic of a hero (t. pp. 371, 377), others by the intensity 
of their own thought could infect powdered lime with power. There is no evidence 
that the islanders ever claimed actually to poison their arrows in our sense of the term. 

There is great variety in the treatment of the shaft in all the arrows of this 
group. Sometimes the rind is scraped and the whole surface blackened, more frequently 
the whole of the upper intemode and the uppermost portion of the second intemode 
are left iutact, sometimes the whole of the upper two intemodes is so left, or the 
uppermost part of the first only, or merely a central band. As a rule the rind has 
been scraped off those parts that are blackened. 

Flo. leg. DeoormtioD on ihAtt of krrovB, Mer. } n&t. nie. 
The part, which is of the natural colour may be left quite plain, or it may be 
decorated with groups of fine encircling lines with occasionally scratched chevrons and 
other simple devices; sometimes the patterns are in zigzagged lines. 

On some shafts the lower third of- the first intemode and more or less of the 
second are decorated by removing several narrow longitudinal strips of the rind (fig. 189b), 
■ For ft difcouioD on the repated poisoning of arrows in HoUnesia of. Be*. Dr B. B. Codrington, TA« 
Melanaiaju, pp. S06— S12; and Jtmm. Anth. IniU x». 1S90, p. 21G. 

d by Google 


or only the first interaode is treated in this manner (fig. 189 a). The bands thos' formed 
shew black upon the yellow ground of the rind that is left ; a line may be scratched 
round the shaft in the upper intemode to prevent the strips firom going too hrK 
Simple patterns of various kinds formed by paring the rind of the shaft may also occur. 
A " poisoned " arrow is called taiai kimva (W.) ; kimus (W., E.) is the bone point, 
it also means the shin in the west, and aapur kitnus is the wing-bone of the flying fox 
which is used for piercing ears. Ima (W.), kaCgob (E.) is the general name for this 
class of arrows, it also includes, I believe, those with bamboo points. 

(a) Arrows with plain heads. 

The wooden head is usually cylindrical, slightly tapering at each end, and blackened 
all over. When it is made of a hard red wood (I believe of tms wood) there may 
be a broad black band and sometimes red gum is applied. Sometimes the head is 
made of palm wood, or a pale wood. Occasionally the head and its lashings are whitened 
with lime. 

In some arrows the head is more swollen in the centre, and flattened in the same 
plane as the barb. 

In the western islands this type is called kaigob or dodu. 

(b) Arrows with heads carved in simple devieea. 

This is not a very numerous class, but the variations which occur are too considerable 
to detail, and I am not aware thst they possess any significance. Parts of some of 
them appear to have been suggested by details on arrows of the crocodile series. The 
Daudai name for this type is iana, a, term which probably also includes the plain barbed 
arrows. Specimens with two barbs are described on p. 180. 

(c) The Claw Arrow. 

This is a very distinct type of barbed arrow, of which the carving consists of two 
series of projections and a beading. In our collection the shortest arrow measures 1*488 m. 
and the longest 1605 m. (584—63} in.). 

This type of arrow was sometimes spoken of as the dugong harpoon arrow, presumably 
from a resemblance of the carving to the plumed end of a wap (pi. XXXII. figs. 8 — 14). 
In the western isluids it is called putil, "having pui" in the eastern potim or poftn, 
jKrf (E.) being a nail, claw of a bird, operculum of a mollusc. I venture to suggest 
that the claw of the cassowary is intended, for this bird can strike out very strongly 
with its legs and the large claws enable it to produce a dangerous wound. The Eiwai 
name is fforagara, and the claws are called orto. 

The upper series of projections varies from about 11 to 16 cm. (4J — 6} in.) in length, 
and consists of from twelve to twenty or more whorls of bract-Uke projections in sets 

> The pTiaaiple ol this ttjh of d«oontioii ia ver; olwracteriitia.or arrov« trom the Santa Ctqe and Solomon 
lilands (et H. Baltonr, "On the ETolotion of a Cbaraoteristio PaUero on the Shafts of Arrowe trom the 
Solomon lelaude," Jovm, AntK Ifut. xvn. 1SS6, p. S3B, and The Svolutfo* of Dteorative Art, 1898, p. 101). 

d by Google 


of four; for lack of a better term I propose to call them bracts, and to describe those 
arrows which have this decoration as bracteate. The head appears to have been cut 
as a four-sided rod which gradually increases in size from above downwards. The upper- 
most portion of this rod is rounded and forms a tapering cone to the end of which 
the bone point is affixed. The bracts have evidently been produced by nicking the 
angles of the square rod, and subsequently their form has been developed by further 
carving. The bracts slightly increase in size from above downwards, and frequently the 
last whorl is perceptibly larger than the preceding one. Occasionally it is not easy to 
determine whether the last bract is really a bract or the uppermost claw of the next 
series. Occasionally the bracts are little more than slight triangular flaps, but usually 
they are quite prominent and stand out well from the Btero and from each other. 

The second series of projections I propose to term claws, as I believe it is from 
them that the name of this type of arrow is derived. The claws may consist of two 
or of three rows of four elements. Typically they have the form of a lanceolate leaf, 
except that there is no basal constriction, but occasionally they are almost rod-like 
(pi. XXXII. fig. 10); they are deeply undercut.. The lowermost are always the largest 
and they vary from 5 to 8 cm. (2 to 3J in.) long, the average length being 6 cm. 
One arrow in the Cambridge Museum has no claws, another has only one whorl and 
this arrow is exceptional in having five bracts and claws in each whorl. 

There are two classes of this group of arrows: (1) those with a simple beading 
below the claws, and (2) those with a cylindrical or barrel-shaped swelling below the 
claws. (1) The beading may be simple or grooved (pi XXXII. figs. 9, 8). (2) The 
cylindrical or bfurel-shaped swelling is usually bounded above and below by a bead. 
The cylindrical swelling is generally decorated by raised vertical lines, between which 
are raised dots, zigzags, etc. The typical decoration of the barrel-shaped swelling consists 
of a symmetrical pair of bowed lines enclosing a row of dots (pi. XXXII. fig. 13). In 
one specimen in the Museum (fig. 14) a face has been carved on the oval space, and 
lines and dots cover the corresponding space on the other side. In another two pairs 
of rings (leyea) occur in the areas between the oval spaces (fig. IS). 

The undercut portions may be painted white or red, the bracts and claws may be 
reddened but they are usually black. One arrow in the Museum has a ring of human 
hair between the first and second whorls of claws. 

(d) The Man Arrew. 

The heads of these arrows are carved to represent a man (fig. ISO). Sometimes 
they are made out of a slender piece of wood, but usually they are of larger diameter 
than other carved arrows ; good specimens vary from 25 — 35 mm. in the antero-posterior 
diameter of the man's head. They are called pandaig or paruag (W.), opop or le op (K), 
both terms have reference to the face. Dupamutu is the Kiwai name. 

The head of the man is surmounted by three or four (at most seven) whorls of 
bracts, usually the lowermost bracts are considerably larger than the othere; rarely claws 
are carved^ (pi. XXX, figs. 7, 10). The crown of the head is usually cut square or 
> It U exUttuely rue to fiod an mtow in which th« bnet* are wanting, m in pi. XXX. fig. 1. 

d by Google 


is slightly bevelled, but ocowionally it tapers gradually to the stem beariog the brscts 
<pl. XXX. figs. 4, 8, 10). 

The type of tbe &ce is very characteristic, beiog greatly elongated, The forehead 
is nearly always a large blank sur&ce; it is absent in pi. XXX. fig. 3, or rather the 
eyes are cut on Uie forehead. The eyes are generally quadrangular, but are sometimes 
ovoid or triangular. The nose is usually an almost imperceptible ridge, but is occasionally 
slightly prominent; the alae are M-shaped; in a few cases (fig. 190b) a line below 
the noee indicates the nose-stick. The mouth ia variously rendered, usually the teeth 
are shewn. Whiskera are always represented and frequently a moustache 
and beard, the latter being the more constant (fig. 190 a). The high 
forehead is bounded above and at the aides by a band which marks 
the beginning of the scalp. The top of the back of the head is 
always plain, below this the hair is suggested by very varied 
combinations of straight lines, zigzags, chevrons and lozenges. 

The face and hair are separated from each other at the sides, 
and often deeply undercut, so that in the better carved arrows they 
form two thin concavo-convex plates with a shaft passing up between 
them. In the smaller arrows the face and hair are separated by 
more or less deep grooves. This interspace is occasionally painted red. 

On the neck there is a marked projection, sometimes very large, 
which represents the thyroid cartilage, "Adam's apple," waiwi rid (W.), 
vjaivKti lid (E.), mango atone; it is rarely absent; in one instance 
another is carved on the back of the neck in addition. The shoulders 
are usually represented by a decorated area, but no shoulder scari- 
fication ia indicated ; I suspect that this decoration is intended to 
denote the sur&ce modulations caused by the aboulder blades. The 
arms are always flexed, but they are absent in a few specimens. 
The fingers are frequently unrepresented or indicated by usually 
five or fewer transverse cuta; longitudinal cuta are extremely rare. 

The chest ia uaually distinguished from the abdomen, the elbowa 
marking its lower limit; the sternum ia shewn by a plain or notched 
ridge. The dorsal vertebne are also usually indicated by a well- 
marked line or row of tubercles, which are more prominent than 
those of the lumbar region when the latter are represented. The 
abdomen is cylindrical or barrel-shaped and decorated with transverse ^ ,^ „. 
lines between which are dots or other simple designs ; very occasionally ^ ^^ ^ ^^^ ^^ 
these run in a vertical direction. In a few inatances the navel ia urom (rom Torm 
represented. A double line round the waist indicates a belt, from Btrwto (Cam. Um. 
which a flat triangular flap projects shghtly in front and usually a^-^' t' ^ *'' 

behind as welL I wae informed that the fix>nt one did not re- 
present the genitals but the pubic shell, in one or two specimens this is clearly tbe 
case; the posterior flap is probably intended for the bunch of leaves or feathers which 
is worn on festive occasions or when fighting. 

Tbe legs are rarely of uniform thickness, the thighs and calves being usually veiy 

H. yoL IV. 24 

d by Google 


broad and the knees veiy narrow; the front part of the legs is always straight. The 
knees are always very prominent; the arrow shewn in fig. 7, pL XXX, is intermediate 
between that of fig. 10 and the usual type as seen in fig. II. Four knee prominences 
are seen in fig. 3; knees are rarely absent. The area between the legs in front is 
filled up with a simple design, and the diamond-shaped or lenticular area behind is 
decorated in various ways. I know of only two specimens in which feet are represented, 
one in our Museum and the other in Oxford (1632). A simple band pattern finishes 
off the carving. 

The foregoing desmption applies in general terms to the majority of arrows of this 
class, but no two are aUke; although there are numerous cases of degeneration, the 
hunum element never becomes obliterated. Sometimes only the head is carved (pL XXX. 
fig. 1), and the rest of the body may be entirely omitted or represented by a simple, 
apparently meaningless design. Some arrows occur in which the head, arms and body 
alone are carved (pi. XXX. fig. 2), or even the head and legs may occur without the 
arms and with little or no body (pi. XXX figs. 3, 5, 6). There is a specimen in the 
British Museum which has a fiice on both sides of the head. 

Dr Uhle* describes three carved arrows in the Dresden Museum, two of which 
belong to this class. He pronounces them "identical," but I find that they exhibit 
the usual variahiUty in design; one of them (No. 6404; fig. 1, la of tJhle) has a 
variation which so fitr as I am aware is unique, since both arms arise from the left 
side of the body and the hands are joined on the right side. A man arrow is sketched 
in the 4lbum, i. pL 267, No. 7. 

(«) Ths Crocodile Arrow. 

The most interesting of all the arrows is the crocodile arrow which is called kodalu- 
paraag (W.), crocodile bee, kodal k^ (E.), and I have heard it called saibri op in Mer 
«at6n being a Daudai name for crocodile. 

In fiY>nt of the main design there are usually a few bracts, much as in the man 
arrows, but these may be considerably increased in number in the more degenerate 
types, or even absent altc^ther. 

It is desirable first to describe the typical crocodile arrows, and it will be necessary 
to call attention to certain well-marked divisions of the total representation : these are 
(i) the snout, (ii) the head and neck (from the eyes, inclusive, to the fore hmbs), 
(iii) the fore Umbs, (iv) the trunk, (v) the hind limbs, and (vi) the tail. In these arrows 
too a simple band pattern generally terminates the whole design. 

(i) The snout is plain; above at the anterior extremity are two elevations, which 
are meant for the prominent valvular nostrils of the crocodile. Occasionally one is 
placed behind the other (fig. 191 a) instead of their being side by side, or even one 
only may be present ; very rarely it is absent. Laterally the jaws and teeth are usually 
characteristically rendered. In one arrow (b) the teeth of the upper jaw on one side 
have been transformed by an easy transition into a zigzag line; teeth are rarely absent 

> "Ueber PMb «u dtr TonwrtrMM," InUrnat. Areh. /. EOmotr, i. ISS^ p. ITS. 

d by Google 



The under side of the snout and head is ornamented with lines uid dots which may 
have a longitudinal or tnuisverse arrangement, or both may occur (b). 

(ii) The head and neck, like the snout, are plain above except for an occafdnsal 
repreeentatiott of scales on the neck (c), and the ventral ornamentation is a continnatioa 
of tiiat of the under side of the snout. The eye is triangular with the apex behind, 
rarely ovat (o), or round ; a band pattern, usually a zigzag, which is always distinguishable 
fix>m the ventral ornamentation, extends from the eye to the fore limb. 


TOW! (Ofunb. Mna, 0. m. H. 8 

(iii) The region of the fore limbs has generally the greatest thickness of the whole 
arrow. The limbs often arise from an ornamental band (a), which repreeents the prominent 
8cnt«8 in this region of the real animal ; the scutes are rarely unrepresented. The fore 
limbe first project backwards, and then run forwards towards the middle ventral line. 
The toee are usually indicated by transverse lines. 

(iv) The trunk has usually a row of chevrons or lozenges running along the 
dorsal and ventral median lines; the lateral ornamentation usually consists of transverse 
lines separated by rows of spots, sometimes these run longitudinally. These three 
patterns indicate the differences between the dorsal, lateral and ventral scales of the 
real animal. 

(v) The hind limbe may be separated dorsally by a triangular area (a), or by 
a row of tubercles (E). They invariably bend forwards and then backwards. The 
enclosed angle contains a row of spots or rarely a plain ridge. I can recall only one 
specimen in which they are absent. 

{vi) ' Typcally the tail is ornamented with three, occasionally two or four, dorstd 
rows of tubercles. The median is a continuation of the median series or the triangular 


d by Google 


area above noted; sometimes the mediaa row is a direct continuatioa of the central 
series on the back of the trunk. The lateral start from the insertion of the hind 
limb. Occasionally the tail is represented by other devices, I have only once noted 
its absence. The proximal portion of the tail of a living crocodile is fbrmshed with 
three rows of scutes. On the under side there ia a large quadrai^ular plate, ornamented 
with concentric lines, the sides of which often extend up to the dorso-lateral angle of 
the tail ; it is rarely absent. 

On comparing a number of crocodile arrows with the animal itself one is atrnck 
with the numerous realistic details which have survived the decorative treatment of 
the design. It must be remembered that one is dealing with a work of decorative 
art, and not an attempt at realistic carving. In one arrow several anatomical charac- 
teristics of the crocodile will be su^featively rendered, in a second other details will 
be more accurately carved, but in the great majority of arrows belonging to this series 
variation has occurred to such an extent that the crocodile becomes almost unrecognisable 
as such. 

In fig. 191, I have drawn half a dozen specimens which I have given to the Cambridge 
Museum ; equally cbaracteriatic examples will be found in many other museums. 

A very typical crocodile arrow is to be seen in a ; the chief variation in this is the 
placing of one nostril behind the other. 

In B, the nostrib are side by aide, and the teeth of the upper jaw are represented by 
a zigsag line. The hind limbs and the tail are entirely absent. 

O is important in several respects. The nostril ia single, the mouth is partially closed, 
but the teeth have not as yet entirely disappeared from the hinder closed moiety. The eye 
is oval — a rare feature — and the dorsal scales of the neck are represented; this is also rare. 
The fore limbs have been converted into a raised zigzag bsnd encircling the arrow. The hind 
limbs do the same, except that the pattern is interrupted in the mediaa dorsal line by a 
double row of tubercles, which represent the prominent dorsal scutes of this region io the 
living animal. The thigh is carved with a curved upper border and a straight lower border. 

There is a gap in the series between c and d; but it is easy to see that the binder 
part of the mouth is closed, and the teeth of both jaws are represented hy different patterns ; 
the front part of the mouth is widely open, but edentulous. The nostril is single. The eye 
has become enormously enlarged, and constitutes what I propose to term an eye-panel ; it 
extends backwards to the fore limb. The pl&in upper surface of the head and neck has 
become much reduced, owing to the encroachment of a double row of spot« on each side. 
The artist mistook the upper for the lower surface when he carved the fore limbs, for it will 
be seen that the toes are above and the dorsal scutes are placed below. Another point of 
interest is the replacing of the central row of caudal scutes by a plain ridge; so &r as I am 
aware this is unique. 

B is a type of a large number of arrows. The front open part of the mouth is quite 
small, and the surfaces of the jaws are scored by oblique lines. The median dorsal plain band 
of the snout is no wider than the lateral bands which indicate the closed hinder part of the 
mouth. In the gape of the mouth an elongated triangle is very generally present; this is 
doubtless intended to represent a tongue, sometimes it is notched. The eye-panels are elongated 
and narrow, and the dorsal median band of the head and neck extremely reduced. The rest 
of the body in this arrow calls for no special mention. Sometimes eyes are carved on the 
dorsal surface of the gaping end of the upper jaw. 

d by Google 


lo the but ftiTow (f) of the series which I figure, the froat part of the month has 
dJsappeftrad, but the hinder part of the head is much the same as in the last arrow. Tbe 
fore limbs and body are abeeat. The hind limbs are narrow, but retain their characteristic 
forward bend; the dorsal caudal scutes are replaced by numerous parallel transverse lines. 

In my Decoratiee Art (^ B. If. 6., p. K7, I have called attention to the danger there is 
in these studies of misinterpretation from the lack of a sufficiently large series of specimens. 

It would be impossible to refer to all the modificatiotiB of this clase of arrow which 
are to be found in museums. The few I have just described indicate the general drift 
of the changes which occur, and will enable anyone to inter{B«t the carving on tbe 
majority of carved arrows from this district. Two features, however, are worthy of special 
allusion, the one is tbe remarkable retention of the projecting nostril, which may often 
be found as a alight prominence in very degraded arrows; and the other is the still 
greater persistence of the tail and hind quarters of the crocodile; I suspect that the 
striking decorative effect of the concentrically marked cloacal plate has led not only to 
tbe retention of that part but also of the neighbouring organs. 

Photographs of twenty arrows are shewn on pis. XXXI. XXXII. ; they are sufficiently 
described in the explanations of tbe plates. 

(/) The Snake Arrow. 

We now pass on to a small group in which the open front part of the mouth 
of such an arrow as fig. 191 £ has suggested a complete head, and so eyes are added 
(fig. 192 ; pi. XXXI. fig, 12) ; the rest of the snout, the head and fore limbs are 
omitted; the body is much elongated, but the bind legs and tail are usually quite 
normal, or subject to merely minor variations; the patterns may run transversely as 
in the figure, or longitudinally. Such a carving irresistibly calls to mind a snake; the 
natives themselves told me it was a snake, calling it elma gudvlaig (like tbe mouth of 
the elma snake) in Tutu and waruwa kep (snake arrow) in Mer^. 

The tail and hind quarters, however, proclaim the crocodilian original. In this 
group of arrows we have a very interesting example of tbe transition from one form 
into another; but hitherto I have not seen a snake arrow which has lost all trace of 
its aanrian ancestry. 

Fia. 193. Snake arrow (Comb. Uob. 0. m. M. 8S], one-third nat. (iie. 
The crocodile arrows may be classed as follows : 

L Complete. 
Those representing all the features of a crocodile with the least amount of modificatiou; 
mouth with teeth along its length, small eyes; all bracteate (fig. 191a, o, pi. XXXI. fig. 1). 
ii. Modified. 
Tlioee in which tbe greater portion of the mouth is closed (except in a very few oases, 
fig. 191b, pL XXXI. 13, XXXII. 6) and the teeth are rarely indicated; eye generally in the 
> Ana ia the MairaU name for a ipedea of snake, and eraaa arva U alio the name of a snake ; elma and 
entva maj be tha same word. 

d by Google 


fonn of a panel (nnall, trlBDgalar in fig. 191 B, pL XXXI. 4, XXXIL 5; a spot ia 
XXXI. 13). 

(a) Flat type : form gener&ily represeDting that of a lisard ; the snout generally ronnded 
and the body di&mond-shftped in seotion; fore limbs very nrely absent, hind Umbo, tail and 
oloaoa preaent; rarely bracteate (pL XXXL 2). 

(6) Shnder, rottnd type (very rarely thick); frequently non-bracteate. 
Head, body, limbs', taiL (Fig. 191 d, i, pi. XXXL 3—10.) 
„ „ no tail (PI. XXXIL 1.) 

„ fore limbs, no bind limbs or tail (Fig. 191b, pi. XXXU. 6, 7.) 
„ toil, no limbs. (Fl. XXXL 13.) 
„ „ no tail or bmbe. (PL XXXIL 5.) 

„ hind Umbs, tail, no ian limbs or body. (PI. XXXIL 3, i.) 
„ „ no tail or fore limfae or body*. (Fl. XXXIL S.) 

Eyfrfwnd, limbs, body, tul, do snoot. (PL XXXI. 11.) 
„ and hind limbs only. (Fig. 191 r.) 

iiL Snake type. 
Those in which the orooodile is so degenerate that it oomes to be regarded as a snake, but 
hind limbs and tail are present; non-bracteate. 

8ton«-headed Olnttt. 

Both Jukes and Macgillivray state that the only weapon other than the bow and 
arrow which they saw in Torres Straits was "the club called gabba-gtx^," the ordinary 
disc-shaped atone club which both aptly describe as "like a quoit"; Jukes adds, "We 
only saw one or two of them " (i. p. 209). Macgillivray says that they are made " of 
hard stone (quartz, basalt, or serpentine)" (ii. p. 19). 

In my "Classification of the stone clubs of British New Guinea" (Joum. Anth. 
Inst. XXX. 1900, p. 221), I drew attention to the great variety of stone-headed clubs 
that odcuiB in British New Guinea, and it is probable that there are other varieties 
which I have not seen. On the whole the clubs of the Central District (Ic. p. 246) 
afford the greatest variety of form with excellence of workmanship. Those of the Gulf 
District are generally not very carefully made, many are extremely rough, and a number 
of unworked or' but slightly worked stones are perforated and rudely hafted. In that 
paper I stated that "hitherto the natural stone clubs, the nnflanged knobbed clubs, 
the triangular, rectangular, and other peculiarly shaped disc clubs, and the large two- 
pointed pickaxe clubs have been obtained only from this district. Flattened ball clubs 
are not rare. A few unpolished ovoid clubs have been collected in this district. All 
flanged clubs are rare. A few flanged star and knobbed clubs have been obtained; 
but the flanged disc clubs are only ' found here. The same applies to the wooden 
knobbed clubs." 

■ The tore Umbi ue frtqnentlf reprBMntod by t, deoorktiTe devioe. 

■ Thia know hu ftlmoit dagenerattd to a bdaIu anow, but the eje puial ia noogniMble. 

d by Google 


When we come to the Fly River and Daudai districts, we find a still greater 
paucity of fbnns, and from these districts the Torres Straits islands cannot be distiognished, 
some clulvbeads appear to have been made in the islands. Owing to thfi absence of 
suitable rock the mainlandere, like many of the islandeis, must have obtained their 
weapons by trade or loot. The various forms of club>bead will be described immediately. 
The universat name for these clubs was gahagaha or g<^)agoba, which was the general 
name, but particular types probably had specific names. According to Mr Ray the two 
disc-shaped stone-headed clubs of Malu were called "Waduli" and "Tamera" (vi. p. 296), 
and those with star-shaped heads were called aauriaauri. 

The bandies, pes (&), of the clubs are rather short, vaiying from about 381 to 
720 ram. (15 — 28} in.) in length. They are usually thick, often of stout ratan, and are 
either cut square below the grip or taper to a more or less blunt point; this end was 
occasionally beaded. 

In the Brit. Mus. (c.c. 6630) there is a disc club, the stem of which beyond 
the bead is carved to represent an animal's head; the grip is served with coco-nut 
fibre string. The length of this club is just over 1 m. (40 in.) ; it is certainly not a 
typical Torres Straits form. Occasionally the head end of the stick may be decorated 
with Abrus seeds set in beeswax. 

A string loop, uru (W.), is generally present, and probably was at one time universal ; 
it was slung over the right shoulder, the head of the club being in front when going 
on the war path. 

The following information was obtained by Mr Wilkin at Mabuiag. All the stone- 
headed ctubs came from Dauan, Saibai and Mer. [I do not believe that they were 
made in the two latter islands, but there may have been a &ctory on Dauan.] A diso 
or star stone-headed club cost one wap or one vmiwi, the star club being stated 
to come only from the Miriam, but more would be paid for a large club. " S'pose you no 
got wap you hungry all the time; s'pose you no got gahagaba by'mby you fight, you 
loee'um life. That's what tor he so dear." 

Flat or biconcave diso heads are common. They are often somewhat irregular in 
outline, but are meant to be circular'. They vary in diameter from about 103 
to 150 mm. 

Fig. 193 illustrates a typical biconvex disc club which I obtained in Tam in 1888. 
The stone is well worked, 103 mm. in diameter and 28 mm. thick. The handle is of 
thick ratan, 706 mm. (27} in.) long, and the grip is decorated with simple incised 
patterns. For a somewhat simitar example see Albam, IL pL 176, No. 7, the head is 
12 cm. in diameter, and the stout ratan handle 767 (30} in.) long, plaited and twisted 
string is served above and below the head; the treatment of the projecting shaft of 
No. 4, is peculiar, it is probably a mainland club. In 1888 I collected at Mabui^ 
a rude club that was said to come from the Tugeri; the stone is an irregular fiat 
disc, 112 mm. (4J in.) in diameter, which was evidently a natural water-worn stcne, 
the edge is blunt, the hole is 8 cm. in diam. and has vertical sides and sharp edges 
as if bored with an iron drill ; the ratan handle is 584 mm. long (cf. Album, i. pL 346, 
■ Tom of Hkbniag informed Ur WilUn tbkt % trungnkr rariet7 wm oalled dUinuut hma, an lutiftQR- 
kUUa word, but it look* aa if it might mMn dolan'* rtone. Ko eumple fa koown to me. 

d by Google 


No, 1). I also obtained In 1888 a beautifully worked club, the atone of which is 
146 mm. in diam., it has a eharp edge and its upper sur&ce is painted with a 6-rayed 
red cross; iron nails are used as wedges, and the projecting end 
of the stiok is ornamented with Abrus seeds inserted in beeswax. 
These two specimens are in the Brit Mns. 

My old friend Maine, the chief of Yam and Tutu (pL V, 
fig. 10), presented to the Cambridge Museum the head of an 
old disc club (15 cm. in diam. and 2 cm. thick, weight 2 lbs.), 
made of lava. It is irregularly circular, biconvex but very much 
flattened ; the hole is oblique, with a diameter of 28 mm. at the 
outside and 22 mm. in the centre. We obtained in Mer a club 
with a short, thick, recent handle, the plano-convex disc head is 
made of fine-grained volcanic ash and coloured black so as to 
resemble the ordinary dark stone of which club-heads are made 
(17 cm. in diam., 29 mm. -thick). I doubt if this was ever a 
fighting weapon, probably it was made for dance purposes. 

Unflanged star-shaped heads are characteristic; some are of 
rather rude workmanship, others are well made with a polished 
surface. The number of rays varies &om four to about a dozen 
(pL XXIX. fig. 3) ; the latter variety practically meiges into the 
class of disc-shaped heads with a notched rim, examples of which 
occur on the mainland, more particularly in the Qulf District. 

Two sacred Malu clubs, aaurisauri (vi. p. 296), are beautiial 
examples of simple unflanged four-rayed stone-headed clubs (pi. XII. 
fig. 2). The rays are fiurly long, bluntly pointed and biconvex 
in section ; the diameter of the head of these specimens is 
28 cm. (11 in.) from point to point. The handle gradually tapers 
to the grip, the total length being 68*5 cm. (27 in.) in one 
specimen and 59 cm. (23^ in.) in the other. The grip end is beaded. When used 
ceremonially the head end was adorned with two white feathers which were split and 
the two halves bent round to form a circle. A similar club-head is shewn in 
pi. XXIX. fig 4. 

On Tarn island I obtained a large four-rayed head (fig, 193) made of white granite, 
which was also used in the ancient ceremonies. The rays are convex above, sloping 
fi^>m an indistinct median keel, and slightly concave below in section. The head as 
a whole is concavo-convex, so that it rests on its four points in one position and upon 
its centre in the other. Its greatest diameter is 224 mm. and its thickness 42 mm. ; 
the outermost diameter of the central hole above and below, is 44 mm. which converges 
to 16 mm. in the centre. 

I obtained at Mer a niffir gahagaba (fig. 194), which may be described as an 
unflanged ball head, with the greatest diameter at an equatorial band, fitim which 
there is a sharp slope polewards. It is 105 mm. in diam. and 97 mm. thick, the oater 
margin of the bole is 37 mm. and the central diameter 8 mm. It is made of n^r 
stone, which is said to be found in Dauan. 

d by Google 


Clulra with ovoid stone heads have been obtained Irom the Fly River, and Tom 
of Mabuiag told Mr Wilkin of a spherical-headed club, gabagaha, and of a crescenttc- 
headed club, malpelau kuma (this is probably mulpcU-au, moon's, 
kuta, stone), but I have not seen examples of these from Torres 
Straits. Partington figures in the Album (II. pi. 176, No. 6) a club 
in the Brit. Mus., probably from the mainland, the head of which 
is a perforated natural pebble of irregular shape. The stone is 
107 mm. in diameter, the total length of the handle is 1-423 m. Fta.l94. Uoflufiedball- 
(56 in.), which is remarkably long for this district. He also figures "h»P*J ^tone head of * 
(Xo. 6) a club, 89 cm. long, with a ball-shaped head of chalcedonous 
quartz 9 cm. in diameter ; the cone at the end of the handle rather suggests that this 
specimen belongs to the Central Pistrict. 

I collected at Muralug in 1888 an unfianged stone head with two rows of knobs. 

I obtained at Mer a curious club, tut (pi. XXIX. fig. 9), made of nigir stone; 
it is 39 cm. (I5| in.) in length and weighs nearly 3 kilos (6^ Iba). It appears to be 
a natural ' stone which was perhaps selected because it had a convenient grip. It was 
formerly used to hit initiates during the Malu ceremonies (vi. p. 311). 

Wooden Clabt. 

Wooden clubs, tutu (W.), tut (E.), are by no means common. They were cut 
from a single piece of wood generally of Mimusops Browniana, the waogai plum, 
ubar (W.), enau or enoa (E.) (vi. p. 6, fn. 6), or kus wood (l.c, p. 27). 

A simple wavy club, (irf, of dark red hard wood (fig. 195) was obtained 
at Mer. It measures in a direct line 525 mm. (20} in.) and weighs two 
pounds. This is the club with which laui killed Boa and Kaidam for 
stealing his coco-nuts, the two marks on the club were cut by Laui as 
a memorial of this deed (vi. p. 190). 

A konoy konor tut made of konor wood (fig. 196 a) is 767 mm. long 
with a greatest diameter of 61 mm. ; the grip is engraved with charac- 
teristic zigzags and there is a narrow beading at the end. Another veiy 
old konor konor tut, also from Mer, has a conical end on which a turtle 
is engraved (fig. 196 b); the length is 545 mm. and the diameter of the 
flat top is 67 mm. ; the grip is broken but probably it was never much 
longer, there is a plaited cane lashing on it. We also obtained at Mer 
a small baton-like club which tapers slightly and has a broad bead at 
each end ; it is 37 cm, long. 

In 1889 I collected in Mer an old, well polished club, panigob', made 
of the hard, heavy, dark-coloured enau wood, 58 cm. (nearly 23 in.) in length 
(pL XXVII. fig. 3; Album, i. pi. 346, No. 4). The head has somewhat 
the form of a bird's head, and an eye is incised on each side, the lines j, ,_, ™^ 
being filled in with lime, the beak is spatulate and plano-convex, the an dab with 
back of the head is marked in the median line with white incised chevrons, tally mukB. 


> It wM called a panigob, axe, probably od aooonnt of its ihape, foi it oonld not have been nsad m an axe. 
H. Vol. IV. 25 

d by Google 



and besidQ it tbere are eight red comb-like markiDf^. White chevrona are cut on the 
upper of the two beads which occur at the grip. This club is of especial interest as 
it and probably the following one are of the type seen by Flinders at Eirub a hundred 
years ago. He says: "Their clubs are made of the ccKuarina, and are powerful weapons. 
The hand part is indented, and has a small knob, by which the firmness of the grasp 
is much assisted ; and the heavy end is usually carved with some device : one h&d the 
form of a parrot's head, with a ruff round the neck ; and it was not ill done " 
{Voy. I. 1814, p. zxiii). Capt. Bampton states that the Erub clubs were about four feet 
long (I.C. p. xxKvii). 

Fia. 196. Old woodeD olaba, Her. 

Another old club made of enau wood (fig. 197) was given to the Glasgow Museum 
. 67 At) by Mr R Bruce, who probably obtained it at Mer. It is 825 mm. (32^ in.) 


Fia. 197. Old woodeo olub, Mer?; QlMgi 

long; the oblong head is 10 cm. long and lenticular in section, it is slightly ornamented 
with engraved designs and decorated with cassowary and bird-of-paradise feathers; the 
handle is round and there is a terminal swelling at the grip. 

Wooden imitationa of stone-headed Clubs. 

I collected two or three specimens which are obvious imitations of the ordinary 
disc-shaped stone-headed club. One obtained at Mabuiag in 1888 and now in the 
Brit. Mus, is cut out of one piece of dark heavy wood; it is 725 mm. (28^ in.) long, 
and the head, 145 mm. in diameter, is a good imitation of a biconvex stone disc 
(c£ Album, I. pi. 346, No. 3). We collected another at Mabuiag in 1898, the disc of 
which is flattened and measures 175 mm. in diameter. We collected at Mer a aekeraeker 
gobagoba, which is a wooden imitation of a star-shaped stone-headed club with six rays 
(fig, 198), and the total diameter of which is about 105 mm. ; it is all in one piece. It 
is probable that all these were used in dances and not for fighting. 

Foreign types of Wooden ChAs. 
On both expeditions I obtained in Mer and Mabuiag wooden clubs which were 
evidently imitations of South Sea forms. 

d by Google 



The more common of these have the head end shaped like an elongated dome 
(fig. 199 c), this ia the ordinary mushroom-shaped club, characteristic of New Caledonia 
and the Loyalty Islands^. At Saibai it was called gorbotut in 1888. Since 1871 Lifb 

Fio. 198. Wooden imitetioD of a star- 
Bhaped atoiK-bwded olnb, Mer. 

^ ^ 

Fia. 199. Wooden clatn of (oTeign tjpea from Torres 
Straits. A, B, Mew Hebridea (nwe {66 and 80-5 cm. 
lonfi). C, Lofalt; Islands type (56 em. long). 

teacheiB have been imported into Torres Straits by the London Missionary Society, and 
other Loyalty Istandeis have followed in their wake. The presence there of this type of 
implement is thus readily explicable. To the same cause is due the occurrence of a 
tjrpical pelican-head club which I obtained in Mer in 1889. It was made by Ned Ware 
in Eruh. This is the Lifu jia (the j^tk in " this "), my informant called it dia. 

Two clubs of known New Hebrides type (fig. 199 a and b) were collected in Mer. 

I The speounen is however not qnite tTpioal, (he end of the grip ought to be like that of A. The same 
objection applies to 6. All the other alabs of tjpe that we collected have the normal handle. Aberrations 
are to be expected in implements made in a foreign land where the control of onstom is absent. 

d by Google 

186 anthbopolooical expedition to t0brb8 straits 


The javelia, kal<^, klak^, was employed solely by che natives of the western islaDda 
from Muralug to Mabuiag; it was thrown by means of a spear-thrower, kobai or ktUmi. 
From their appearance it would seem that many of the javelins and spear-throwers 
were imported from Cape York, indeed Macgillivray states that the Muralug men obtain 
them from the Oudang tribe, and I confirmed this statement in 1888. It is likely 
that these weapons also found their way to Mabuiag, but it is possible that the 
. Mabuiag, Badu and Moa people made many of their own weapons; a specimen in the 
Brit. Mus. from Mabuiag measures l'S7 m. (73^ in.). There is no evidence that their 
use extended northward to Dauan, Saibai and Boigu, or eastwards to Yam, Tutu and 
the central islands, but Macgillivray (il. p. 34) says, " The spears and throwing sticks 
[of Nagir] are perfectly similar to those of Cape York from which place they had 
probably been procured." 

So far as I am aware, this is the only instance in which Papuans have borrowed 
from Australians; the innovation was a wise one, as there was in 1888 a general 
concensus of opinion that the javelin is a more formidable weapon than the arrow. 
I was informed that it generally took three or four arrows to render a combatant 
hors de combat, whereas one javelin usually had that desirable eflFect, and, further, a 
better aim could be taken than with bow and arrows. Again I heard at Muralug 
that in fighting the white man javelins were found to be more efficacious than arrows. 
[According to d'Albertis (l. p. 417) the natives of Ynle Island, New Ouinea, "prefer the 
spear to the bow and arrow, which is becoming obsolete among them."] 

These javelins were the favourite weapons of the legendary Kwoiam (fig. 205), and 
it will be noticed in the legend (v. pp. 71—83) that his antagonists were never 
mentioned as using these weapons which, in the final sentence of the narrative, were 
relegated by general consent to Australia, whence they were derived — indeed •' all he 
did was Mainland fashion." When I was in Mabuiag, Kwoiam's island as they were 
proud to call it, in 1888 a laige number of Badu men came for some " Sports," the 
chief feature of the friendly contests being a match of javelin hurling. The mark was 
a tree stump, 125 mm. (5 in.) in diameter, and the distance was about 40 paces (about 
275 m.). I reckon that about ten per cent, of the javelins struck the stump, some being 
hurled with such force that the points projected through on the other side. The greatest 
distance thrown was about 100 paces (about 98 m.), 

MiicgUlivray (ii. pp. 18, 19) givoa the following information: "The Eowraregas [Prinoe of 
Wales Istanden] obtain bows and arrows from their northern neighbours, and occasionally use 
them in warfare, but prefer the spears which are made by the blacks of the mainland. We 
saw three kinds of spear [kalak, general name] at Cape York ; one [rtuii] is merely a sharpened 
stick used for striking fish, the two others, tipped and barbed with bone, are used in war. 
The principal spear {kal<^ or alka*) [tvn] measures about nine feet [276 m.] in length, two- 

> A vater-apoDt, 5ai'u (t. pp. Bfi, 334, S60), is also called in Habaisg klak ntarkai or spirit*' spesr, it is 
thus regarded ■■ a fighting-BpesT ud not ss a fiab-ipear (p. 157). 

* This may be algadi, the barb of the kalak (the ihsft is called gvapi or jMuf (W.)), bot it may be the 
■ame word as alkir, % Qaeensland name, see p. 198. 

Digitized by VjOOQ IC 


thirds of which are made of she-oak or casuarina, hard and heavy, and the remaining third 
of a very soft and light wood; one end has a small hollow to receive the knob of the 
throwing-Btick, and to the other the leg-bone of a kangaroo six inches [152 mm.] long, sharpened 
at each end, is secured in such a manner as to furnish a sharp point to the spear and a long 
barb besides. Another spear [Coiu], occasioaally used in fighting, has three or 
four heads of wood each of which is tipped and barbed with a smaller bone than 
is used for the kalak. 

"The throwing-stick in use at Cape York extends down the N.E, coast [of 
Queensland] at least as far as Lizard Island ; it differs from those in use in 
other parts of Australia in having the projecting knob \kuhai ngur or kubai pii\ 
for fitting into the end of the spear parallel with the plane of the stick and 
not at right angles. It is made of casuarina wood, and is generally three feet 
[915 nim.j in length, an inch and a quarter [30 mm.] broad, and half an inch 
[13 mm.] thick. At the end a double slip of melon shell \_kubai lal], three and 
a half inches [9 cm.] long, crossing diagonally, serves as a handle, and, when used, 
the end rests against the palm of the right band, the three last fingers grasp 
the stick, and the forefinger and thumb loosely retain the spear. With the aid 
of the powerful leverage of the throwing-stick a spear can be thrown to a distance 
varying according to its weight from 30 to 80 yards, and with considerable pre- 
cision ; still, if observed coming, it may easily be avoided." 

My Muralug informant gave me to understand that there were several 
varieties of javelin or kalak : the rud, or small form with a simple wooden 
point, which is probably the rOda (p. 156), fishing spear, not used in 
fighting i the tun, or large barbed variety; the taku, or pronged javelin 
with barbed points; and the waki, similar to the last but armed with 
the serrated spines of the Bting-ray, waki. He also said that the taku 
was mainly aimed at the side of the neck, evidently to have a better 
chance of severing the jugular arteries, the tun at the back, probably 
because it was the strongest of the three, and the waki at the fix)nt of 
a foe. When imbedded in the body of a victim the gum, ierka (W.)*, 
which surrounds the barb, aXgadi, was stated to dissolve and thus to leave 
it in the wound when the javelin was withdrawn. 

The description by Macgillivray of the spear-thrower is sufficiently 
detailed. The specimen shewn in fig. 200 is 80 cm. long (31 j in.) with 
a maximum width of 4'5 cm. (I j in.), a band of yellow orchid skin, batn^an (W.), Fio.aoo. Speai- 
Dendroblum, is fastened at the lower edge of the gummed head. Abrus thrower made 
seeds adorn each side of the shell handle. This may be taken as a somewhat * , Imnorted 
extreme form of the narrow type of spear-thrower; an exceptionally broad into Torres 
one is shewn in fig, 205 and in Vol. v. pi. iv. fig. 2. 8tr»it«. 

The following remarks by Dr Walter K Roth ("N. Queensland Ethnc^raphy, Bull 13." 
Records of the Austr. Mue. (Sydney), vii. p. 192) are of interest as shewing the wide dis- 
tribution of this type of spear, "The Princess Charlotte Bay, Cape Bedford, BloomGeld and 
^Middle Palmer River spears may be dealt with collectively, with the Cape Bedford ones, about 

' itTlut'^vfXeea, reeiii, '■milk belong wood," used in fliiuR the heads and joints ot epeara uid the peg ol 
the Bpear-thtower, 

d by Google 


which Teij reliable information is known, as the type." Local generic names for spears are 
kalka for the Koko-rarmul (Hinterland and coast of Princess Charlotte Bay), Koko-yimidir 
(Cooktown, Cape Bedford, etc.), Eoko-yellanji (Butcher's HillX Koko-minni (Middle Palmer 
River), and filklr for the Eoko-wara (Hinterland and coast of Princess Charlotte Bay). "Straxtge 
to say, kalka is the generic term for a spear amongst the coastal blacks (Qunanni) between 
the Mitchell and Staaten Rivers." "All the following Cape Bedford spears are made of a 
distal (shaft) morticed into proximal (butt) portion, the one extremity of the barb invariably 
forming the very tip of the completed spear." Then follow descriptions of various spears, 
illustrations of which are given on pL LVIIL 

Various spear-throwers are described on pp. 197 — 201; those on the Pennefather River 
are similar to the kuhai and "are known by the general name of ar&i-i. Used both as a 
spear-guard and as a spear-thrower. The blade (pi. LVIII. 6g. 15) varies greatly in width, 
but with greater width there has, of course, to be a correspondingly larger peg; greater width, 
however, is not considered an advantage. It is manufactured from five difibrent timbers... [the 
construction is then deecribed]. The handle is covered with [Canarium australicum] cement, so 
as to prevent it from slipping through the hand. The cement at one or both extremities of 
the blade may be occasionally decorated with the dried (yellow) strips from the outer covering 
of the 'Rock Lily' {pendnAiv/m bigilAmn, Lindl.) orchid. The shell-haft is formed of two 
oval-cut pieces of pera shell {Mdo diadema, Lomk.) attached with beeewaz, while a few Abnu 
preeatoriue beans may help to ornament the edges in-between ; the angle at which the shell-haft 
is affixed varies a good deal, and appears to depend on individual caprice." The spear-thrower 
at Cape Bedford, on the Endeavour and Bloomfield Rivers and at Butcher's Hill is somewhat 
similar and may be provided with two pieces of Melo shell. The method of fastening the p^ 
to the shaft in the Cape York instrument is similar to that figured by Dr Roth (pi. LVIII, 
fig. 19) from these three places and from the Pennefather River. The Cape Bedford name 
of the implement is mUbir (t.c. p. 199). See also R. Etfaeridge, jimr., Proc. Linn. Soc. N. S. W. 
(Ser. 2), VI. 1891, p. 699, vii. p. 399', viii. p. 299, and F. von Luschan, Bagtian-FetUchriJt, 1896, 
pi. X. fig. 6. 

With the exception of javelins, speara do not seem to have been employed aa ordinary 
weapons, but on occasion fish-spears (p. 156 ; V. p. 71) or even the dugong harpoon 
(v. pp. 15, 21, 93) might be used for killing persons. The Miriam zab is stated in 
the Vocabulary to be a war spear, but probably this is merely a fish-spear, dot (p. 156), 
and the kaigob, "spear, javelin," is probably only a large arrow with a bone point — 
such arrows are frequently termed "spears" by white residents. In a folk-tale (v. p. 45) 
we read of a simple toy spear, duktin, made of the hard dukun wood, and we were 
informed that a k&ca was a long stick with a sharp point that was used in fighting 
(v. p. 94). On the other hand. Flinders quoting from Bampton's MS. Journal (1793), says 
that in addition to bows, arrows and clubs, the Enib natives bad "spears and lances of 
various kinds" which were "made of black, bard, wood. Some of the lances were jagged, 
fi^>m the sharp point to a foot upward; and most of them were neatly carved" (Voi/. L 
p. xxxvii). This is the only record I have seen of a type of spear that is known in 
New Guinea and Australia. Dr Rutherford saw "long wooden spears" in Mer in 1833. 

' Etb«ridge tAja: "1. Uth-8liBp«d, mouiited vith Melo. Cape York; Agate Creek, Oilb^ BiTsr; C*f» 
OreDTillei Herbert River; BftUvia Biver, Onlf of CarpentarU" (p. i(t2), and l.c. p. 170, this wotnerah "eit«iid( 
throQgboat Cape York Peninsula ; tbe Oilbert Biver being sitnated at its eiticme baae." 

d by Google 


Beheading knife and head-carrieb. 

The bamboo bebeading knife, yjn (W.), kwoier (E.), was said to be aimilar to tbe 
ordinary bamboo knife, but I do not know whether the same knife was actually used for 
bebeading and for domestic purposes. 

An ordinaiy beheading knife (fig. 201, pi. XXIX. figs. 5 — 8) is cut out of a stout 
bamboo, and averages about 37 — 40 cm. (144 — ^H i"-) l°°g- The handle is about 
14 cm. long, and is made by putting a piece of 
fibrous pith (?) into the concavity of the bamboo 
and firmly surrounding both with a lashing of 
twisted Btring, perhaps of luali. The string is 
knotted at intervals in such a manner as to form 
raised zigzag cordings which run along the length 
of the handle and serve to give the hand a firm 
hold when the knife is reeking with blood. 

I was informed that the edge of the knife 
remained sharp only long enough to cut off one 
head, consequently a fi^esh edge must be made for 
each decapitation. This is done by cutting a notch 
at tbe handle end of the edge and splitting the 
opposite end with a piece of quartz, uz (v. p. 71), 
or a shell, in such a manner that a narrow strip 
is peeled off the old edge, this was often done 
with the teeth. One result of this method of 
sharpening the knife was that there was a notch 
for each new edge, and the series of notches was 

necessarily a tally of the number of persons de- F">. 201. Bamboo bebawlmg koifc and ba^- 
capitated by means of that particular knife. The ^^= J^ '"^ ^ «"^ ^"8- ""*'^ 
greatest number of notches that I have seen is 

eleven, but it must be remembered that as knives could be used for cutting up dugong 
and other purposes each notch may not always represent a bead cut off. It is probable 
however that roost of the genuine knives in collections are bebeading knives, as these were 
kept as treasured possessions by the natives whereas the ordinary knives would not be 
preserved. Further, most of the knives are associated with the cane loop, and when this 
is the case all doubt may be set at rest. Figures of the two implements are given in 
Jukes, Vol. 1. p. 277, and Album, i. pi. 336, Nos. 1, 2. 

The head-carrier, aingi or sungei (which averages about 345 to 400 mm. long), consists 
of a loop of cane or ratan, the two ends of which are lashed on to a cross piece of 
wood about 15 cm. long. Besides the direct lashing of the loop there is frequently 
a supplemental lashing of braid or string of coco-nut fibre (fig. 201) ; the former is alone 
present in the illustrations in Jukes and the Album, Very frequently the cross-bar is 
composed of an old harpoon dart ; in a specimen which I collected at Tutu in 1888 each 
end is carved to represent a human face (cf. Album). 

d by Google 


I was informed that the loop was passed in at the neck and out through the mouth 
of the decapitated head. 

Sir William Macgregor, in liis account of the natives at the mouth of the Bamu River 
(^nn. Report, C. A. I. 1893, p. 63), gives the following description of the use of these implemente : 
"When going into action they cut out with a small clam shell a notuh about a fourth of an 
inch deep in the edge of the bamboo at the spot where the blade joins the handle, and they 
make a slit in the upper end about two or three inches long in such a way that when the 
enemy falls pierced by an arrow they can at once catch hold of the splinter of bamboo, tear 
it off down to tiie notch, and thus have a sharp, clean-cutting, bevelled edge. The same edge 
is never used for two heads, and knives can be found that have cut off as many as seven 
or eight heada When the head is cut off the loop is brought through the fioor of the 
mouth until the toggle reste on the lower jaw. It b then carried homo by the loop. The 
kwuere and garaora seem to be the most valued heirlooms of the old families, at leant in 
many tribes." 

The Rev, J, Chalmers in a MS. states that at Tureture and Mawata a bamboo knife, 
were, and sling, garaoro, are carried round the neck when going to fight, and when an enemy 
is killed the head is cut off with the former and the sting is inserted under the jawbone to 
carry the head home. The head is hung over a fire and all the hair singed off. While this 
is being done all the young girls of the vill^;e assemble near the fire, join hands, and dance 
in a ring near to (but not round) the fire, and with singing, nekede, the head is taken and 
all the flesh removed ; after the skull has been washed it is hung in the house. A notch, nepiri, 
b made on the were for each head cut off. A lavori e ipa shell is used to take a strip off and 
to sharpen the uteri. 

Decoration for Wartare. 

When going out to light the warriors decorated themselves in order to produce 
a fearsome appearajice. Bed paint seems to have been the main colour with which the 
skin was painted, and there is no doubt that it bad a special significance in this respect. 
We read, for example, in the folk-tales that when a Western man wanted champioas to 
espouse his cause and to attack his enemies, he put some red paint in the kwod (the 
ceremonial ground of the men), and those who took some and rubbed it on their bodies 
thereby signified that they were prepared to fight for their friend (v. pp. 21, 43). 
I understand that there was no uniform method of painting among the Western Islanders. 
Maino of Tutu described a koi gerkital garka, " big fighting man," as having his face, 
body above the belt, and upper arms above the mu»ur painted red, the remaining 
portions of the body and limbs being painted black. From information obtained in other 
western islands I gathered that the entire body might be painted red, or only the upper 
portion and the legs below the knees, or the head and upper portion only. Probably the 
rest of the person was painted black in most cases. The Miriam painted with yellow 
ochre a cross on the face, one line extending down the forehead along the nose to the 
chin, the other running across the face at the level of the eyes. 

Some of the accoutrements of a warrior (fig. 202) consisted of objects of ordinary 
attire, such as the crescentic pearl-shell ornament, arm-bands, leglets and belt; otbera 
were articles that were worn in war and other dances, for example, feather head-dresses. 

d by Google 



croesed Bhoulder-belte, bow-atring guards, groin-shields, aad bunches of leaves (frequently 
crotoD or dracsna) or of cassowary feathers inserted in the belt at the back. Mr Wilkin 
gives in Vol. V, p. 311, an interesting description of the 
Mabuiag warriors accoutred for a fray. 

Perhaps both kinds of the white feather head-dress, dSrx 
(p. 37), were worn in warfare, but I am under the impression 
that the one in which the framework was provided with a 
solid front was more particularly a war head-dress ; as a 
nde the decoration of feathers was less elaborate in this 

Cassowary head-dresses, daguf, were most generally worn 
(p. 36); these sometimes bore a central plume of bird-of- 
paiadise feathers. The most notable feather war head-dress 
I collected is that which belonged to Kebisu ("Kabagi"), the 
chief of Tutu, which was given me in 1888 by his son Maino 
on the condition that it should go to the British Museum 
that "all men may see it" (pi. VII. fig. 3; and AU>um, I. 
pi. 339, Noe. I, 2). It was called baixb, a word which means 
" eyebrows " or " a rain cloud." A somewhat similar double 
head-dress was one of the Ewoiam emblems (figs. 203, 205) 
which were worn by the two head men of a war-party in 
Mabuiag (v. p. 372), but in this case the one with vertical 
plumes was called boibu (boip, baib or baiib), and the depending 
one ear, " branch " or " bough." The &ot6w was kept doubled arrow, 
up in a case of tea bark, ufru (Melaleuca). 

Frontlets of tu (p. 35), that is of the etiolated sprouting leaves of the coco-nut palm, 
were worn in war dances and presumably also in warfare. 

Tu was also employed for making crossed shoulder-belts, 
armlets, leglets and anklets (pp. 52, 59). There was 
traditional authority for these accoutrements, as Ewoiam 
always put them on before killing people (T. pp. 71, 76). 

The left forearm was encased in a kadig (W.), kadik (E.), 
adigo (Eiwai). This is an arm-guard or bracer which protects 
the arm against the recoil of the bow-string. Probably when 
going out to fight a spare bow-string was frequently carried 
doubled up in the bracer (pp. 56 — 58). 

The bow and spare arrows were held in the left hand. 
A stone-headed club was generally carried, which was 
suspended from the loop over the right shoulder, with the 
head in front, when not held in the hand. A beheading 
knife and cane loop were slung round the neck and hung 
down the back. 

The groin-shield, alidan, Idda (W.), alida, ebeneop ("in front of penis") — sometimes 
this is pronounced ebeneaup (E.) — vedere ere (Eiwai), is made frx>m the outer whorl of 

H. Vol IV. 26 

Fio. 209. Dmring of a wurioT 
by Umdo of Tnta, Mune siie. 
He U wearing a ouBowory 
leather hMd'^reu, mai, txm- 
lets, kadig with aomething 
etaok in it, belt, lonUt, Inlets, 
anUetB, and oarrleB k bow and 

Fio. 70S, Drawing b; Oizn of 
MabnUg of a boi&u in its tmse ; to 
the rigbt are the etringa for tjing 
on the head-dnM. In the middle 
of the boibu ia some hail of a 
young Dum, (his makee a jonng 
man oome to be killed by the 

d by Google 



a very old melon or bailer shell (Melo diadema), of which the outer sur&ce has either 
become white or of which the outer coloured layer has been scraped ot£. They are 
triangular in shape, but the form varies considerably (fig. 204), all are more or 
less concavo-conveK bom side to side. A short distance bam the upper border two 
holes are pierced, into which a string is passed which is tied on to the boat of the belt 
in such a manner that the shell protects the sexual organs. The shield may be plain, 
but most frequently its upper portion is decorated by means of horizontal lines, in the 
intervals between which are various decorative devices of a simple character, mainly 
consisting of lines or dots. Frequently also the design includes a central downwardly 
projecting triangle. OcxMsionally the ends of the string that appear on the outer sur&ce 
are furnished with a tassel of coloured wools or even with large beads. The shields 

Fio. 304, aroiI)-Bllielll^ Mer, i nftt. aice. 
average in length bom about 21 to 25 cm. (8J to 10 in.), the extremes being 178 and 
295 mm.; the breadth averages from about 13 to 14 cm. (5 to 5^ in.), the extremes 
being 11 and 15 cm. When going through the bush on a fighting expedition the 
shield was pushed on to the hip so that it might not form an impediment when 

It is probable that part at least of the decoration of a warrior had a significance 
to which we apply the term mt^cal, in addition to the world-wide motive of making 
a brave show. I was however unable to discover any objects that were commonly worn 
which could be definitely stated to be " magical " in their function. The following special 
objects may &irly come under that designation. 

Two crescentic objects of turtle-shell (fig. 205), which according to the legend were 
made by Ewoiam (v. p. 70), had such distinctly supernormal qusdities that they were 

d by Google 


termed augwd, which ia the- name by which totems are called; references to them will 

be foand in the saga of Kwoiam (Vol. v. pp. 70—76, 79), and in the account of the 

feud between Mabuiag and Moa (v. pp. 310, 311, 313). The moral valne of the ai^vd 

in war was very great and the natives themselves recognised it, as a Mabuiag man 

said, " S'pose we no got aagud, how we fight ? " • 

On one occasion (l.c. p. 313), the victorious 

Mabuiag men refused to fight the Moa men on 

accoant of the temporary absence of the two 

aagud men. The Moa men also had magical 

emblems associated with Kwoiam, but they were 

not considered so efficacious as those belonging 

to the Mabuiag men (v. p. 372). Each avgud 

was worn by the head man of a war column 

together with the boihu and zar (p. 201). 

The use of boars' tusks for ceremonial 
purposes and as armlets was mentioned on 
pp. 50, 55, but they were also employed aa 
accoutrements in connection with war&re. Al- 
though I have no definite information on this 
head, one is justified in assuming that, as in 
other parts of New Guinea', they had a magical 

significance, probably being worn to give strength p^a„ D»win«otK-<.lMn\.yWwi»otM.b<i»g; 

and courage to the warrior. They were known he weus the augtid on Mb ohest, on hia head ue 

as gi (W.). gir (E.), both words meaning a boar's *« *"'*" *n^ "*''■ *>« ™"» " ^^" ^^ •honlder- 

1 belts, he bokia a throwing-stiak uid ■ thrae- 

pionged JBTeliu Bimilar to the one with whioh he 

PI. XII. fig. 4 A represents two tusks lashed ^m^j ^w mother (y. p. 71). 
together below, fium which spot a bunch of 

fibres projects and a long cord of twisted human hair, a two-stringed tassel of halved 
his seeds and white beads depends from the point of one tusk and one of white beads 
from the other. It was called gidang, boar's tusk, and held in the mouth when on the 
war path in such a manner that the points projected upwards and the hair string hung 
down. It formerly belonged to Kebisu of Tutu (p. 201), (cf. Album. I. pi. 339, No. 3). 

I also obtained at Tutu in 1888 an amulet (pi. XII. fig. 4 b), that was carried 
in the mouth when fighting, but on other occasions it was worn as a pendant. It 
consists of two tusks joined together at their base by plaitwork, to this is fastened 
a flap of bark cloth, a tassel composed of strings of halved kua seeds with a feather at 
the end of each, and a double cord of plaited human hair, the two strings being 
bound together at intervals by lashings of yellow orchid skin (cC Album, i. pi. 340, No. 5). 

At Mer I obtained a specimen composed of four very fine natural tusks, the bases 
of which were lashed together and covered with red calico (fig. 206). Probably it was 
held in the mouth when fighting, but judging from the two strings it was also worn on 

1 J. Eohnei, " Notea on the Beligioaa Idesa of the Elema tribe of the Papuan Onll," Jovm. Anih. Inaf. 
xsxii. IMN, p. 437. 


d by Google 


the cheat, in which case it would hang in a different position. One pair of teeth measure 
157 mm. across and the other 142 mm. 

A chest pendant fivm Mer (fig. 207) is composed of five pain of tasks fastened to 
a bar 255 mm. long, at its upper end is a loop for sospension and at the lower is 
a tassel of calico and bark cloth. 

Via, 106. Amolet of hnu tnalu, Her. 

An. 907. Amulet of (sn tniks, Her. 

All these specimens were imported from New Guinea, aa there are do wild boais 
in the isUnds. 

When fighting the Miriam wore a whole trimmed pearl-shell, kemerkemer mai, on 
the chest (pL VIIL fig. 2). Mr Wilkin was informed that the Mabuiag men on such 
occasions wore pearl-shell ornaments, nun', on their chests inscribed by means of sharks' 
teeth with the owner's totem (of. fig. 60). In all the drawings of warriors by natives a 
large deep crescentic nun is indicated. 

d by Google 


Tbansport by land haa always been effected solely by eimple canying, mainly done by 
women. Baskets are used for canying small objects, but two baskets are never fastened 
one to each end of s cartying-pole nor is any other device employed, except a plaited 
strap for canying firewood (p. 69), and probably most of these are imported from Daudai. 
When a heavy load is carried on a woman's back the strap is passed across the vertex 
of the head. There are no vehicles, nor are hammocks or similar contrivances known. 

Definite roads were not constructed, but footpaths came into being when there was 
sufficient traffic. No necessity arose for the construction of bridges. 

Transport by water was efficiently performed by canoes ; rafts and floats were unknown. 

The Canobs of Torres Straits. 

Oanoes of large size {pis. XXIV. — XXVI.) were formerly used for fishing, trading 
and fighting. There were no special war canoes. Small canoes were and are used by 
the women to go fishing on neighbouring reefe. The laige canoes are still used, though 
now some natives own, or have a share in, an ordinary fishing lugger ; these they employ in 
pearl-fishing and dagong harpooning. Many canoes have proper names which in some cases 
are now painted on the gunwale, but I do not know whether giving names to canoes was 
an old custom. Sailing by night was very rarely attempted, but the natives had definite 
ideas of steering by the stars when occasion arose (see Astronomy). 

The large cuioes of the Torres Straits Islanders of former times must have been 
very imposing objects when painted with red, white and black, and decorated with white 
shells, black feathers, and flying streamers; and not less so when quickly paddled by 
excited noisy naked sav^es adorned with cassowary coronets and shell ornaments, or when 
swiftly sailing, scudding before the wind with mat sails. 

DumODt-IVUrville {Voy. au P6le »ud, IX. 1846, p. 235) saw about thirty oanoes when 
on Toad (Tutu). One was more than 10 metres (33 ft.) in length, hollowed out of a single 
tree. All were ornamented with rude carving, the prow of one representing "an old man 
with a long beard of fucns." The Tutu canoe figured in pL 190, Alias piU., is of the usual 
Western type both as regards rig and decoration. What appears to be a porpoise is painted 
on the starboard bow; the decoration in front is typical. The sides of the central platfonn 
are bnUt up with bamboos horizontally placed, the upright poets which support these are 
steadied by oblique poles which, starting from the middle line of the platform, project upwards 
and outwards; no crates are indicated; two flags are drawn. At the stem there is a large 
stem-poBt with a Mnge and two flags, dadu. 

Flinders gives the following account of the Hiriam canoes as he saw them in Sept. 1792 : 

d by Google 


"Their csnoee art about fifty feet in length, sad appear to have been hollowed out of a 
single tree ; but the pieces which form the gunwales are planks sewed on with the fibres 
of the cocoa nut, and secured with p^s. These vessels are low, forward, but rise abaft; 
and, being narrow, are fitted with an outrigger on each side, to keep them steady. A raft, 
of greater breadth than the canoe, extends over about half the length ; and upon this is 
fixed a shed or hut, thatched with palm leavee. These people, in short, appeared to be 
dextrous sailors and formidable warrioura ; and to be as much at ease in the water, as in 
thdr canoes" (Flinders, i. 1814, p. xxiij). Captain Bampton mentions canoes fifty to seventy 
feet in length, some of which were "ingeniously carved and painted, and had curious figures 
at each end." 

The account in the Naat. Mag. (vi. 1837, p. 754) repeats in the mun that of Flinders, 
with the following additions: the canoes are "so narrow that the men cannot pass each other 
without crawling between tJieir legs, in tjie bottom of the boat.. .with a fresh breeze they 
are obliged to stand ont upon the outrigger to windward, to keep the canoe upright In 
pulling they had no chance with the schooner's boats, in consequence of the stem and bow 
being encumbered with matfi, which hang into the water. These mats, called aoo goo, are 
made of the young leaves of the oocoa-nut split into shreds [p. 215]; the sails of the canoes 
ar« made of the same material. 3ome have the head carved with the figure of a man, ornamented 
with strings of cowries." 

Judging from the excellent drawings of H. Melville (Jukes, L pi. facing p. 169; Melville, 
SkettAat in duttralia, pis. XYII., XIX., the last is reproduced in pL XXIY. fig. 1) the 
Eastern canoes were precisely similar to the Western craft. Less satisfactory is the drawing 
by W, Westall, the artist to the expedition commanded by Capt. Flinders of H.M.8. 
Tnvealigalor, 1801—1603, reproduced by R. T. Fritchett (Pen and Pencil Sketchet of dipping 
and Craji, 1899, p. 211). A small, rudely made Miriam canoe is shewn in Vol. vi. pL XXVI. 
fig. ti (taken from Jukes, i. p. 133), A canoe seen at Erub in 1887 is figured on p. 41^ of 
Tht Lait Voyage, Lady Brassey (1889), the illostration on p. 423 is wrong in every detail 

MacgillivTsy says: — " The largest canoes which I have seen are those of the Murray and 
Darnley Islanders, occasionally as much as sixty feet long; those of the Australians are 
small, varying at Cape York between fifteen and thirty feet in length. Even the Kow- 
raregaa have much finer canoes than their neighbours on the mainland; one which I 
measured alongside the ship was forty-five feet long and three and a half in greatest width, 
and could cany with ease twenty-five people. The construction of a canoe in the neighbour- 
hood of Cape York is still looked upon as a great undertaking, although the labour has 
been much lessened by the introduction of iron axes, which have completely superseded 
those of stone formerly in use. A tree of sufficient size free &om limbs — usually a species 
o{ Bombaa: (ailk-cotton tree) or Er^thrina — is collected in the scrub, cut 'down, hollowed 
out where it falls, and dragged to the beach by means of long climbetB used as ropes. 
The remaining requisites are now added" (ll. pp. 15—16). He speaks of the poles of the 
outriggers as being fourteen to twenty feet in length, and from sis to ten feet apart. There 
is a long float of light wood on each side, pointed and slightly turned up at the ends. 

At Sue Island, Waraber, Macgillivray saw a canoe which "was rather singular in form, 
with greater beam than I had ever seen in one,.nor did the sides tumble home as usual; 
the bow was sharp, but the stem square, as if effected by cutting a very \b.t^ canoe 
in halves, and filling up the open end " (iL p. 40). In tjie Kwoiajn saga (v. p. 76) it is stated 

d by Google 


that the Gebeu- men cut a new canoe In half to make it more serviceable for Kvoiam ; 
a short canoe of this kind was called gabo, perhaps becatise the cut end would require to 
be fitted with a gab to make it seaworthy. A canoe of this sort is called pau or pao 
by the Miriam, this name was given to the canoe mentioned in Vol. vi. pp. 16, 2&. 
A small canoe in Mer is called waaar. In the Western folk-tales other forms of canoe are 
mentioned, these are : guguha wake, kim and karar mad, a straight canoe (v. p. 29), gaho, 
a canoe which has been cut down (v. p. 75), kauta, one side of a canoe that had been 
split in half (t. p. 104). 

Although canoes were locally made in the Cape Tork district, in the Prince 
of Wales group and at Nagir, as Macgillivray informa us — and I too have seen a small 
canoe which was made by a Muratug native-T-this I believe was only occasionally done, 
and those there made were probably of small size. There is no doubt that all the large 
canoes are and were obtained &om New Guinea. The details of this trade are described in 
Vol V. p. 296. The hulls were hollowed out in the vicinity of the Fly River and fitted, I was 
told, with a single outrigger, as they are only used for river navigation (pi. XXXIX. figs. 1, 2). 
If a canoe was traded to the most westerly islands by the Saibai route, it was refitted 
with two outriggers, and the original gunwale (if there was one) was removed by the 
Saibai men and a more seaworthy, one put in its place ; an attempt at decoration was 
also made (v. p. 296). The figure-head, d^gai^ (pi. XXV. fig. 4), was festened on and 
other bow ornaments, together with white shells and cassowary feathers. The canoes were 
further ornamented by the later purchasers, as they used to pride themselves on their fine 
canoes and the Saibai decorations, having a purely commercial significance, were rather 
scant Further details regarding the ornamentation of canoes are given on pp. 213 ff. 

Id the following description I have given the Western names for the parts of a oanoe, 
the Eastern names are given in brackets (cf. pis. XXIT.—XXVI. and fig. 209). The bow 
is called bttai {larim) and the stem kttn (kor). The hull, gar {g«m), of the canoe, gut {nar), 
is oat ont of a single tree-trunk, the ends gradually slopiug up and coming to a blunt point, 
that at the bow is called ngaaa {larim garbad or op). The sides are generally heightened 
by a gunwale board about 101 mm. (4 in.) in height, garbad (the gunwale, or at all events 
the front end of it, is called bag, cheek, by the Miriam. They call the upper edge of the 
canoe or gunwale maunur, but very few if any of their canoee now have a gunwale). The 
smoothed lower edge of this is laid on to the strught edge of the hollowed hulL A split 
bamboo, mavmau (tarob), is placed rind outermost against the joint, and the gunwale Is lashed 
on to the hull by string, uru (ied), which pasaee through holes, twa (ne6), previously bored 
opposite one another in the gunwale, and in the upper edge of the hull ; a long triangular 
weather-board, tabi {loerem sab), is similarly added to the gunwale at the bow. A vertical 
flat end-hoard, gab (garbad)', is inserted in the bow and at the stem, kuna gab (kor garbad). 
The front part of the sabi has a small deck-covering, auxtr (lak aop), which is supported by 
a croes-bar, gvb (gob). & hole is bored in each saH through which a stout stick is passed, 
that end which projects on the starboard side is called gudagvh and serves for the attachment 
of the anchor rope. The central platform, natar {tarn), and outriggers are next added. Four 
bamboo poles, 3-66 — 4-57 m. (12 — 15 ft.) long, constitute the thwart poles, tug (tug), of the 

' Mr J. BmM intorms me that the Sigai figure-head is not Miriam, but belongs to the veet. 
* Mr Btom aayi that tli« garbad, bag Mid totrem tab tre ooUeotiTol? termed uni irltti op by the Hiriam. 
]t wu on this that M»ln (or Bomai) floated to Mer, see Vol. vl p. S7. 

d by Google 


outrigger, and at the same time form the framework of the platform. Two of the poles, 
from about 1-07 m. (3 ft, 6 in.) to about 1*83 m. (6 ft.) apart, project a foot or two on one 
Bide of the canoe and stretch out some 3-05 — 3-66 m. (10 — 12 ft.) on the other, and the other 
two are similarly placed on the opposite side; the front poles are called buai ttig {ta/rim tvg) 
and those behind kuna tug (kor tttg). A double-pointed float, »aima {nrib), aboat 2-44 — 3-66 m. 
(8 — 12 ft) long, made of the light wood of the pasei tree, is fastened on to the end of each 
pair of thwart poles; the ends are often gently turned up, and frequently the upper surface 
of the float is slightly raised at the spote where it is fastened to the thwart poles. Two 
pairs of sticks, saiu pat (ka^), spring like a V from each end of the float and embrace the 
pole, to which they are securely fastened with string. The platform, ruUar (tarn), is made 
of lengths of bamboo, tabu puU, which run transversely to the length of the canoe. Each 
side of the platform is bounded by a peculiar kind of crate or wattled basket, kutii (aat), 
built on to the platform. It consists of two rows of short vertical sticks, the front ones 
are (»lled kniktt miii, and an outermost row of long ones, taiil, occasionally 1-22 — 1'52 m. 
(4 — & ft.) in height (usually they run much shorter now than formerly). Long sticks are 
woven between the uprights, and the ends are also enclosed. Thus two long narrow receptacles 
are formed along the outer edge of each side of the platform : the outer one, uMtarau lamid, 
IB the firewood compartment as its name implies; the inner one is divided by partitions into 
three compartments, otn^ tamvi, "food compartments": the front one, buai lamul, is where 
the "mate," buai-garka, keeps his food, the middle, dada tamul, and the hindmost, kuna 
tamul, contain the food of the crew'. The "captain" keeps his food in the stem of the 
canoe. Bows and arrows were formerly kept in reodinees by being placed on the kutit. 
Projecting obliquely upwards and outwards from the ends of the crates ai-e two sticks, adaka 
taiU, which support the bamboo poles, Auru (imiU), which are used for punting the canoe 
when on the nref or in shallow water, they are thus stored out of the way of the crew. 
When on a voyage a fireplace, mui lam (ur memeg:), is kept on the platform. 

Sometimes an awning {mud moder, under mat) was built over the platform ; it had a 
mat roof and was supported by sticks which were fastened to the crate. 

A pair of crosa-tiee, datni or duam {zerem or xirim), strengthen^ the middle of the canoe. 
The paddles, kaba (wt«f), are about 1-5 to 1-85 m. (5 — 6 ft.) long with an ovoid or elongated 
oval blade, they are very clumsily made of paaei wood, and without any oroamentation, except 
in some cases a simple beading at the end of bhe handle (pU. XX lY. fig. 1, XXVI. fig. 2, 
XX. fig. 2). A large flat board or a large paddle is used as a rudder, kuli {korixer or 
hor-uxer), at the stern on tbe windward side when sailing. The anchor, iadi (par), is a large 
stone attached to a hawser and kept in the bow. 
A flat oblong perforated stone (vi. p. 42, pi. I. fig. I) 
is said to be the stone anchor of a canoe that came 
in search of Bomai. " The cable is made of twisted 
olimbera— often the FlagdUma I«diea" (MacgiUivray, '^«- ««■ ^V*-^" "' "«»-'"" n«d « s bwler, 
n. p. 16). The stem of the Queensland bean, Entada **^'' ""■ " "^^ ''^■ 

scandens, lirip or tirtb, is used for canoe cables as it is particularly strong (vi. p. 47). 

Cordage and other gear are kept in tbe crates. The shell of the Melo diadema, alup 
(««r) (fig. 152), of which the columella has been removed, is generally used as a bailer, bat 
the spatbe (gem) of a coco-nut palm leaf (fig. 208) is often so employed (huapi). To bail 

' Then do not appear to be an; names among the Mir ia m tor th« oomportmenta, tale ntbgt, of tbe eiate ; 
a modal of a crate is deaoribed in ToL ti. p. 378. 

d by Google 


irater out of a canoe is called sal pamai, lit. dig bilge-water (um depaupti, um is urine or 
bilge-water). Mats are often placed on canoes when on the beach as a protection against 
the S1ID. 

The Bails, toa/cu or gulngu teaku (moder), are large mate; Macgillivray (ii. p. 20) states 
that "the large mats used as sails are made by the women from the fallen leaves of the 
pandanus." They are about 3-66 m. (12 ft.) in height and about 1-62 m. (5 fb) in width. They 
are supported along their sides by two long bamboos, vxtku tag (naitoe), to which they are fixed 
by numerous skewers. A ring or gromet, guguba {gogob), is attached to each side pole between 
one-half and two- thirds of the way up, it enables the sails to slide up and down the 

A canoe with one sail, teaku (narbet moder), is rigged {»e$eri derem) as follows : A step 
or shoe, lira (ti/r)', is placed in the bow immediately behind the awar, in this are erected a 
stout mast, rangad or rod (morgober or morgobar), which slopes towards the port or windward 
side*, and a simitar mast, karaa {karaa}', which is approximately vertical or slopes to leeward. 
The two masts are kept from diverging too far apart by a cord, lumtdam {alakobi lager), 
which is fastened by its ends to the rangad and loops round the karat. A guy, gauxU uru, 
" rope of the gau," passes from the head of the rangad to the outer end of the fore pole, buai 
tug, of the port outrigger; but sometimes, instead of being tied to the end of the buai tug, it ia 
fastened to the end of a pole that projects to windward in front of the port outrigger. This rope 
is in charge of a man who stands on the buai lug, and steadies himself by means of a pole 
which is fastened to the crate. This man is called lugu-kwiku^arka or gauau-garka, Le. the 
chief man of the tug or the man of the gau or windward outrigger ; perfa^is the term gau 
should be restricted to the pole just mentioned, the "temporary outrigger" of Macgillivrey. In 
the Miriam canoes this pole (jfrau), which is made of mangrove wood and is about 3*66 m. 
(12 ft.) long, is a constant feature; it is fixed between the two masts and projects to windward 
on a level with the deck. It is used for balancing the canoe, a man stands on it and shifts 
out or in according to the force of the wind. A stay, nmgadal uru {morgober lager), passes 
from the head of the port mast to the base of the hind pole, kuna tug, of the port out- 
rigger, central to the crate; this is manipulated by a man, amvrgarka, who stands on the 
platform. Another stay, karaail uru (karaa lager), passes from the head of the starboard moeti 
karas, to the base of the bind pole of the starboard outrigger, kuna tug, and ia held by one 
ol the crew, paaara, standing on the platform. In setting sail, the two backstays, rangadal 
uru and karaail mtu, are passed through the gromets of the sail, which is pushed into position by 
means <^ two bamboo poles with forked ends; when the soil is home, that on the port side is 
retained, the fork fits into the gromet and the pole, parungaizinga (nariiet akmeret lu), props 
up the sail against the masts, its lower end being lashed to the port bow about halfway between 
the bow and the outrigger. A kupal uru (tMl rope) or sheet is attached to the starboard lower 
comer of the sail and made fast to the front thwart of the platform. 

When there are two sails, as is usually the case, the second and smaller one, dada t*tutu 
(ieimer moder), is placed close behind the principal sail. A third mast, pai^xt tdrai*, is erected 

> The t(T toi snpportiiig the lower end of th« matta, karat and morgober, ii a piece of wood with four 
hotel in two of which th« feet of the masts {latri) aie itepped, the other two are spare holee in cue of 
braaksge, A wooden i(ep, $aua peire, with two square boles, from the mouth of the FI7 Biver, ia figured in 
the AUnm, n. pL 194, it is SB em. long. 

> Thii ia placed to wiudwaid, becaDse "rad he Btrong." 

* Morgober is made of mangrove wood and karat of bamboo. 

* Mr Brooe informe me that there ia do maat for the keimer moder in the Miriam rig, it ia supported 

a Vol. IT. 97 

d by Google 


in the step and slopes to aUrboard. The sail is run up on the two backstays in a similar 
manner to the front sail, and it too is propped up by a forked bamboo pole, dada waku 
parttngaieinga^ (heimer akmeret lu), the lower end of which is fastened to the gunwale on the 
port bow behind the former prop. The backstays, which now pass through the gromete of 
both sails, are attached as before. A sheet is attached to the starboard lower comer of 
the liada wahi, and this sheet is made fast to the front thwart of the platform; I am not 
sure whether the same sheet serves for both satis, judging from the photographs on pL XXVI. 
I think they must each have one. 

A centre-board, im/unffa (pi. XXVI. fig. 2), is fixed on the starboard bow when sailing 
and is kept in place by a bamboo which is lashed to the outside of the canoe; the holes 
in the side of the canoe used for this purpose are called ku/jntmau lim. Wilkin (Vol. v. 
p. 297) B&ya that " the holes, Jbupuntou tirat, for the canes by means of which the canoes 
were launched were filled with fmat, the root of the tapi tree." Biiu appears to have been 
used for caulking canoes (v. p. 73). 

Details of Mabuiaq Canoes. 

"Adi" (pL XXV. figs. 2—4) has a total length of 13 m. (42 ft 6 in.), at the centre 
the inside breadth of the hall is 46 cm. (18 ia), but the average beam is from 61 to 71 cm., 
the central height is 115 cm. The gunwale ia 76 mm. (3 in.) high, the higher forward portion 
ia 13 cm. (the sudden angle which occurs when these two meet is called ngur pagami). The 
outriggers extend on each side to 3-65 m. (lift. Sin.), the floats are 3-76 m. (12 ft. 4 io.) 
long and 30cm. (1ft.) broad, the outrigger poles are 99 cm. (3ft Sin.) apart, and the distance 
between them and the float is about 2S cm. The steering-board ia 2-19 m. (7 ft 2 in.) by 
27 cm., and the centre-board is 2'61 m. (6 ft. 8 in.) by 63't} cm.; both consist of a simple 
plank straight, or nearly so, at one end and rounded off at the other. 

One canoe I measured at Mabuiag in 18S8 was just upon 15-24 m. (60 ft) long ; the 
hollowed trunk was 2*69 m. (6 ft. 6 in.) in circumference, with an opening 30 cm. (1 ft.) wide. 
The platform was 2-08 m. (6 ft. 10 in.) across and 221 m. (7 ft 3 in.) long. The inner side 
of the platform ciate was 30 cm. (1 ft) in height, and the outermost 74 cm. (2 ft 5 in.). 
The inner and outer receptacles were respectively 166 mm. (6} in.) and 89 mm. (3} in.) wide. 
The thwart poles of the outrigger were 1'6& m. (6 ft. 5 in.) apart and projected 3-78 m. (12 ft 6 in.) 
beyond the gunwale, or 2-92 m. (9 ft 5 in.) beyond the platform. The float was 3-66 m. 
(12 ft) long. One old canoe at Tutu was 20-66 m. (67 ft 9 in.) long, the trunk was 107 m. 
(3 ft. 6 in.) aci-oes in the widest part and 79 cm. (2 ft 7 in.) deep. 

I was much puzzled when I first went to Torres Straits m 1888 by occasionally seeing 
at Thursday Island a canoe with a single outrigger. I afterwards found that it belonged 
to a native of Ware, one of the New Hebrides, residing at Mabui^, and that he had 
re-outri^^ a native canoe according to the fashion of bis own people. Later on, when I 
was staying at Mabuiag, some natives of that island were fitting up a canoe with a single 
outri^^r in imitation of it. Thus foreign custom was beginning to be imitated, but ten 
years later we found that many of the Uabuiag canoes still had the double outrigger. 

CI1I7 bj tha two bamboos, aaitee, to wbiob it is skewered, and hj tbe keimer aktnertt lu. He &lio 8»7a that 
the; Bometinies erected thne luli 11 the; bad aioe light wlndaj tbe; were aalied narbtl- (elder), «p. (middle), 
and keimer- (;oanger} moder (nil). 

> pari! means faoe or front, bat I ms also told that it means port side. 

d by Google 


Among the western islandB European sails bad not quite supplanted the original mat 
sails in 1888, bat we did not see any of the latter in 1898. 

The Miriam canoes also are in an interesting transitional stage. In some instances the 
old double outrigger has been retained, though the crates are not now so carefully made 
as fonneriy. In these as in all the other sailing canoes the mat sails have given place to - 
European sails, there being a mainsail, foresail and jib (the foresail still retains its old 
name of narbet moder, "elder brother sail," referring to the time when it was the more 
important sail, while the mainsail is termed keimer moder, "younger brother sail"'); there 
is DO bowsprit. 

On being questioned the Miriam admitted that the single outrigger had been adopted 
in imitation of canoes rigged by South Sea men, indeed they attributed its introduction 
to Mataika about 1873. They said that with it a canoe was less liable to capsize, could cany 
more people, and could sail fifty or a hundred miles. The outrigger was always to windward 
and the large paddle was always held to leeward. In some instances the float is exception- 
ally long and thin, in one case it was practically as long as the canoe (pL XXY. fig. 1). 

Colonel A. Lane Fox (Pitt Rivers) says : "It is necessary that the outrigger should always 
be on the windw&rd side. The outrigger acts aa a weight on the windward side, to prevent 
the narrow canoe from being blown over on the opposite side. When it btowB very hard, 
the men run out on to the outrigger, to give it the additional weight of their bodies. Wilkes 
says that whenever the outrigger gets to the leeward side, there is almost invariably an upset. 
The outrigger probably is pressed too deeply into the water, and meeting with too muoh 
resistance, breaks the poles. To meet this difiQoulty both the canoe and outrif^er are, in some 
parts, made pointed at both ends. When they wish to tack, instead ol luffing and coming about, 
they bear away, until the vessel gets on the opposite quarter, and then by shifting the sail, 
they sail away again stem first. This system is pursued in Fiji, in parts of New Guinea, and 
northward, in King's Mill Islands (Wilkes)'." I was told that a canoe with a single outrigger 
can travel both ways. 

In many Miriam canoes the numerous thwart poles of the single outrider support an 
almost continuous platform from near the float to about an equal distance on the other 
side of the canoe, the latter extension is what Lane Fox (Pitt Rivers) terms a " weather- 
platform." The crate is absent. 

" In Samoa the canoes are built with bow and stem, and the outrigger is pointed towards the 
fore part only. As these vessels cau only sail one way, the outrigger, in tacking, must necessarily 
be sometimes on the leeward side; to meet this, they rig out a platform corresponding to the 
outrigger platform on the opposite side : this, for distinction's sake, we may term a tveather 
plaiforrn. It has no outrigger log, oor does it touch the water, bat when the wind blows so 
heavily aa to press the outrigger down on the lee side, they run out on the weather platform, 
kod counterbalance the effect of the wind by their weight. This contrivance is used in some parts 
iA New Quinea, where, it may be observed, the varieties of the outrigger canoe are more numerous 
than in most of the other islands. It is also used in Solomon Isles, where the weather platform 
b of the same width as the outrigger platform [as it is in Torres Straits]" (Lane Fox, I.e. p. 430; 

' CI. Vol. VI. p. 94. 

> "On Early Uodei of NBTigation," JoMm. Anth. Irut. it. 1876, pp. 439, 4S0 (reprinted in The Evolution of 
Cviturt, by Lt.-Qeii. A. Lane-Fox Pitt-Biven, Oxford, 1906, p. 239). 


d by Google 


and 1906, p. 222). The vftriatioas in New Guinea noted by Lane Fox occur loainly in the eaatem 
and ■outh-eaatern regions where Oceanic influence has made itself felt for a long apace of time. 
The modifications in the rig of the Torres Straits canoes are quite recent and are due to die 
Melauesians and Folyneeians who have been introdaced by Europeans, or have followed in their 

SaUlng with mat uili. 

The following are the only accounta we have of the method of Bailing canoes in Tones 
Sta^itB. I am afraid nothing can be added to them as in 1888 I believe there were only 
one or two canoea with mat sails among the Western Islandera and none either in tfae 
Hurray Islande or Erub, and in 1898 none existed in Torres Straits. 

"The two masts [of the Miriam canoea], when not wanted, are laid along the gun* 
wales; when set up, they stand abreast of each other in the fore part of the canoe, 
and seemed to be secured by one set of shrouds, with a stay from one mast bead to the 
other. The sail is extended between them ; but when going with a side wind, the lee 
mast is brought aft by a backstay, and the sail then stands obliquely. In other words, 
they brace up by setting in the head of the lee mast, and perhaps the foot also; and 
can then lie within seven points of the wind, and possibly nearer. This was their mode, 
so fiu* as a distant view would admit of judging; but how these canoes keep to the 
wind, and make such way as they do, without any after sail, I am at a loss to know" 
(Flinders, ll. pp. 110 — 111. The accompanying plate shows three canoea sailing, two with 
one aail, and one with two oblong sails, and one canoe with an outrigger being paddled. 
The drawing is not sufficiently detailed or accurate to be worth reproduction). 

Macgillivray, describing a Muralug canoe, says : " When desirous of making sail, the first 
process is to set up in the bow two poles as masts, and on the weather side a longer and 
stouter one ia laid across the gunwale, and projects outwards and backwards as an outrigger. 
These are further supported by stays and guys, and, together with another long pole 
forked at the end, serve aa a frame to support the presaure of the aaila, which are uaually 
two in number, made of matting of pandaaus leaves, and average four and a half feet 
in width and twelve in height The sails have a slender pole on each side to which 
the matting lb secured by small pegs; when set, they are put up on end side b; side, 
travelling along the backstay by meana of a cane gromet. When blowing fresh it is 
usual to keep a man standing on the temporary outrigger to counteract by his weight 
the inclination of the canoe to leeward. From the whole sail being placed in the bow 
these cannot make much leeway, but when going free may attain a maximum speed 
of seven or eight knots an hour. Except in smooth water they are very wet, and the 
bailer (a melon shell) is in constant requisition" (ll. p. 17). 

To manoeuvre the sails of a canoe is called gul wakulwiga pungai. Sails are reefed 
by rolling up the mats from below and lowering the remainder by pulling them down the 
supporting bamboos, waka tag (naiws). 

Windward ia paipa or paipa kid (on the windward side, paipal), but this also means 
on the right hand ; probably the latter meaning is derived from the former since, facing 
north, the south-east wind, which prevails for eight months in the year, would blow on 
the right. Paru is port side, hence parungaieinga is the prop set up to port Paupa or 

d by Google 


paupa kid ia teeward {patyta asi, decline [of day], go down [of Bun]), pavpa tdrai, is the 
mast that is "set up to leeward." Kal is starboard. 

The owner of a canoe in the western islands is the " captain," who stands at the stem 
and steers, and directs the operations of the crew ; but when turtle were being caught 
by means of the sucker-fish (pp. 162 S.) the buai-garka takes charge. 

The next responsible person is the buai-garka, or "mate," who stands in the bow of 
the canoe. He is nsnally the imi^ (or foiling him an ira*) of the owner. According to 
Dr Rivers (v. p. 148) his duty is to hoist the sail, weigh the anchor, cast the anchor, 
light the fire, prepare the food, indeed as the natives admit he has very hard work to do. 
His duties in connection with the sucker-fish are given on pp. 163 — 1. 

According to one Miriam legend (VL p. 3) the owner or captain of a canoe stood in 
the bow and speared fish, the mate was in the stem. According to another tale (VL p. 51), 
the following is the sacred method, zogo tonar, of turtle fishing (p. 161): the men in the 
bow are to pole the canoe, those at the stem are to paddle. One man must keep a 
sharp look-out. One man must dive for the turtle, and the other men have to pull 
him up by the rope fostened to his arm. 

The ownership and inheritance of canoes in Mabuiag ia referred to in Vol. v. pp. 286, 7, 
and the sharing of turtle, etc., by the crew of a canoe in Vol. v. p. 289. 

Th« Decoration of a Canoe. 

Canoes are no longer fully decorated in the old style, but in Habuiag I have seen 
two or three canoes which were more or less decorated in the ancient foshion; unfortunately 
modem conditions do not encourage indulgence of esthetic notions. 

I was informed that the hulls of the imported canoes were usually painted and carved 
by their New Quinea makers, and I have seen such painted designs inside a canoe ; the 
carving is referred to later. 

The hull is painted in various ways according to the taste of the owner. One Mabuiag 
canoe named lawaikan had each end, the foredeck and the bamboos which ore placed 
over the joints painted red, and numerous red lines encircled the hull; the gunwale and 
weather-board were black ; the crate, tips of the poles of the outriders and their pegs 
were red. Examples of the painting of Miriam canoes are seen in Fl. XXV. fig. 1. 

In many Mabuiag canoea an incised pattem, gant minar, mns from the bow to the 
stem immediately below the upper edge of the hull ; the left side of the lower sketch 
in fig. 210 and PL XXV. fig. 4 show the typical pattern of a double zigzag or stepped 
line, the lower one being toothed, but it may be simplified. About the level of the end 
of the weatber-board this pattem is interrapted by a design called a koimai (p. 23), 
variants of which are shewn in fig. 210. From this series it is evident that the concentric 
triangles of fig. 209 are in reality degenerations of the human face. When shewn in 
Mer these sketches were called m^kgt op (vi. pp. 273, 4). These incised patterns are out 
by the maker of the canoe in New Guinea. 

' An iMi i> roughly ipeakinfi » brolher-io-Uw, bat it hu $, more extended signiflranoa, knd baa therefore 
no EogliBb eqniTBlent (et. Bivere, t. pp. 186, 148). 

* An jro ii ft member o( • gronp of men, to ooe of whom ma term hther-in-law oonld be applied (t. p. 137). 

d by Google 



The forepart of the gunwale and the triangular weather-board are the areas on which 
jbfae owner gives himself full play. I was informed in Mabuiag in 1888 that the owner 
might represent his totem on the gunwale. In fig. 211 A, B I have sketched two 
variaats of what is evidently the same design; I was informed that they represented 
the bones of a fish, the lozenges being the end view of vertebne and the oblong the 
side view of a vertebra. The tails are similar to that of the gapu (p. 162). The zigzags 
of C were also said to be the bones of a fish, and the design beside them was a cross- 
section through the body of a fish, the central lozenge being the section of the vertebral 

Fia. 909. Bow of ft tmnoe nuoed ^uiBn sketched at MnlmiitB hj tha mthor in 1SS8. 

Fro. 911. DedgDa engnved Mid punted on the bows of MftbniH; oi 
Muioe, flg. 809. 

uid C from the Aiain 

column and the four encircling designs probably indicating the muscles as seen in section. 
My informant bad no object in misleading me and I see no reason to doubt his expla- 
nation, though it certainly appears a somewhat strange one. Of late years the decoration 
seems to be meaningless. 

The decorated bow of an Erub canoe is shewn in pi. XXIV. fig. 1, judging firom this 
drawing it was painted to represent an animal's head. 

The figure-head, rfijro*' (ffope, meket op), was always a separate carving that was lashed 
> For ma Mwoant of the bogeje oalled dtigai. He Vol. *. p. 858. 

d by Google 


OD to the top of the gah (fig. 209)^ The specimen shewn in fig. 212 was made by Mariget 
of Mabuiag ; the heads, ngagalau kunk, represent the sea-eagle, ngagalaig. A similar carving, 
ki9u kwik, occurs on Ann's canoe (pi, XXV. fig, 4), portraying a hawk-lite bird, kiaulaig. 
Id nearly all cases a human face is carved on the figure-head, the one &xim Saibai 
shewn in pL XXVII. fig. 1 being in the round, supported on a neck. In this specimen 
the bars, hair and centre of mouth are painted red, the face and ears are black ; a red 
line mns round the lower border of the eyes and across to the ears the groovings of 
which are red, and a yellow line runs round the upper border of the eyes and above the 
former line to the ears ; a yellow groove indicates the cheek-fold. When complete there 
were two bunches of cassowary feathers on each side of the face, five, mainly of bird-of- 
paradise feathers, along the top of the bar, and twenty bunches of cassowary feathers 
behind these. The total width is 413 mm., the tsae being 13 cm. long and 148 mm. 
wide including the ears. I gave a somewhat similar specimen from Dauan to the British 
Museum {Album, I. pi. 232, No. 2), but the board is fiat, and together with the &ce is 
coloured red, blue and white ; it is profusely decorated with cassowary feathers, and measures 
26 X 10 cm. The figure-head of the Mabuiag canoe (fig, 209) belongs also to this type. 

Fio. S19. Pignie-head of a MabnUg oanoe. The tront part of the faM Ii whits with a nd border, tb« month 
nd with ft bUok border, the under aide of the neok is white, the aides red, and the top aud rest of the 
bettd are black; thia coloration is extended to the central block. 48 cm. long. 

The regular decoration of the bow end of the hull of a Western canoe consists of 
a stiff V-shaped band which continues the line of the upper border of the hull, starting 
from where the gab is attached to it; a similar band projects upwards and forwards, 
thus forming a kind of jaw, as is indicated by the name of the ornament, gud, mouth 
(fig. 209, pi. XXV. fig. 4, Album, I. pi. 323, No. 2); a vertical stick, paipa za, connects 
the apices of the jaws and keeps them apart. Both bands are generally decorated 
with a row of small grey cowries, uza, and a row of large white cowries, bttbiam, is also 
attached to the lower band and one or two to the upper, the details naturally vary. 
The top of the vertical stick is furnished with a tuft of cassowary feathers, and a fringe 
of shredded young coco-nut leaves, tu (»u), depends from the lower jaw, which according to 
the Miriam represents a beard, imaa (pi. XXIV. fig, 1). Tufls of cassowary feathera may 
also be inserted along the sides of the end-board, along part of the junction of the 
weather-board and gunwale, and at the end of the former; these feathers are called 
"whiskeis" by the Miriam. The posterior border of the deck-covering of the weather- 
board in the Mabni^ canoes may also be decorated with tufts of feathers. 

The stems of the canoes have also their characteristic decoration. The stem-post, 
hm, of the canoe shewn in pL XXV. fig. 3 consists of three vertical sticks, two of 

d by Google 




which at their upper extremities clamp a more or less horizontal stick, while the third 
is laehed to its anterior end; to the posterior end of the horizontal stick are lashed 
two vanes, which make a lai^ fish's tail, wapi pakai, resembling that of the king-6sh 
or one of the allied gigantic predacious mackerels. Between the tail and the post is 
fastened a long streamer of "grass" which is made faat below to the stem of the canoe. 
The top of the post is furnished with cassowaiy feathers and a bunch of gba rattles ; 
the whole is steadied by a guy-line, zez, decorated with pieces of calico. In front of 
the stem-post four carved staves, gozed, project upwards and outwards, two on each side, 
being lashed to sticks which extend across the gunwales. Each goged has a serrated 
edge and terminates in a Inau kwik, below which are tied feathers, calico and a white 
cowry. Two shell trumpets, bu, are placed between the four goied. 
Projecting behind on each side of the canoe are two other gozed, 
two of which are plain and two deeply serrated. A small slat of 
wood (fig. 238) was suspended to the stem-post, it strangely re- 
sembles one type of British bull-roarer but is dissimilar from the 
Torres Straits type (see Sound-producing Instruments); I was in- 
formed that it did not mean anything. 

Unfortunately no direct information was obtained concerning 
the meaning of the goged, but there can be little doubt that they 
had at one time a magical significance. The upper end was generally 
carved to represent the head of the frigate bird (fig. 213) or that 
of the sea-e^le (Album, I. pi. 323, No. 4). Both these and the 
king-fish and its allies are voracious catchers of fish, and the repre- 
sentation of them would therefore be obvious to the native mind. 
Their use would therefore be analogous to that of the canoe charms 
mentioned in Vol. v. p. 337. Similar carved staves were employed 
by the maid le in sorcery (VI. p. 230), and probably also by the 
maidelaig (v. p. 321). 

In 1889 I obtained at Dauan a fiat board, 1'26 m. long, it was 
perforated with numerous holes along its length, and the lower 
portion was carved to represent a human fiice {Allmm, i. pi. 323, 
No. 3). It was used as a kun and also as a tobacco charm (V. p. 346, 
Tl. p. 207). Perhaps the atem-post of the canoe shewn in pi. XXIV. 
fig. 1 was somewhat similar. I was informed in Mabuiag in 1888 
that a piece of the skin of the owner's totem animal might be 
hung on to the kwi of a canoe. 

The crate was sometimes painted red ; the terminal and some 
of the intermediate outer long vertical, sticks, usually those opposite 
the partitions of the food compartments (p. 208) were provided 
with a tuft of short cassowary feathers (pi. XXV. fig. 2). Sometimes 
stiff rod-hke plumes of cassowary feathers, paupusa (W.), were em- 
ployed. A frequent decoration consisted of flags, dadu (W.), which 
were made of pinnules of coco-nut palm leaves, lashed on to a 
pole, and provided with a tuft of cassowary featheis above and 



Fio.218. Oottdkalapitrova 
tha fltflm of B oanoe. owred 
to reprewat Ui« pod of 
the "QneeDBUnd boui," 
kalapi, and the head of 
the frigate-bird, vomer. 
Copied from a pliotograph 
of a ipeoimen in the Ant- 
traliao Mawam, Bjimj. 

d by Google 


below (figs. 214, 255, Bee also Vot. V. p. 357X A flag in the Brit. Mus. &om Erub is figured 
in the Album (ll. pi. 322), the pole is 8 feet long, the upper part of the pole is carved 
with a coDe-in-cone oroament. 

Pio. 214. DrswJDg bf SondAj of M&bnug of the turtle ceremon; on » canoe at Qomn (v. p. 331), abewing 
the oidinuy deooiation of a Mooe, and two flags on tb« onto. 

d by Google 


The knowledge of natural pheaomena and objects possessed by the Torres Straits 
Islandera was similar to that of most nature-folk, in other words they had a sufficient 
knowledge of nature for all practical purposes. This knowledge, which was utilised for 
gardening operations, fishing, ceremonial occasions and the like, must be distinguished 
from the tales that are told about the heavenly bodies, rocks or other noteworthy objects ; 
the former may be regarded aa their science, while the latter represents their literature 
or teleology. 

The following subjects are dealt with in this section : 

ABTROttomr (W. H. R. RtvBtia) 


Calkndar {S. H. Rat) . 
gsooiuphy .... 
Gkhbrai. and Spicial Tibms 
Natural Hibtort . 

DiKKrrioNS of Spacb .... 832 

Rblation of Man to his Scrroundinos 232 

Ndhbkals, Countino and Rbcords . 232 

ccrrbnot s36 

Wbiortb and Measdreb ... 236 

HroiENB 237 

W. H. R. Rivers. 

Only a very fragmentary account of the astronomy of the people was obtained. 
It ia not easy in a short time to map out the sky completely according to native ideas, 
and in both parts of Torres Straits the difficulties were increased by the &ct that the 
natives were forgetting their star-lore and were uncertain about the identity of stars which 
we know to have been of the greatest importance in the old life of the people. All my 
observations were made in the evening and I now very much regret that I did not adopt 
the obvious measure of getting up some morning before daybreak when other regions 
of the sky would have been visible. 

The material collected is, however, sufficient to demonstrate the great development 
of the ideas of the people about the heavenly bodies and especially about the stars, 
a development which is definitely connected with practical importance, for the islandera 
not only used the stars in navigation but also in connection with agriculture and in the 
regulation of the times for certain ceremonies. Further the constelUtions of the people 
stand in a definite relation to their mythology, while certain facts pointing to the close 
connection between certain stars and certain localities and petsons indicate an even closer 
relation to the religion and sociology of the people. 

d by Google 


The chief interest of what wae obtained lies in the fact that the people group the 
stars in coastellations wholly differ^it &oin those recognised by ourselves but with such 
a general similarity that we may conclude that the grouping owes its origin to ideas 
similar to those of the people from whom our constellations have been derived. 

Little could be discovered about the ideas of the people on the nature of the sno 
and moon. The people used to tell Mr Bruce that during the night the sun returned 
to its place in the east by going under the water but they will no longer talk about this. 

The moon is regarded as the husband of Ilwel, the evening star, the stoiy of whom 
has been given in Vol. vi. p. 4. 

Planets were recognised as different from stars, or at least it was pointed out that 
Venus did not twinkle (epreki) while a star did. 

The evening star was Ilwel, while a planet, said usually to be seen in the morning 
and called Gegemeaaur^, is probably the name for Venus as a morning star. 

The constellations of which an account was obtained in Mer were the following : 

Beizam or the Shark. This consists of certain stars of Ursa Major together with 
Arctunt8(?) and Qemma of Corona Borealis. The seven chief stars of the Bear form 
the body of the shark and two small stars which in our customary representation form 
part of the mouth of the bear (t and k Urase Majoris) were its eyes. Qemma was at 
the extremity of the ventral tail-fin. There is some doubt about the inclusion of Arcturus 
in the constellation. According to one account this star fonns the extremity of the 
dorsal tail-fin but it also had a special name Dbgai representing a being believed to 
have much influence on the weather in the north-east season by swinging the tail of 
the shark (see Vol vi. p. 271). 

A drawing of Beixam was made by Debe Wali but the stars represented in it bore 
no resemblance to the actual arrangement of the stars in the sky. There is a shrine 
at Babud connected with this constellation (see Vol. VI. p. 269). 

Tagai. This is a very large constellation, including many of our constellations, which 
illustrates the chief personages of the story of Tagai given in Vol vi, p. 3. The whole 
of the constellation was visible in the evening during our visit. It represents a man 
T^ai, standing in the fore part of a canoe formed by the body of Scorpio, while 
Antares represents one of the personages of the story, Kareg, sitting in the stem of the 
canoe. Tagai holds in his left band (the Southern cross) a fishing spear, and in his 
right hand (Corvus) some fruit of the kupa. Tagai himBelf is represented by stars of 
Centaurus and Lupus. Those most definitely identified were 7 (?) and ft Centauri 
representing the eyes, 0, of the same constellation, the chin, and ij Centauri the navel. 
Two stars close together, k Centauri and j8 Lupi, were said in Mer to be the testes. 
Sagittarius was said to be the aachor(?). 

Below the canoe was a sucker-fish formed by part of Scorpio and possibly of Telescopium 
(see p. 221). 

I In the VoosbiilAi; gtrigtr nttau is th« moniiDg Etai and ki netau the evening tiar ; Di Haddon obtained 
gertgtr %e»a>tr for tba moming itor. 


d by Google 


At the time of our ^it the coDstellationa Usiam and Seg representing personages 
of the story of Tagai were not Tisible in the evening, but they had been identified by 
Dr Haddon during his firet visit with the Pleiades and the belt of Orion respectively. 

Naurwer or Mahersor or the Brothera. The two chief atars of this constellation 
are Vega and Altair. Vega is the elder brother or 7taH>et and Altair the younger or 
keimer (see Vol. vl p. 96). With Vega are associated /3 and 7 Lyrse, representing two 
sticks which the elder brother is holding on one side of his body, and with Altair are 
associated /9 and 7 Aquilae, representing two sticks which the younger brother is holding 
out with arms outstretched. 

This constellation was said to belong to certain villages and its constituent stars 
to individual persons. Mahersor is its Ulag name and it was said to belong to this 
village when it comes up; Naurwer on the other hand is its Sebeg name to which 
place it belongs when it goes down. Narbet or Vega was said to be the property of 
the Mamoose (Zaub 2) who had inherited it &om a man named Katu, while keitner 
belonged to Eaige (Saugiz, 6) to whom it had come from his &ther Obra. There are 
certain stones connected with the two stars on the land of these men, both of whom 
belong to the neighbonrhood of Sebeg, and it is possible that thero are also Ulag men 
who claim the stars as their property. 

This account of private property in stars was only obtained during the last day or 
two of our visit and the clue so given could not be followed up, but the constellation 
Beitam probably has a shrine associated with it (see Vol. Ti. p. 270), and according to 
Hr Bruce the star Dbgai belongs to the district of Ebmet. This association of stars 
and constellations with persons and social groups is highly suggestive and may be 
a survival of a division of natural objects between social groups such as is found among 
many peoples. 


At the time of our visit to Mabui^ some of the constellations which had been 
visible in the evenings in Mer had disappeared, but it was possible by means of drawings 
to be satisfied that there was a very close correspondence between the constellations of 
the two communities, though the names were often different and there were difierences 
in detail 

Fio. 31S. Tha Baidam ooniteUfttioo drawn by Haii. 

Baidam or the Shark. This evidently corresponds closely with the Mer grouping, 
but the two small stars in front, which were the eyes in Mer, were in Mabuiag said 
to be tuopi, two small fish which swim in front of the shark — pilot-fish. Further, names 

d by Google 


were given to several atara represented in the drawing of Naii (fig. 216), stars which 
it 18 unfortunately impossible to identify. The two small stars at the snout of the 
shark were oalled ipioli^; the two at the extremities of the pectoral fins were komaei 
and that at the end of the fore dorsal fin, kuikuitugu. The star at the end of the 
dorsal tail-fin, probably Arcturus, was called Duga or Dugiia, corresponding to the Dbgai 
of Mer. In Naii's drawing as in that made in Mer it is not possible to identify the 
seven chief stars of the constellation with any certainty. 

Tagai or Togai. This constellation evidently has the closest similarity with that 
of the Eastern Islanders. It had largely disappeared in the evenings at the time of 
our visit to Mabuiag, but the canoe, its anchor and the sucker-fish were still to be seen, 
and four stars in the tail of Scorpio (x, X and v Scorpionis and 7 Telescopii) were 
pointed out as the body of the sucker-fish, its head (gapukvnk) being &r away near 
the Pleiades (possibly the Hyades, see p. 223). Sagittarius was identified with the anchor 
as in Mer. Drawings were made of the whole constellation which are shewn in figs. 
216, 217. 

Flo. 316. Tbe Tagai eonstaUktion, dnwn 
by Hariget; lednoed to (. 

>. 217. Tagai and Karig in their oanoe draws b; 
Qiza; rednoed by one-half. In thii dnwing the 
canoe, Karig and the BDoker-flah are repnaented 
tbe wrong way loiind. 

In fig. 218 is shewn a plan of the constellation constructed by Waria, The identi- 
ficationa of the lower part of the constellation were made in Mabuiag but those of the 
parts of Tagai himself are derived firom Mer and their exact correspondence with th(»e 
of Mabuiag cannot be guaranteed. In the drawing made by Gizu the canoe has been 
reversed mirror-wise, a common feature of the drawings of the islanders (see Vol ll.). 

In the plan made by Waria a group of stars is given near the bead of the scorpion 
to represent a reef and two stars near its tail were called d^^coHr, but neither of these 
groups can be identified with certainty though the latter evidently form part of Ara, 

Dbgai. This Mabuiag constellation corresponds to tbe Brothers of Mer. It consists 
of Tega and Altair with the same associated stars as in the Mer constellation. Vega 

d by Google 


is called / and also Dbgai vmiiraiaig and Altair is Meta&orab or Dbgai kukilaig. The 
associated stars represent arms, as in Mer, one in one case and two in the other. 
Bu or the trumpet shell This constellation corresponds to our Delphiii. 

/ * 

Fio. 218. Diawii^ of the Tagai oonstelUtioii b; Wuia. |. 

These two constellations are connected with the folk-tales recorded on pp. 12 — 16 
of Vol V. The one arm of / and the two of M^akorah are shewn in the drawings 
which illustrate these tales. A drawing of these constellationB made by Qizu is given 
in fig. 219 A, B, It illustrates very well the imaginativeness of this artist, the two stars 

Flo. 219. DnwiDgs of conatoUfttioiu bj Qiza; rednoed }. A, Dbgai wauralaig or J. 
B, Dbgai kuhiUtig or Uttakorab. C, Ba. 

associated with each of the Di>gai having become four in one case and six in the other 
while the five staia of Dolphin have increased in number to a still greater extent 
(fig. '219 c). As it was possible that in this case Gizu was representing smaller stars of 
Dolphin which he could see I enquired into the number of stars which several native^ 
recf^nised and found that they saw five only as we do. 

Kek. This star represented in fig. 220, is one of great importance in the livee 

d by Google 


of the people and yet there is much doubt whether the identification derived from 

several natives is correct. There was much agreement that Kek was Achemar 

(a Eridani) which was visible during the evenings at the time of oar visit, but the great 

size of the star in Qizu's drawing and its many radiations 

sorest a brighter star than Achemar and raise the proba- « v 

bility that Kek was Sirius or Canopus, and it is in accordance ■ o 

with this probability that one man definitely denied the 

identification of Kek with Achemar. In view of this '^ « 

discrepancy of evidence, it may be useful to record that ^ 

Kek rises at daybreak at the beginning of the south-east ^ 

season immediately over the hill called Baudar in Moa. 

It disappeais at daybreak in the middle of the north-west ^ . ^ 

monsoon. In the drawing two stars are represented called 

Keaken tonar, " the manner or fashion of Kek." These , 

stars rise shortly before Kek and indicate its approach. * " 

These are stars in Phoenix if the identification with ^ 

Achemar is correct. In his drawing Qizu has also repre- ^ ^ 

sented Kek as surrounded by a circle of stars but here, » 

as other drawings have shewn, exactness of representation Fio. 330. Drawing of Kei by Qim. 

is not a characteristic of this artist. 

The natives of Mabuiag recognise a group of constellations called respectively Usal, 
Gapukwik -and Ltdeal dose to one another, while a fourth Oetulai or Oetalax is probably 
not far distant (fig. 221). Of these the only constellation visible in the evening at the 
time of our visit was Uaal which was undoubtedly the Pleiades. Qapukmk was described 
as being to the' soutti of JJgal and as coming up shortly after and it is probably to 
be identified with the Hyades. 

Fio. 331. DTBiringa.ot aonstollUlone, A — b; Oitii, D — Q b; WarU; rednoed \. k, Wapi. 
B, D, Ottalai. C, d, Dideal. F, Oapukirik. 

Ifideal or Dedeai was described as being to the south of Gapukwik and as rising 
shortly after it. This suggests that it may be the whole or part of Orion, but the 
stars in Dideal were stated to be very little larger than those of Usal and its identification 
must be left uudetermined. 

Oetalai or the crab was represented as near these three constellations and the drawings 
given in fig. 221 would seem to indicate a nebula rather than a star. It was described 
as a large star looking like the ashes of a fire. It has a star on either side, its get 
or arms, and two 'stars called wapi or pilot-fish come up before it.. 

d by Google 



Usal and Gapuktoik are both connected with the story of TogaL Oapnkwik was 
said to be the head of the gapu of that stoiy and iB probably connected with Bome 
incidents of the tale not given when it was related to Dr Haddon. 

A star called Panauna graz was identified with Fomalhaut. 

Nothing could be discovered indicating that stars or constellations are the property 
of persons or groups as in Mer. 

The function of these stars and constellations in marking the onset of seasons and the 
proper times for certain ceremonies are recorded on 
pp. 225 ff. The most important star was vmdoubtedly 
Kek, the rising of which not only gave the sign 
for the beginning of much ceremonial but also the 
appropiiate time for planting the new crops. The 
rising of Dbgai, viz. Vega and Altair, seems also to 
have been an important event. Whenever in such 
a connection the people speak of the rising or setting 
of a star at a certain season they mean the time of 
the year when the star or constellation in question 
just appears or disappears on the horizon at daybreak. 

When the rising of a star is expected it is the 
duty of the old men to watch. They get ap when 
the birds begin to ciy and watch till daybreak. In 
the case of Kek and probably of other important pig. 333. Drawing of the moon with a lub, 
stars and constelhitioos the appearance of certain other kubwii, by Oisn. }. 

stars is taken as a sign of the near approach of the 
object jbr which the old men are watching. In the case of Kek it is the two small 

Flo. !I2S> DrawiDp of the mjh uid moon bj WbtU. A, Sun at lenith; B, Son at horison; 
C, Uooa at lenith; D, New moon; E, Moon at horizon. 

d by Google 


stars csJled Ktaken tonar which serve this fiiDctioD and when they appear on the faorizoD 
at daybreak it is known that in a few days Kek will appear and the look-out becomes 
especially keen. 

The setting of a star is watched in the same way. 

In the western islands the figure seen by the people in the moon is that of 
Aukwum (or Aukum) with a coco-nut (urab) between two pandanus trees {kausa). This 
is [nvbably the woman whose story is recorded in VoL v. p. 56. A drawing of the 
figure by Oizn is shown in fig. 222 and by Waria in fig. 223 c, here Aukwum is 
seen holding out her hand to the centre of the disk, the lines in front of her are 
the kauaa. In fig. 223 are also shown Waria's drawings of the sun and moon at horizoa 
and zenith which shew veiy well that Waria had observed the enlargement and alteration 
of shape at the horizon. 

According to Hr Bay the following designations of the phases of the moon occur 
in Mabuiag: dang mulpal, tooth moon, for the new moon when first seen it was also 
described as unmarried ; a little later it is called kisai and termed young ; the half moon 
is called ipi laig, married person; the moon in her third quarter is kan laig, person 
with child, and was described as having one child (probably meaning that she was 
pregnant); and the full moon is badi, which was said to be kaiza ipilaig, big one 
married. In Mer the crescent moon when first seen is aketi nwA, md) digemii when 
in the first quarter, in the third it is m«& sisimi, nearly full moon is eip mfh and full 
moon giz rneb. 


There are two main seasons in Torres Straits which are usually spoken of by 
Europeans as the south-east and north-west monsoons respectively. 't\xe former is 
the dry season, and ranges from about April to November; it is characterised by almost 
continuous wind from the south-east The latter is the rainy season, and is accompanied 
by storms from the north-west and by intervening calms; at night lightning is seen 
flashing almost continuously in the north. 

There was no division by the natives of the year, vnet (W.), urut (El), into months 
or days, aad the years were never counted. Time was usually reckoned by suns (or 
days) and by moons (or months). Natives who have learnt how to tell time by the 
clock can usually estimate the hour very accurately by noting the height of the sun. 
When wanting to make an appointment at a given time, they do so by indicating 
where the sun will then be in the sky. Sunrise is now usually called in jargon-English 
"small daylight" or "little fellow daylight." There is no artificial method of measuring 

The seasonal appearance of certain stars or constellations was noted, and their 
rising regulated particular ceremonies and various horticultural events. For example, 
Dr lAndtman was informed in Badu that when only the tail of haidam (p. 220) is 
above the horizon the north-west wind begins to blow a " little bit " ; when the tail 
has gone down altogether the people begin to plant yams ; when baidam comes ap 
agun yams, sweet-potatoes and bananas are ripe. 

H. VoL IV. 29 

d by Google 


In the legend of Ewoiam {v. pp. 67—69) we find that one day he tent his maternal 
uncles Koang and Togai (or " Good Eye " as he was also called) in a canoe to get 
some turtle-shell for him. They could both make fine weather, but Togai could mahe 
the best weather. The crew, cousisting of Utimal, Usal, Kwoior and Eeg, stole the 
water belonging to Koahg and Togai, and on a subsequent voyage they were killed in 
revenge. The two old men spake thus to the dead men: "Usal, you go to New 
Guinea (Daudai) side, when you come up there will be plenty of rain. Utimal, you 
go to New Guinea side, you have to bring rain. Kwoior, when you come up over 
Bum (Mangrove Island) just before the south-east monsoon sets in there will be rain 
in tbe morning, then the wind will shift and it will rain in the afternoon. And you, 
Kek, will come up in the south between Badu and Moa, and it will be cold weather. 
When you go round this way and when you come up, then the yams and eweet- 
potatoea will be ripe. You all have work to do." According to this folk-tale these 
dead men were transformed into stars or constellations whose function was to usher 
in certain seasonal changes when they first appeared on tbe horizon. One informant 
said that Utimal and Usal form one group of stars ("make one lot"), of which Utimal 
is the koi nel, or comprehensive name, and Zugubal the magi nel, or special name. 

Th« Saoioni of the Wettem Zslanden. 

Here than once when natives were questioned about the seasons of the year, they 
began with aurkd, and I beheve that the recognition of the beginning of this season 
is their nearest approach to the beginning of a new year. 

Snrlal'. This is the name given to the turtle when copulating; while in this 
state they float on the sea and are readily caught. The importance of turtle as an 
article of food has given rise to the name of tbe season when they are most abundant 
(p. 160). It Usts from about the middle of October to the end of November, but Uie 
limits are not constant. 

The haidam or shark constellation now appears ; it is described as haidam gata 
widan, tbe shark constellation is close to the reef. Everything is dried up and the 
yams are ripe. The sounding of the first thunder {doiom) is the signal for planting 
yams ; when the Habuiag people hear the thunder they say, " Baidam sibui tidi," 

Ras was described as "time of die," that is, the season when leaves die down. 
The first portion of it is called duau-urma in Uabuiag, that is the falling of the cashew 
nuts, dva (Semecarpus heterophyllns). There is then an interval of fiue weather and 
the wind is shitty; it coincides with about Christmas time. This is the time when 
the yams which have been planted begin to sprout; in Muralug it is called malgvi, 
which is the exact equivalent of our word spring. Wallaby of Muralug said that they 
could no longer eat yams or sweet-potatoes, as they were watery when their leaves 
sprouted. The next division is called dob (the last of growing things) or kusi kuln 
(medusae of the north-west), tbe latter name being due to the large numbers of jelly- 

> Hacgillivray girea tdUtr m the Dftme of the green turtle in Muialog (whiah is onlj putiftU; eorreot) uid 
ttlangi u tbe teuoa of tbe year when it ii moat plentiful. Quite independeotlf I had ootad nirlol u being 
" turtle bat," and $arUtnti al the wmod when the turtle ie " fftst." 

d by Google 

80IBNCE 227 

fish that float on the sea. The runners of the yams now grow. The time immediately 
after this ia called parimugo, apagap or kerne in Muralug. 

Knkl. This is the season when strong winds blow intennittently from the north- 
west, accompanied by deluges of rain. During the intervals of the storms there is no 
wind, and the sea becomes calm and glassy. The damp heat is rather trying. The yams 
and sweet-potatoes are too watery to eat, and so recourse has to be had to unsavoury 
&I1U (p. 135) and to the seeds of the Queensland bean or kolap (p. 134). 

The constellation Dbgai kukilaig or kai Dbgai of Muralug is the Dbgai metakorab 
of Mabniag =: Altair and the associated stars and y Aquilse (p. 220). The appearance 
of these three stars hnralded the beginning of kuki. The constellation bu (trumpet 
shell) = Delphinus is associated in the legend with Bbgai Toetakorab. "Dbgai go first, 
bu come after. When 6u go down blow comes, wind not strong." An informant said : 
"Kukilaig go first, wauralaig come after; when kukilaig go rain come; when wauralaig 
go down south-east begins." 

Although one informant said that Used and Utimal were stars of the south-east 
season, yet their first appearance according to the legend of Kwoiam must be during, 
or towards the close of, the rainy season, the end of which is marked by the appearance 
of Kwoior. 

Albaud. This is the dry season when the roots become strong and food is abundant. 
The south-west wind, uraur, begins to blow steadily, and from this fttct the first part 
of the season is called waur' and perhaps merits as much a distinctive name as ru^. 
It is marked by the appearance of a constellation consisting of a large and a small 
star, the magi Dbgai or Dbgai wauralaig of Muralug, which is the Dbgai I of Mabuiag, 
= Vega with /3 and y Lyrae (v. pp. 13, 16). As food is abundant the natives are in 
good health and spirits and perform various ceremonies, the time for these being signalled 
by the appearance of the star called Kek, which is our a Elridani ; the Dideal and 
Uaal constellations also appear. During the waur season the Torres Straits pigeon 
(pp. 152, 153) migrates to New Guinea, and the natives begin to eat taro, sweet- 
potaboes, and wild yams (boa), and cultivate varieties of yams such as tauur, kutai, and 
other edible plants, for instance ikur. 

I obtained from Tom of Mabuii^ the following list of stars associated with this 
season : — " Kek, come up and is the sign [me^] for everything to be done, start meeting," 
that is ceremonies which are dependent upon a good food supply can now be held — 
at the present time the "May meetings" take place at this season; Oil; Usal (the 
Pleiades) — at this time the ovaries of the turtle enlarge; Pagaa and dede (Betelgeui). 
Utimal ; and Wapil. Towards the end of aibaud the large constellation of baidam (shark) 
is seen, and then the Torres Straits pigeon migrates from New Guinea to Australia, 
as do the birubiru birds when the crab constellation, Qitaiai, appears. 

■ In Uoralag latiaaw ii the nkme for the beginniiig of the sonth-ewt trioda, it wm deMribed m "Bmall 
Mow Matb-CMt"; when Kek ftppean (atfuour (fftther-waitr), the "big fellow wDth.east," begitu. The term 
iMi ie applied to a tatber and all tnale relatiTeB of a genention above the apcaker who are not nufwani to 
him. 8mI is probably kati, child, the aoneeponding relation ot the generation below (t. p. 1S3). 

d by Google 


The Beasotu of Uw Saatem Idaaden. 

Our information about the aeasouB recognised by the Miriam is unfortunately very 
scanty. The nortb-west season is called koki; gaibar is tbe spring, from gain, all 
things, bar, spring up. The Uaiam constellation (Pleiades) indicates tbe beginning of 
the turtle season and of early food, and gardens are now prepared. Mag is the season 
when tbe new leaves of tbe yam are sprouting, tifni ia that when tbe yam tubeis 
swell, I believe tbis is indicated by tbe near wer, girl star. GHe nur, beginning of 
harvest, is the time when the yam leaves wither and tbe yams are matured, nur is 
harvest time. Eged or egeb, about the end of Augnst to November, is the season when 
the gardens, gecLvb, are cleared of bosh ; the signs, tonar, which indicate this season, 
are mentioned on p. 145. Tbe fishing at naiger time (November) is referred to on 
p. 158. 

Calendab of Western Islands of Torres Straits. 
S. H. Ray. 









J>«au un- 

DbgaO kullcaka 

Wind aU waja. 


Ilttauurma (limit 

Bat dua trait 


SUrt N.W. 


of dua tree ia 

i)a«at go down 
in thfl Wert. 

ripe). GnM 

Dob ""-I 

growi a little. 

KMt kuki n. 


begin to grow. 


Bu (Ddphin) 

Wind not itroiig. 




K»H . 




Eat biiu and 


AH (»in). 






fifntUnt bird 

Eat (tnii, hto, 




uiivw (Irom 



N.W. wind 


Women di«j«ri 

p. 286). 





Wini ffugadar^, 

flower, new 

,8iui WoM- I 


" SniOl fcUow 

All diy^U»Te« 

Ontboab. Make 


gaiiimi fly to 

tall, new yama. 

dngong plat- 


blow for long 

navai dogam 

form on reef. 



(i.e. N. to Naw 

plant aweet- 
potato in 



"= Piep. 


Oial, OmI ud 
Wapial Appear. 

Fine weMher. 




Iran IToar 


K,k riw.^d 

"Big fellow 

Tarn leaTsi dead 


■eU jtut kftei 



" Hay Meet- 



ings"), formerly 
aeaaon for sen. 


Birubiru j 

Qitalai ftppean 

Plant yama. 

kltei Ffan mt 


gai«a» fly to 

rial dogam (ia 





Tnrtl* oopa- 

Turtle aeawm. 

ammer, lit. hot eeaaon; aibaud, harreat; mm 
a appear to be fomewbat earlier tban thaae. 
■ Note that Dbgai ia oonpoeed of 3 atais, benee tbe dual verb*. 

d by Google 



The natives have a Btrongly-ioarked appreciatioQ of geographical features and can 
readily make mape or portray the esseotiaJ characteristics of an island from memory. 
These bcuUies are probably to be accounted for by their being navigators, and their 
fondceea for drawing animals has made them good draaghtsmen. 

On p. 163 of VoL v. I have given a native map of Mabuiag, which should be 
compared with the sketch map of that island on p. 7 {I, c), but I should m)t like to 
vouch for the extreme accuracy of the latter. I have several other maps of parts of 
the island, and in order to illustrate a folk-tale Papi of Mabuiag drew a sketch map 
of the uncharted coral-ree& between Mabaiag and Bum (v. p. 60)'. 

Qeographical ideas were somewhat crudely represented by the Miriam in the divinatoiy 
shrine of Tomog xogo and in the sacred ground at Dam (vi. pp. 260, 303). When 
Dr Rutherford was at Mer in 1833, he wrote that Madeau or Secure " was easily 
made to undeistaod the meaning of a chart of the straits, and was highly gratified when 
Murray's island, laid down in it, was pointed out to him" {UnUed Service Joum, 1834, 
pt IL p. 202). 

All the islands however small, every large saodbank and many coral reefe are named. 
Banks Island, however, did not have a single name, its hilly eastern portion being known 
as Moa, while the western low-lying land was called It — the former term is now applied 
by Europeans to the whole island. Each island of any size ia further divided into a 
large number of named districts, sometimes of very iimall extent; these may be grouped 
into larger districts, as in the case of Mer (VoL vi. p. 170), thus the position of house- 
sites, gardens, &uit trees, etc. can be located. Qeographical features such aa hills, 
streams, swamps, water-holes, capes, coves, prominent rocks and the like have their 
distinctive names. 

Qenebal and Specul Tebhs. 

It is important to note that terms are classified (more particalarly perhaps among 
the Miriam) into those which signify a group or general term, koi net (W.), au net (E.), 
or big name, and those which denote a special name, magi nel (W.), kebi nei (EL), or 
little name. The following Miriam examples will suffice to illustrate this (cf. pp. 53, 
230, VoL HI. pp. 17, 59). 

Lor is the au nei for fish, but the name of each kind of fish is a Jwfrt net. 
Lewer (food) is the general name for all kinds of yams and u for coco-nut, but each 
variety (pp. 133, 136) has its special designation which is called a A»&i nei, 

A further distinction is sometimes made with regard to what are called au au 
nei, or very big names, as an example of which Mr Bay was given ^u, which includes 
lu (properly plants), meta (house), baker (stones), and such things as kokea (boxes), lan^Kt 
(lamp), eik (floor); ^u, meta and baker are themselves au nei. 

The au nei of the masked perfonneis in the initiation ceremony of the Bomai- 
Malu cult was Offud, Malu was a ktbi nei and was known to everyone (Tl. p. Z7), and 

Dine of Waria'i drawiDHi of the geognphiral futnrei at Hmbnikg wiU b» fonnd in Hut A. 
DU StabkarteH der Af an AoU-Inmlancr (Hamburg, H. O. Paniabl, 190S). 

d by Google 


Bomai a gumik tin or secret name. One disguised performer in these ceremonies had 
the kebi nn of Hagur, but his togo nei (sacred name) or au nei was Ih (vi. p. 312), 
which also was a gumik neL 

Natural HisTOBY. 

The natives are good field naturalists and have names for a large numbei* of plants 
and animals. A considerable number of plants are utilised in one way or another, more 
so than we have mentioned in these Reports. Although the land &ana is deficient 
in forms of economic importance, the natives have names for animals which are not 
of value to them, and are acquainted with their habits; their knowledge of the natural 
history of marine animals being very extensive. The uses and properties of most of 
the plants are known to them. 

As will be seen in the section on Decorative Art, the drawings of animals are 
remarkably accurate, especially when one remembers that the majority were drawn from 
memory; but although the essential characters of the external form are recognised we 
did not discover that this has led to a system of classification. A remarkable' carving 
in coral of a fish's skull is described in Vol vi. p. 284, pi. VIL fig. 4. 

The only domesticated animal was the dingo, as is the case with Papuans and 
Australians; at present they have a mongrel breed of dogs. Jukes (I. pp. 202, 209) 
refers to the keeping of a cuscus in a kind of cage by the Eastern Islanders (pi. IIL 
fig. 3). Cats and fowls have now been adopted. 

Bampton (Flinders, i. p. xxxvii) says that in Erub "on all parts of the ree& there 
were bamboos set up with pendants of dried leaves," the meaning of which he could 
not ascertain. (Captain Lewis found they were merely ornaments, " 6ach pole is surrounded 
by a string of shells. Ireland says they are placed there for birds to build upon. 
The birds he describes to be with long necks and legs and wings, Uke cranes : perhaps 
the blue heron" [cf. p. 153] (Naut. Mag. 1837, p. 755). Accepting Ireland's version, it 
would seem that the object was to attract the sir in order that it might be more 
easily shot for its plumage. This may be regarded as a first step towards domestication. 

A large number of animals have separate names, but in some instances the same 
name is given to different species, mainly, I believe, because they are used for a similar 
purpose and possibly any special designation may be of no consequence. For example, 
in the West id is the name for small bivalves such as Tellina staurella, while Lucina 
exasperata is called warkid id, that is, another kind of id', similarly Barbatia fusca 
(fig. 72) is called tepe and Modiola subramosa wakid tepe. Of two species of spider 
shell, one (Pterocera lambis) is called aaor (E.), while another with carved spines (P. 
chiragra) is called tiaasor. Pasi of Mer wrote for Mr Bay a list of " many shell fishes* 
names," he gives sorsor as the au n«i and enumerates 42 kinds as k^ net. The 
Miriam au nei for clams of ■ the genus Tridacna is mi: beieam mi is T. serrifera; terpa 
mi is T, elongata (terpa being the coral-rock oyster, Ostrea mordax), but the giant 
clami T. gigas, is called taikor. The male of the sun bird (Nectarinia australis), ti, was 
called by the Miriam hapi it, dark ti, and the female nurU> ti, sere U, on account of the 
throat of the one being dark and that of the other yellow (TI. p. 8). 

d by Google 

: aCIEHOE 281 

As the natives of Mabuiag constantly cut up dngong and tnrtle they have distinct 
oames for parts of the body, organs and different portions of the intestine ; we collected 
a cohsidetable number of these names, which were api^ed in most cases to the 0(}iTesponding 
structure in both animals. 

Owing to the utility of the coco-nut palm there is a distinct designation for its 
variona parts and for the various stages of its flowering and the ripening of the nuts. 

One expects to find names for meteorological phenomena such. as rain, art (W.), 
irmer {%.) ; rainbow, kuruai, oripard, nsurt (W.), tuaeri (K) ; wind, guba (W.), woff (E.) ; 
and clouds eitt, bat (W.), bag (E.); but it is interesting to note that the Islanders 
have names for particular kinds of clouds, thus a cuUulus cloud is amat (W.), xioff (K) ; 
bSffai or bige (W.) is a bill-shaped cloud denoting fine weather; margor (K) is a cloud 
appearing during the north-west season betokening fine weather ; a rain cloud is baiib 
or bmib (W.), golegole bae (E.) ; torob (E.) is a wind cloud with a little rain ; a road 
or path is iaba-gud (W.), hence goigoi iaitu-gud is the expression for white stratus clouds 
at sunset, and kiAUau, kulkan, and midpalan iaim-gud for dark, red, and yellow stratus 
clouds at sunset ; neder is the Miriam name for a stratus cloud ; a small cloud, " half 
way in sky " is iara eia (W.) ; a clustering of clouds is Toei (W.) ; a scud, a driving 
cloud or squall is rat (W.); while a bright cloud is laruai (W.); a cloud on the top 
of a hill is ataer (E); dad is a still white cloud in the night sky, or the Milky Way. 
An invocation for various clouds to gather is given in Vol. Tl. pp. 198, 199. 

There are al86 names in Mer for dark, night, dawn before and dawn after sun-rise, 
light, day, noon, setting sun, red sky at sunset, etc., and there seems to be greater 
discrimination in Mabuiag, as the following list obtained by Mr Ray shews: 

Ar pu, morning twilight [ar zUami, getting light, lit. dawn runs']; ar hiika, early dawn, 
lit. dawn is red ; gidub dada^talamin, aun juat rising, lit. gtdub peeps or comes through — the 
metaphor is taken from the gidub or kernel of a oooo-nut just shewing when the nut is slioed 
off— [goiga or palami, lit. sun outa dawn] ; magi batainga, sunrise, lit. little morning ; goiga 
kadaka-palgin, sun baa risen \^oiga toorogi lanori, sun above stopa]; gotga haclaka-menui, sun 
has gone up [goiga kadaka-ulaig, sun goes up]; goiga dada »enu., noon, lit. sun (in) middle 
there \dada goiga tanori, middle sun stops]; goiga vtagikia paufta atin, about 3 p.m., lit. sun 
a-litde-way gone downward; goiga paupa fuin, later, lit. sun gone downward [^oiga pawpa 
anka, sun downward goes] ; goiga pudaiica mska, sun b^;inning to go down ; goiga Jtup-MUt/t, 
sun sitting down ; goiga pudi, sun fallen [^otga pudiz^ ; kuta buia, evening twilight, lit. ending 
light [^oigaa buia mala punka, inurau padaiginga, lit. sun's Ught still hanging, darkness not 
fallen] ; \magi inur, little darkness] ; inur, darkness ; laihil, night \kubU, inur, darkness] ; dada 
kubil, midnight. 

Being sailors the natives have named the winds which blow from definite quarters 
and these give the directions of space, p. 232. Among the Miriam we find : wag, wind, 
nw, strong wind or storm, wi, squall, balgup, catspaw. 

The Hiriam name the tides, meg, as follows: meg ogeri, rising tide, meg tawerge or 
au m^, flood tide, m^ omarida, ebbing tide, meakep, low tide, au msskep, low spring tide. 

> The Tata eqalTaleols ms givMi ia bncbts. 

d by Google 


DiBEcnoNs OF Space. 

The following directions of apace were recogmsed, the chief determining foctora being 
the direction of winds. For the sake of convenience I have taken them as equivalent 
to the more important points of the compass. 

Uabuiaq. Hibiah. 

Korth: naiffai dogam', sab. 

North-east: naigai id. naiger. 

East: waura dad\ naiger pek*, soger pek. 

South-east: waura dogam, 8ager\ 

South : ziai dogam, gored (or eiai pek). 

South-west: huki ada^, eiad (? logcA). 

West: ^u^' dogam, koki giai. 

North-west: kuki koki. 

Relation op Man to bis SuRBOUNDmos. 

Unfortunately we have vety little direct knowledge concerning the ideas of the 
Torres Straits peoples as to the relation between inanimate objects, human beings, animab, 
plants and inanimate objects. I was informed in Mer that no animal, plant or inanimate 
object had a tamar or spirit (vi. p. 251) as human beings have, but lamar were associated 
in an unexplained way with carved stone images; similarly the mart of the Western 
Islanders (v. p. 365) were the spirits of men. Thus, so far as we know, all the Islanders 
recognised that the possession of a ghost, which ultimately became a spirit and dwelt 
in a land of spirits, was a distinguishing feature of man. This belief was held by the 
Western Islanders who at the same time recognised that a man belonged to the same 
family as his totem. One result of totemisra was to place the totem animal of a clan 
in a different category from all other animals so far as the behaviour of that clan was 
concerned. The relation of human beings to individual stars or constellations is mentioned 
by Dr Rivers (p. 220). 

Numerals, Cocnting and Records. 

Throughout Torres Straits there are practically but two numerals, urapun (W.), netai 
(EL), one, and vkasar (W.), next (E.), two. The first is generally pronounced urapuni or 
warapuni in Muralug and sometimes SrSpun or Srapuni in the Western Islands. Ukasar 
may also be pronounced kiuasar, kOaa, okdsa, etc. Higher numbere are expressed by 
repetition, thus: two one (3), two two (4), two two one (5), two two two (6), after 
which my informants would generally say a lot, fcm (W.), lakuh, gaire (E.). I obtained 
the term badagUi in Muralug for three, but this simply means another one or some. 
Jukes (if. p. 302) says that nets " repeated three or four times rapidly means an indefinite 

' Ada, oataie; dad, middls ; dogaxi, side; pek it equlTAlsDt to -1; in northerly. 

* Aooordinfi to Hr Ba; tager inaludei k11 the winds trom B. to S. fared toga, 

light, to tha 8.E. of Ha there ie nothing bnt the open aea ; »ager ptk, points of the 

d by Google 


large number, twice only means ' a few,' ' Bome,' as we should say, ' three or four.' " 
Nomoa and Waria of Mabuiag also gave me wa (or war) getal (fingers) for five and 
okataT getal for ten. 

For further information on this subject, see Joum. Anth. Inst. xix. 303 ; and S. H. Ray 
in Vol III. pp. 46, 86. 

There are two methods of counting, one by enumerating parte of the body and the 
other by a bundle of sticks or tallies. 

In the first method when only a few counts, say up to ten, were needed they 
counted on the fingers, begiuning with the little finger of the left hand. I am not 
aware that the toes were ever utilised. There was another system of counting on the 
body, commenciog at the little finger of the left hand: — 

The Western Islanders enumerate in this way as follows: 1, kutadimur (end finger); 
then following on with the fourth finger, 2, kuiadimur guranguzinga (a thing following 
the end finger), ring finger ; 3, il get, middle finger ; 4, klak-nitui-get (apear-throwing 
finger), index finger ; 5, kagabet (paddle finger), thumb ; 6, perta or tiap, wrist ; 7, kudu, 
elbow (they point to the inside of the elbow joint); 8, tugu kuik (upper arm head or 
basis), shoulder; 9. suau madu (breast flesh), left nipple; 10, kosa dadir, sternum; 11, 
wadogam euau madu (other side breast flesh), right nipple ; and so on in reveres order 
preceded by wadogam (other side), the series ending with the little finger of the right 
hand. These names were obtained at Mabuiag; those used in Tutu and Muralug are 
somewhat different. This gives nineteen enumerations, of which 11 to 19 are merely 
inveree repetitions of 1 to 9. Nomoa and Waria of Mabuiag in 1898 gave the following 
enumeration : 1 — 5 as above, 6, wadogam kutadimur (other side end finger), 10 being the 
thumb of the right band, 11 is the left wrist, and so on, ending with 19 for the right wrist. 
I believe the former method to be the correct one. The spelling of these names is that 
adopted by Mr Ray in Volume iii. ; in my original publications (JourTi. Anth. Inst. xix. 
p. 305 ; Proc. Roy. Irish Acad. (3) iv. 1897, p. 162) I adopted a somewhat different spelling. 

I obtained the following in Erub in 1889 : 1, kebi ke (little finger) ; 2, k^ eip ke 
(little middle finger); 3, eip ke (middle finger); 4, baur ke (spear finger), index; 5, au ke 
(big finger), thumb; 6, keii kok ne (little bone inside of joint), wnst; 7, au kok ne (big 
bone inside of joint), inner part of elbow ; 8, kenani, armpit ; 9, tuger, shoulder ; 10, 
gSlid, pit above the clavicle; 11, nirk^, pit at the root of the neck; and then passing 
in the reverw order on to the right side, ending with the little finger, using the same 
names; this gives a total of 21. Mr Ray obtained at Mer from Jimmy Rice precisely 
the same enumeration, except that he added: 12, nano, breast or nipple, and 13, kopor, 
navel, making a total of 25 by counting the nSrkSp a second time. 

The mamoose of Mer gave Mr Ray a somewhat more complicated series (iii. p. 86), 
but he was often a somewhat vague person. It is possible that the method may vary 
with different individuals'. 

' The Bev. A. E. Hant tajt: "The ooDDtiDg oommenoed at the little finger ot tha left h&nd, thenoe 
connttog tbe digits, WTiEt, elbow, umpit, shoulder, hollow aboTe the cl&Ticle, tborai, and tbenoe in leverse 
order dowD tbe right um, ending with little finger of right hand. This givea twentf-one. The toea are then 
resorted to and these give tea moie. Bejond this nnmber the term 'gain' (oiuif) would be used" (Joarn. 
Anth. Init. xivni. p. 18). 

H. Vol. IV. 30 

d by Google 


The names are simply those of the parts of the body themselves and are not 
numerals. In counting the part mentioned was touched with the hand, or if a finger it 
was bent down. This system can only be employed as an aid to counting and not as 
a series of actual numbers. In Mabuiag the elbow, kudu, may be 7 or 13, but I could 
not discover that kvdu really stood for either of those numbers, although in a matter 
of trade both men would remember how for along the body the tally of a former number 
of articles extended, and by beginning again on the left little finger the actual number 
of articles could be recovered. Only the older men are acquainted with this method as 
all the numerals now in use are borrowed from the English and spelled phonetically. 
I have experienced a surprising amount of difficulty in getting reliable information on 
this apparently simple subject. 

I have noticed in several islands a decided tendency on the part of the natives to 
count by twos or couples. In Mer seg signifies things tied in a row with string, two 
bunches of bananas, and se<j degari is to hang in a row, tie coco-nuta on a string in 
tens. I suspect the tying in tens is due to modem South Sea influence. Mrs Walker, 
of Badu, informed me recently that a native who had 183 pearl-shells for sale said: 
"17 shell stop in water, suppose they get them, they get 200 shell." 

When it is desired to keep count of a comparatively large number of objects or 
events recourse is had to sticks. These are 
frequently portions of arrow-heads or small 
trimmed sticks, which are usually nicked so as 
to enable them to be tied on to a piece of 
string. A bundle of such sticks is called a 
kupe or kopei (E.), (fig. 224; Album, I. pL 342, 
No. 6), and has generally a stouter stick in 
the centre round which the others can be 
wrapped, frequently it is a broken harpoon dart 
In some specimens certain sticks are marked 
by a piece of rag being tied to them. Occasion- 
ally two sticks are tied together ; on my en- 
quiring what this signified my informant said 
" he have two women that night." The sticks 
are generally rather rough, but sometimes they 
are neatly cut out of hard dark wood ; in one 
example of the latter from Dauar the sticks had 
an average length of 11 mm. 

A tally of this kind was employed for various purposes, such as reckoning the 
number of diigong or turtle killed, number of amours, etc.' Formerly the Miriam men 
did not keep a tally of the dugong caught off the coast as dugong were not common 
there owing to the absence of suitable feeding grounds. Now a few men on Dauar have 

' Jukea says (i. 194); ■' MammooB eoqaired ODe da; of Lieut. Bisk, the namber of all our large veaaels, or 
'ow Bbippo' and amaU onei, or 'kabbj abippo,' aa they call the boats. Od Lieut. RUk eDameratiiig them, he 
took little pieoee of stiok, and made a Uttle bundle of loui, for the Fly, Bramble, PriDoe Qeorge, and Uidge, 
and of the requisite number of amaller pieces for the boats, amoanting then, I believe, altogether to eleven." 

d by Google 


some as they are able to go further afield in fishing, but they are so small that it is 
considered of do importauce. The Miriam are so fond of making a show that they take 
no interest in a tally unless it is a big one. 

Mr J. Bruce informs me that the Miriam had kape for turtle speared or caught in 
deep water only, those turned over on the sand-beach did not count. There was also 
a tally for the number of king-fish, ffeigi, speared, but not for those caught on a line. 
A kvpe was also kept for the number of bunches of bananas that were wrapped up, 
mpem kaha (p. 147). One kupe was employed as a fornication tally, another for aop»m 
kaba, and a third for turtle and geigi, the two counts being separated by a dividing mark. 

The kupe, at all events in Mer, were usually employed to keep count of adventures 
with women, and a large tally was greatly prized. I believe the men used sometimes 
to go over the tally before congenial companions giving the name of each woman or 
girl that a stick represented. The specimen in the Cambridge Museum (fig. 224) contains 
ninety-two sticks. Mr Bruce informed me that " these tallies are left entirely to a man's 
bonoar ; a man may have doubts about the correctness of another's tally, but he would never 
dare single-handed to express an opinion doubting it. Sometimes a crowd may chaff a 
man whom they suspect of cheating and get him to confess that be was only ' gammoning.' " 
One kupe which I obtained in Mer, now in the Oxford Museum, was said to be used 
to denote the number of persons injured by sorcerj'. 

When a man died and was laid out (VI. p. 130), the kupe tallying his illicit con- 
nections with married women and unmarried girls was suspended from a croton plant 
placed at his bead or feet. This was to honour the dead, as the unimportant kupe were 
not displayed, for example those which recorded the number of turtle or king-fish (geigi) 
he had caught or his wealth in bananas. The kupe descended to his son, or &iling a 
son to his next heir, the keeem le. 

The idea of competition and the desire to excel others, to which I have alluded, 
led the Torres Straits Islanders to keep the skulls of dugong or the carapaces of turtle 
as evidence of their skill in fishing. To take but one or two examples: in Mabuiag 
the shelb of turtle were placed on a long platform (agu), and as each canoe had its separate 
agu the crew that could shew the greatest number of turtles at the end of the season 
acquired the greatest glory; a drawing of an agu by a native is given in Vol. v. p. 331. 
In the account of the cult of Kwoiam (v. p. 370) it was stated that there was an active 
rivalry between the two phratries concerning the acquisition of sknlb for the " elder " 
and " younger basket." The lower jaws, however, were private property, and were taken 
to the house of the warrior, where they served as a proof of his valour. 

Although there were no personal monuments the natives had some idea of per- 
petuating their own memory; for example, a large stone bird in Mer was said to be a 
memorial of a certain man, and in the kwod on Tutu there is a large old tree in which 
pieces of bones, mainly of turtle, have been inserted and are now more or less over- 
grown by corky tissue. Part of the plastron, kHpai, of a turtle had been placed there 
by a man named Rusia or Rosir, and another man, Wedai, being exceptionally tall, had 
stuck his iron harpoon dart into the tree 20(3 m. (6 ft. 9 in.) above the ground ; these two 
objects are now in the Cambridge Museum (cf. Vol v. p. 208). 

The decorated skulls of deceased relatives may perhaps be regarH"d as a crude 


d by Google 


attempt at portraiture (Vols. v. pp 251, 258, 362 ; vi. pp. 126, 148), and certaiD scarifications 
were also commemorative (p. 14), as were also the bud lu (VoL vi, p. 158). 


Various articles in daily use or which were occasionally worn had a Tecogniaed 
exchange value, which necessarily depended upon the fineness of the specimen or its 
rarity. These considerations must be taken into account when attempting to make out 
a table of relative values. The following articles were the units of highest value, and 
were approximately of equal exchange value, each being about equivalent to a wife, 
and were used for bride purchase, more especially among the Western Islanders: the 
dugong harpoon (p. 169), canoe (p. 203), cone-shell armlet (p. 56), urcu shell necklace 
and necklace of dogs' teeth (p. 41). The dUndibi pendants (p. 44) were aiso used for 
this purpose. 

The only object which was not obviously useful or ornamental, but which was used 
as currency in Mer, is a cone-shell from which an o has been 
made and of which the top has been removed probably to 
make a dibtdUn (fig. 226). I had considerable difficulty in 
procuring this specimen as the owner said he wanted to buy 
a canoe with it fi^m the Fly River. In the British Museum 
there is a similar wauri (Albmn, I. pL 306, No. 6) 163 mm. 
(6^ in.) in length, which was collected at the Fly River by 
the Rev. S. MacFarlane. The label of this specimen says, " Fly 
River armlet, very valuable, worth a man or a canoe." The 
real interest in these objects lies in the fact that their value 
must be conventional, thus representing the first step towards 
actual money. 

The details of the internal and external trade are given 
in Vol. V. pp. 293—7 and Vol. vi. pp. 185—8. 

No arithmetical processes were employed other than those 
mentioned above. ^''■^- H'-'iri, Mgt, m mm. 

Weights and Measures. 

There is no system of weights and measures. The only unit of length is the fathom, 
hn getalwaga (big with the fingers), kaza (W.), kag (&), which is meaanred from tip to 
tip of the fingers of the outstretched hands; to measure in fathoms is called gOa 
minami (W.), or hand measure. 

d by Google 


Od the whole the houses are kept very clean and tidy, and the spaces round them 
are freed from filth. 

The natives frequently bathe in the sea, but I do not know that they make a daily 
practice of it. Fresh water is boo scarce to be uBed for bathing. Meidu, in the folk- 
tale (vi. 13), "as was her wont, went into the sea to wash herself and to kill the fieas 
in her petticoat When she returned she changed her wet petticoat for a diy one." 

Mosqnitos are very troublesome during the months of the north-west monsoon, 
especially in December, and there were certain men in Mer called lag sogo le who were 
credited with having a control over these pests. To remove the plague they put a 
mixture of coco-nut oil and water, sabid, on the stones of the water-bole at Lakop (vi. 218). 
Saifid is greatly used in all togo ceremonies, but in this instance if the oil spread over 
the water it would form a film which by preventing the larvae of the mosquitos from 
breathing would cause their death. I do not know whether the natives have noted that 
oil spread on water lessens the numbers of mosquitos, or whether it is merely a coinci- 
dence that they adopt the modem scientific method. 

The Miriam recognise that one place may be better than another for a sick person. 
For example Kiar (vi. 13), when he cut his foot and could not get well on his own 
island of Dauar, said to his fiiends, " More better you take me to Ulag, north-east wind 
he come here all the time." So they took him to Ulag on the far aide of Mer. It is 
probable that the salubriousness of this brees^ point of Mer {see Map, vi. p. 170) was 
the real reason why the remarkable practice of curative "medicine" by the kekuruk le 
was performed at Ulag (vi. p. 237). Patients used to resort thither and sometimes 
remained there several days. 

The diseases and therapeutics of the natives are described in Vol. i. 

d by Google 



A. MuRBAi laLAND 236 

1. Introductory 238 

2. Origin and CUssilication of the Songs 239 

a Methods of Analyaia 242 

4. The Songs transcribed in Musical Notation .... 244 

6. Detailed Analysia of the SoDgs 247 

6. Deductions from the AnalTaea 255 

7. Summary 260 

B. Thb Wbbtebn Islanw, and Saibat 261 

C. Oenbbal CoHCLuaioNB 285 

D. Appendix. Words op the Sonus 266 

A. MuRBAY Island. 

1 . IntroductoTT. 

The songs of the Miriam or Murray Islandeie, which form the subject of this section, 
are of considerable interest from the standpoint of musical bistoiy and development. For 
they differ among one another not only in complexity of structure but also in date of 
compoeition and place of origin. They thus afford an opportunity of tracing the changes 
in musical expression which may occur in course of time within a primitive community. 
They also shew evidence of the great traffic in tunes which may go on between the 
inhabitants of neighbouring islands, thus raising the general question as to how far the 
fundamental characteristics of the music of a given people are fixed or are modifiable, 
temporarily or permanently, by the importation of foreign airs. 

The primitive character of Murray Island music is sufficiently attested by the fact 
that the drum is probably the sole native instrument the islanders possess; at all events 
the drum is the only instrument ever used as an accompaniment to their tunes, and then 
only in ceremonial and dance music Flutes, pan-pipes and jews' harps are occasionally 
seen, the first possibly introduced by visiting South Sea Islanders, the two latter coming 
from New Guinea (see following section). But they are now seldom played upon and 
are never used in orchestral combination or at musical festivals : they have evidently exerted 
little or no influence on Miriam music. 

d by Google 

H0SIC 239 

The compass of the majority of the Miriam male voices is baritone. A few have 
a light bass voice ; but none could be described as having a true tenor. At times, however, 
thejr pitch their songs (especially their Christian hymn-tunes) so high, that their voices 
are forced into an artificial compass. The songs which are about to be described were 
sung by men only. There are doubtless children's and women's songs, but these were 
not heard or recorded during the stay of the Expedition. In church, the women emit 
the most piercing treble notes ; the tempo is excellent, but the volume and quality of the 
sound are very unpleasant and there is a general tendency to lower the pitch in the course 
of the hymn. 

By the aid of a phonograph records of many songs were secured, Mid of these twenty 
have been subjected to detailed examination. Some of the songs are religious and ancient, 
otheis secular and relatively modem. Of the secular songs, some are sung at games, 
others at dances. Unfortunately no songs, specially designated as dance songs, were recorded, 
but from what could be ascertained during our visit it is probable that any hvourite non- 
sacred tune (cf. Song XIII a, p. 246) could be sung at dances. The rhythm of the songs 
actually performed during dancing was strikingly independent of that of the dancers' 
movements. In the dance the drum appeared of more importance than the melody. 

There can be no doubt that the tunes belonging to their religious ceremonials have not 
been affected by contamination with European music. It is true that hjnnn-tunes sung 
in the Miriam church have been introduced by Christian missionaries during the last 
35 years, and that lately one or two tunes have been brought into the island from 
the Salvation Army at Thureday Island. But the songs of the Malu cult were so sacred 
that no native woman or child might hear them; indeed no white man bad ever heard 
them before the arrival of the members of our Expedition. It was only with the greatest 
difficulty that the natives were induced to sing them. 

In other words, the Malu songs were sung so seldom and so secretly that it is 
impossible to believe that modem European music had affected them. In securing the 
records for the phonograph, great care, moreover, was taken to ensure that they were 
obtained from the older men who were alive in the times when the ceremonies were 
still being performed, of which these songs formed part. The other ceremonial songs 
were probably also sacred. Traces of European influence may be suspected in one or 
two of the secular songs, but the evidence is not strongly in favour of this view. 

3. Origin and ClaMlflcatlon of the Bonga, 

The first four songs are songs of extreme solemnity belonging to the Malu funeral 
ceremonies. They have already appeared in these Reports (Vol. vi. pp. 151, 152), The 
circumstances and manner in which they were sung are there described {ibid. pp. 145, 
146, 312, 313), and the translation of the words, so far as that was possible, is also given 
{ibid. pp. 299, 300). 

Song I was sung into the phonograph by Ulai. Of Song II three records ate available, 
of which two were sung by Ulai and one by Gasu. The two records of Song III were sung 
by Ulai and Wanu. Song IV was sung by Enoka. 

Song IV A was sung during the exhibition of the sacred masks of Bomai and Malu 
(Vol VI. pp. 289—292, 306 — 308), and again during the dance of the Bezam Boai {ibid. 

d by Google 


pp. 309, 310). The inranslation of the various words sung to this air has already been given 
{Md. pp. 297, 298^), and is here re-printed (see Appendix). Three records of this tune 
have been obtained, two of which were sung by Ulai and Qasu. 

Songs V — XII have also been published (Vol. vi. pp. 152, 153). They belong to the 
kfi>er ceremonies {■Aid. pp. 126 — 144). Songs VI^ VII, X^ appear to have been connected 
with the zera markai (i&td. pp. 133, 134), Songs VIII and IX' with the m^iet siriam 
{ibid. pp. 142, 143, 273, 274). These two songs were sung into the phonograph by Wanu. 
Joe Brown and Ulai were responsible for most of the others. Song XIII is an old song from 
the island of Dauar, sung by Gasu. Perhaps Song XIII A may be included in this group 
of songs. It was sung at the funeral of a nogle {" one of ' the people,' who was not zogole or 
tamileb"). But it was also used as babanet, i.e. as a prelude to a dance. The words were 
introduced from Tutu. 

Songs XIV and XV are sung while the natives are sitting in a circle spinning their 
kolap or tope (see section on Games). Song XVI was composed by a woman named Akoko ; 
it was described as "a new song from I^s." Joe Brown who sang Song XVII claimed 
to have composed it. Song XVIII appears to have been introduced by Boa. It is said 
to have been known at Tutu. The words of these twenty songs are translated into 
English, BO far as possible, on pp. 266 — 268. 

The songs conveniently fell into three divisions, the Malu songs (I — IV a), the 
keber songs (V — XIII a), and the "secular" songs (XIV — XVIII). That the Malu songs 
have been sung in Murray Island for a longer time than the keber songs admits of no 
doubt whatever. It was Waiet who introduced the ceremonies of the zera markai and 
the keber. The natives emphatically state that Waiet came after Bomai and Malu, and 
that he came from the western islands (Vol. vi. pp. 279, 280). 

A study of the words of the Malu and of the keber songs confirms this estimate 
of their relative age in Murray Island. With one or two exceptions the words of the Malu 
songs clearly belong to the language of the eastern islands. These songs thus present 
a striking contrast to the keber songs which, with the exception of Song XI, are invariably 
in the western language. Mr Bay finds many obsolete words and forms of grammar in the 
words of the Malu songs : he has dealt with this subject at length in Vol. III. p. 51. The 
original significance of the Malu songs is for the most part forgotten by the Murray Islanders 
of the present day. They describe some of the words as " Malu words," and attempt to 
translate them into the vernacular (Vol. VI. pp. 296, 297), 

Inasmuch, on the other hand, as the words of all the keber songs (save Song XI) 
belong^ to the language of the western islands of the Torres Straits, they are naturally 
even less intelligible than those of the Malu songs to the Murray Islanders. The same 
feature characterises the majority of the secular songs. These are said to be introduced 
from the western islands of the Torres Straits or from Saibai, often during the voyages 
made by the Murray Islanders while serving on the luggers engaged in the pearl-shell 
industry. The original words of these songs are more or less correctly retained on their 
introduction into Murray Island, but their meaning is lost. The Murray Islanders appear 
1 B; miatake the words "AJr I, p. Ifil," "Air 1," "Air 1, p. 151," sppMr on theM two pages om the tiana. 
Ution or the worda lang during the exhibition of the aoored maska. The onl; words snng to the first aoDg ue 
thoae given and truialated on p. 369. 

' I WM unable to obtain the words Tor these aongs ; at. p. 268. 

d by Google 

HUBIC 241 

mvariably to attach little or no importance to the words of their eon^j they wonld often 
Bay, "it is ooly the music that matters." 

It was seldom that a Miriam (i.e. Murray Island) origin was claimed for the secular 
soDgs, but in some instances this claim appeared questionable upon enquiiy. For example, 
Boa declared himself the composer of Song XVIII. Yet the words are clearly of the western 
language. They were admittedly known at Tutu, and seem to have been introduced 
from Muralug. Similarly Song XV was given to me as composed by Matud, a Murray 
Islander then deceased. But later it turned out to have been introduced by Matud from 
Erub whither it had been previously brought from Masig ; Matud sang it first at a Miriam 
marriage- feast. 

The words of seven secular songs, the music of which was not recorded, were also 
obtained. Of these three came direct from Tutu, a fourth from Saibai, a fifth from 
Aurid or Masig, while a sixth was traceable fr>om Masig, thither from Purem (Paremar), 
and finally from Tutu. The words of these songs are given in the section on Songs; 
they are in the language of the western islands. 

But it is difficult to accept the conclusion, which the evidence thus adduced 
appears to indicate, that the modern music in Murray IsUnd is enttrely of exotic 
origin. There is still the possibility that the Murray Islanders may often have made 
use of the incomprehensible language of the western islands (to the words of a song, 
as has been already pointed out, they attach no importance) and set them to their own 
music. This possibility is strengthened by the words of one song (tunc not recorded) 
which were said to have been brought from Tutu (indirectly from Saibai) " when Oroto 
was ten years old" (about 35 years ago) by men who happened to be sailing under 
Douglas Pitt, a West Indian owner of a pearl-shelling lugger. "Akoko then made 
new music to those words." Akoko, it will be remembered, cUimed (p. 240) also to 
have composed Song XVI, the words of which are in the Murray Island language. 
She was said to dream new songs while asleep, and was evidently recognised by her 
fellow-islanders as a composer of music Joe Brown also appears as the composer of 
Song XVII, one of the words of which is also in the Murray Island language; when 
younger, he had the reputati(m of being the best singer on the island. 

It is conceivable that after long familiarity with the songs and words borrowed 
from the western islands, the Murray Islanders fell into the habit or adopted the 
fashion of singing incomprehensible words to their own music. The words of the Malu 
songs, as we have already explained, are so archaic and symbolic as to have long lost 
most of their meaning; it is therefore not so strange if the custom has grown up 
of singing the words not merely of the keber but also of the secular songs in the 
(more or less distorted) language of the western islands. We may reasonably look 
on the Malu songs as representing ancient Miriam music'. The question arises, of - 
course, as to bow far the kd>er and the secular songs may be respectively regarded 
as specimens ' of " medieval " and " modem " Miriam music. If once it be admitted, 

> It will W rem«mber«d that the colt ot Hala wm introduced from the west, ponibly indeed otiglnftUy 
from New Oninpa (lee theae ReporU, Vol. vi. p. 383), bat aX saoh a remote epoch thkt we may regard the 
Ifa]a tunes m TlrtasU? representative of ancient Miriam made, eren though the introdaoers ot the onlt bioiight 
their tnnea with Ihem. 

H. Vol. rv,' 81 

d by Google 


as I think it must, that some of the modem secalar songs are written hy Murray 
Island composers to the western and a few to the eastern language of the Torres 
Straits, it may well be that other tunee which purported to have come from the 
western islands . really have a Murray Island origin. For aught we know to the contrary, 
it may be that only the words of many of the keber songs were introduced, and that 
the music was composed hy the Murray Islanders, or that, at all events in course 
of time, it became modified to suit their aesthetic needs. If this hypothesis he accepted, 
the Malu, the k^w and the "secular" music may be regarded as representing three 
different stages in the development of music in Murray Island. At the same time, 
however, it must be doubtless admitted that the characteristics of these three kinds of 
music have been determined or influenced by the free communication of the inhabitants 
with the western and other islands of the Torres Straits. 

3. Hethoda of Anal^li. 

The twenty songs which are now to be closely examined have been studied from 
phonographic records, by aid of a metronome and an Appun'a Tonmesser which consists 
of a box of metal tongues any one of which can be made to vibrate at will by 
means of air driven &om bellows. The tongues aVe carefully tuned so as to give 
tones which are successively diflerent by two vibrations. The box contains 66 tongues 
giving as many tones ranging from 128 to 256 vibrations per second. The pitch of 
the tones emitted by such tongues is remarkably constant, despite the inevitable varia- 
tions in temperature and in wind pressure. 

A given song is first written down approximately in European notation. It is 
next subject to more careful examination. The pitch of the more important and more 
prolonged notes is determined as accurately as possible by means of the Tonmesser. 
Any one tone can be prolonged on the phonograph by holding up the lever which 
usually rests on the spiral steel thread and is driven along it. When this lever ta 
held up, the glass style remains stationary instead of travelling along the spiral groove 
cut in the wax cylinder. The mean of several determinations, made both by upward 
and downward changes in the tones of the Tonmesser, is taken as the required pitch. 
The tempo and rhythm of the song, and of the accompanying drum beats when pres^it, 
are at the same time noted by aid of the metronome. 

Then the quotient or ratio of the vibration numbers of successive different notes is 
calculated, so as to determine the interval between them. Supposing that two consecutive 
notes are of 200 and 300 vibrationa respectively, the quotient becomes I'S. The higher 
tone is always selected as the numerator, so as to express the quotient in the form 
of a whole number. 

The size of the interval is also estimated in cents'. A cent is the hundredth 
part of our tempered semitone ; hence an octave is divisible into twelve hundred cents. 
For purposes of comparison of the ratios and cents hereafter calculated fittm the records, 
the following table may be useful : 

> TheM are readily oaJoulated from tha interval tatJM by the aid of the Ublei given by the late A. J. EUis 
In his edition of Helmholtz's SenMatiom of Tone, London, 1B99, pp. 44S— 4S1. 

d by Google 

Tempeied ■emiCone 

Just lenulone (IG : 16) 

jDst miDOT tone (9 : 10) 

Tampered tone 

Just najor tone (8 :9} 

Tempered minor (bird , 

Jait minor third (6 : 6) 

Just mftjor third U : G) 

Tempered major third , 

Jnst fonrtli (3 ; 4) 

Tempered Toorth , 

Jo»t tritoDe (33 : 46) 

Tempered tiitone 

Tempered tifth 

Jnat filth (3 r 8) 

Tempered minor sixth 

JuM minor sixth (S : 8) 

Jast major sixth (3 : 5) 

Tempeied major sixth 

jQHt minor seventh (9 : 16) . 
Tempered minor seventh ... 
Jnst major seventh (8 : 15) 

Tempered majoT serenth 

Octave (1 : 2) 

996 091 



We ahall employ the usual Qomeoclature in referring to the names of the musical 
notes, the tones in the middle octave of the piano heiug written as o', d, e' ...b', those 
in the two octaves immediately below it being written as c" ...b", C, ...5„. The letter b 
ia employed with the customary English signification. 

The numbers and letters which we shall hereafter use having been- thus explained, 
a few observations remain to be added as regards the notation of the songs. Of the 
five Main songs four have already appeared in musical notation in Vol, Vl. (pp. 151, 152) 
of these Reports. The fifth Malu Song IV a was performed during the exhibition of 
the sacred masks and during the dance of the Beizam Boai after the initiation 
ceremonies. Only the words of this song have been published in Vol. vr. pp. 297 — 299, 
where by error in each case they are ascribed to the air of Malu Song I. The 
notation now presented does not always tally exactly with the earlier version, which 
was intended only to convey to the European a rough idea of the character of the 
songs*. In this section an attempt is made to give a more exact value to the pitch 
and duration of each note. The sign + or — over a note means that the pitch of 
the tone is slightly higher or lower than would otherwise be indicated'. The sign v 
indicates a breath pause. Asterisks denote the drum beats. A number of asterisks 
massed together implies that the drum is beaten as rapidly as possible. When two 

notes are connected by two ties 

(- n\ 

a well-marked gltsaando, or continuous 

change of pitch in passing from the one note to the other, is indicated. The songs 
are for convenience sake written an octave higher than they were actually sung. 

> No attempt has been made to convey a very aoonrate idea of gong IT a. It wm felt that a more elaborat« 
system of notation would do more harm than good. 

* The sign + or - is not repeated when snooeoaive repetitions of the eMue nota ooonr. 

d by Google 


4. The Songt tranicrlbed In Musical Notstl^. 


On tb* pUnoforte the notes of thia song ue beat pitiable as a series of deseending whole-tones. 

d by Google 

d by Google 




Digitized by 


6. Detailed Analysis of the Songs. 

Song I. This song conaists of a descent through a succession of intervals, each 
of which (with the exception of the last) is slightly emaller than our whole tone. 
When the descent has been made approximately through an octave, the verae is 
repeated, A sudden rise to an octave may occur anywhere in the song when the 
pitch has become uncomfortably low for the singer. The successive intervals between 
the notes used have the following values: 

YibratioD>ntio8 l-lIS 1-103 1-122 1-108 1-131 1-188 
Ceote 198 168 199 179 197 334 

We have a succession of alternate larger and smaller intervals, averaging 196 and 
173 cents, none exceeding our whole tone, excepting the last where a larger interval 
is employed apparently in order to reach (approximately) the pitch of the original 
starting note. 

If we could treat these differences in size of successive intervals as due to accident, 
we might suppose that it was the purpose of the singer to descend by six intervals, 
each of 200 cents, until the octave had been completed. But, as we shall presently 
see, this conclusion is not warranted by other facts. 

It is, however, certainly the case that a distinct recognition of a "principal tone" 
(? =9 tonic) appears in this song, but the function of this tone is not so much to end 
as to commence and recommence the melody. Each repeated verse starts with the 
principal tone, and the descent is continued until this tone is reached once more. 
The following values in cents are the distances of the successive descending tones from 
the principal tone:— 0, 193, 361, 560, 739, 936, 1160. 

The drum beats are sounded just after the note to which they belong. The 
rhythm of the song is not strictly reguUr until the words emarer, emarer, etc are 
reached. The drum beat is sounded just between the syllables em and an The 
tempo is increased at the start of the descent and remains constant until the song 
is restarted. 

Song II. This song is closely similar in construction to the previous one. The 
transcript here published is derived from the only record of several available, which 
was distinct enough for the pitch of the notes to be accurately determined. In two 
verses even of this record I had to be content with determining the interval between 
say two tones a and c, omitting the interval between a and b, and between b and c, 

d by Google 


owing to the insufficient audibility of b. The following table gives the 
intervals in cents employed in different verses in the same record: 



— 64»- 





Ver« rv. 

343 313 
337 319 







340 337 



The following values in cents represent tbe distances of the various notes from the 
initial note, the latter being given the value zero: 

307 41G 666 B83 IIM 

Here the range traversed is still less exactly an octave than in the former song. 
Indeed it approximates more nearly to eleven semitones (1100 cents) than to six whole- 
tone intervals. « 

The basis of division, moreover, is different. In Song I, it will be remembered, 
an interval of 1160 cents was 'divided into six intervals. In this song an interval of 
1134 cents is divided into five intervals. In both songs the last interval is larger than 
the rest, apparently in order to reach approximately the pitch (octave differences being 
allowed for) of the initial note. 

The drums are sounded as in the first song. The tempo remains fairly constant 
throughout the song. 

()ne of the several versions obtained of this Bong follows the record here published, 
descending (with an octave rise) through eleven semitones, the second verse being sung 
a semitone higher than the first, and the third being similarly sung approximately a 
semitone higher than the second. 

In another rendering the song began on d' and descended to B^ without any octave 
rise during the descent. The second verse began again on ^, descending to d"^. 

In another rendering the song began on c', descending (with an octave rise) so 
as to begin the second verse on b'. In this verse tbe tune descended past the octave 
to below ff„. The third verse began on d' and descended (with octave rise) so that 
the fourth began on e'>. 

These several versions shew how variable are the intentions of the singers and how 
inaccurately they execute their intentions. 

From a study of these two songs there seems little doubt that the general aim 
of the Murray Island singers was to descend through successive intervals until the song 
could be begun again at the initial pitch. In the second song, however, the size of the 
interval is distinctly larger than in the other, so that the octave is traversed by five 
steps in this, by six steps ia the first song. This difference in subdivision is made 
clear in the following diagram: 


Ml MO 730 8M IIM 

[ ''" 1 ■ 

d by.Google 




l-09ft- ? 


1-088— 1*106 



Average i~n2 

1-087— 1-102 

HUSTC 249 

Song in. This extremely simple song consists only of three notes B^, e", d", with 
a broad descent gliasando at the end of each verse. It begins with a prolonged c° and, 
after the intervention of the unimportant ascent to d", descends to B^. I have two 
records (A and B) of this song sung by different natives. The following table gives 
the vibration-ratios of the two intervals formed between these three notes in four verses 
of the song: 

Bboord a. Bbcobd B. 

l-lia ? 

I'lOl 1-121—1-093 

1-180 ? — I-IOK 

1-187 1073— 1106 

1-120 1-094— 1-102 

The interval c° — B^ occurs three times in each verse and varies each time in size, 
generally increasing as the interval is repeated. Both tones tend to rise absolutely in 
pitch, the second more than the first. 

The average value of the interval c" — d' is 190 cents, while the interval c" — Bv 
varies from 153 to 167 cents. We may perhaps suppose that the final or larger value 
gave the singer most satis&ction and represents the interval be really intended to sing. 
It will be noted that the values 190 and 167 closely agree with the two different values 
of the interval used in descending the octave in Song I, viz. 196 and 173 cents. 

The final, resting, note (? = tonic) is clearly B^. Hence we have a scale of three 
notes, Bjf, c°, d°, the distance of successive notes from the tonic being 167 and 357 cents. 
The tempo of this song is not strict. The drum beats follow the enunciation 
of the syllable to which they belong. The long notes are accompanied by a rapid 
series of bests, again not beginning until after the note hue been sounding for a short 

Song IV. This record contains a descent of a large fourth by four notes. The 
three intervals thus sung have the vibration-ratios of 1-135, I'lOi, 1*098, equivalent to 
219, 172, 162 cents respectively. 

Beckoned from the initial note, the intervals consequently have the following cent- 
values : 

319 891 558 

It is curious, but probably accidental, that the interval between the first and third 
notes (391 cents) is almost exactly that of a just major third. 

Probably the lowest note is accidentally sharpened owing to its very low pitch. It 
is obvious from the record that the performer bad great difficulty in singing it. 

The number of descending intervals appears to be entirely at the discretion of the 
singer. Sometimes fewer or sometimes more were sung than occurred in the one 
record which it was possible to investigate with care. Ultimately so low a note was 
reached that the singer stopped ; he then whispered the sacred words given on p. 267. 

Song IV A. This is a very interesting song, in many respects different fi-om the 
other Malu songs. It is far livelier in character, and the characteristic intervals instead 
of being whole tones are fourths and fifths. 

H. Yol. IV. 32 

d by Google 


Apart from the opening phrase, which is repeated, it coosists eaaentially in an 
ascending gliasando through an interval approximately of a foortih and in descending 
by an exceediogty drawn out gliatando approximately through a fifth to a piolonged 
note. This rise to a fourth and descent tturough a fifth are repeated, the pitch being 
consequently lowered by about a tone after each repetition, until the song becomes 
too low for the performers to sing it, or until the words belongii^ to it have been 
exhausted. A series of sacred words (see p. 267) are then whispered and the song 
ends with a number of short high-pitched shouts, bua, bua, bua, as indicated in the 

Consequently it is possible to maintain that in this song we are observing the 
evolution of whole-tone intervals, since a series of (approximately) whole tones is 
obtained from the successive rises through fourths and fidls through fiJUis. I have little 
doubt, however, that this idea is incorrect (see p. 260). 

Four records of this song are available, which will be called A, B, C, D, They were 
sung by different singers. 

We may consider the intervals occurring in this song under five heads: (i) the 
whole tone ascended and descended in the opening phrase, (ii) the approximate whole 
tone separating the difierent repetitions of the main tune, (iii) the abrupt &U through 
a fourth occurring in the opening phrase, (iv) the gliasando rise to a fourth, (v) the 
glissando &11 through a fifth, the last two in the main part of the tune. The first 
four intervals may be considered together in successive pairs, the last separately. The 
quotients for the first pair are as follows: 

(ii) 1-132 1-181 

1-118 1-126 

1-116 1-128 


Avenge '1-138 1-118 

Eqnivftlent to 201 193 oi 

The mean of these three means is 196 cents, an interval closely identical with the 
larger of the two intervals with which we have met in Songs I and IIL 

The following table gives the quotients for the second pair (iii) and (iv): 

(iii) i-8eo 



(it) 1-814 


















Average 1-416 



d by Google 

uasic 251 

The average for the above interrals of (iii) and (iv) amotmta to 1'399 or 581 
cents, a slij^htly flattened tritone (32 : 45). 

It will be noted that the intervals in (iv) tend to become larger and more 
uncertain towards the end of the song. This was noticeable in all the heard versions 
of the song, and maj be ascribed in part to careleesnesa and in part to the increasing 
lowness of pitch of the note Irom which the rise to a fourth was taken. The note 
became so uncomfortably low that the singer evidently exaggerated the rise in his 
effort to get away from it. 

If we omit the intervals (bracketed in the above columns) due to these causes 
the average amounts to 1'361. Precisely the same average is reached if we limit 
ourselves to the fourth sung (without gtitaando) in (iii). The ratio 1*361 amounts to 
534 cents. 

Similar difficulties attend the exact determination of the remaining interval used 
in this song, the glissando descent through an approximate fifth. The two intervals 
bracketed below shoald certainly be excluded ; their smallness is obviously due to the 
bet that the singer had reached so tow a pitch that the descent through the required 
interval was quite beyond his power. 

(T) 1-487 


















age 1-M3 



The average of all the intervals (those bracketed being excepted) amounts to 
1-562 or 761 cents. 

The drum beats immediately follow the note to which they belong. They recur 
irregularly in the opening phrases of the song, but soon take on a regular rhythm, 
only qoickening during the glissando descents which are accompanied by a very rapid 
succession of drum beats. 

Song V. This song, at first limited to the. two notes c', d', contains a descending 
gUasando through a fifth &om c to /°. The song ends on e". Thus the notes com- 
posing the song are e", f, g", c', d. The intervals could not be accurately determined. 
The drums are beaten with fairly regular rhythm, quickening on the last note to a 
rate of about three beats per second. 

Song VI. The essential feature of this song consists of a descent thix>ugh 
an approximate fifth, ft to B^, with the insertion of the intermediate tones e" and d". 
The tune then rises to the octave 6°, falls abruptly to the lower B^, and ends by repeating 
the previous sequence /"%, vC, £,, the common chord of our minor scale. These three 
important notes /"%, d", B^ form the following downward intervals : 

d by Google 


Thus the average values of the intervals amount to 405 and 342 cents, compnaing 
a large fifth of 747 cents. 

The scale composed of the notes of this song runs B^, (T, e", /It, b". 

Song VlL In this curious song, after a prolonged dwelling on f, relieved b; 
two turns, there is a gHasando rise from it through a major third to a° with a &11 
of a major sixth from a° to c°, and hence to £,, the final and fundamental note' of 
the song. The important intervals and their valuee toe thus: 





1-aSS or 862 oentB 

1-710 or 939 oenta 



1-244 or STB oentg 

1-673 or 890 ont> 


Their average values in cents are 370, 910 and 86 cents respectively. 
The scale composed of the notes of this song runs B,, c", (T, /", a". 

Song VIIL This ia an extremely simple song, the note f, adorned with turns, 
being prolonged until the end of the aong which ends by a descent of a fifth on 
the lower B^. We have therefore only the interval f — B^ to consider. The 
vibration-ratio of these two notes is exactly 3 : 2. The ratio is consequently 1*5, 
equivalent to 702 cents. 

Song IX. The tones used in this song are Bj>, e", rf°, f% g°. It starts from d°, 
rising to the fourth g", and then descending through /°f to d", whence it descends 
by two further whole-tone intervals to c" and B^, The notes form a pentatonic 
scale, the fourth and seventh being omitted. The actual pitch of these five notes, 
Bjf, c", d\ ft, g", was 226, 256. 284, 352, 880 vibrations per sec respectively. The 
intervals which ttey successively form are: 

i-iss l-lOV 1'3S9 l-OSO 

Equinlent to 816 179 S71 1S3 oenta 

All these intervals (if the unimportant grace notes be excluded) are actually sung 
in the song. In addition to these, there is the interval of a fourth from <r to ^% 
the ratio of which is 1338, equal to 504 cents. 

The intervals of the scale, reckoned from the final and fundamental note B^ a» 
zero, have tlie values: 

316 S95 766 899 

Song X. The important intervals in this song are the ascent {torn, the prolonged 
note c* to the fourth f°, and the descent fi^m c° through whole tones to A^. The 
values of these intervals are 1339, 1255, equivalent to 505 and 393 cents respectively. 

The notes used in this song run Ap, B^, c°, f. The notes of this scale are identical 
with four of the five in Song IX. 

Song XL Here as in Song IX the scale is pentatonic, consisting of f, g°, a", c, d, 
if we except the once-occurring grace note e. As in that song, the fourth and seventh 

' So little feeling for tonality esUts in these mors that the woid " tonio " Id tbe ssDHe o( " kejnote " would 
be out of place. 

d by Google 

MUSIC 253 

are wantiDg, and the lowest tone is the final note of the song. The iiiterv»1s, how- 
ever, between the third and fourth, and the fourth and fifth notes of this scale, are 
somewhat different. Structurally it coosiste in a play on the notes if, c', repeated a fifth 
below on the notes g', f. 

The important intervals used are tf — c', c' to 3^, ff" to a", g' to /". The actual 
pitch of. the tones f, g", a,°, c', d used is 176, 196, 220, 268, 304 vibrations per sec. 
The intervals which they successively form are consequently: 

1-114 I '138 1-31B 1-lSfi 

Bqoivaleutto 187 300 MS 330 oenti 

or reckoned fix)m the lowest tone f : 0, 187, 387, 729, 949 cents. 

The interval (f to g" which occurs in the song has the value of 1*367, equivalent 
to 542 cents. 

Song Xn. Here again the melody consists of tive notes A„, e", (f, /°3t, g". It 
includes the ascent of a fourth cT — g', the descent of a fourth d" — A^, the descent 
through a major third ft — d', and of a minor third c" — A^. The record, however, was 
too feeble for these intervals to be determined exactly. 

The song opens, like Song IX, with the ascent of a fourth, and, like it, makes 
use of the tone below the initial note, but it ends on the fourth instead of on the 
major third below the latter. 

Song Xni. The notes of this song also were not determinable with accuracy. It 
begins on gt and, after alternating for some time between g'% and a", descends &om the 
former abruptly through a fifth to ct, rises suddenly again to the gt and descends 
by a run of notes once more through a fifth to c°$ on which the song ends, after 
touching on the note b". The important notes employed are 

cit, fft, 0°. b". 

Song XIII A. This song has d" as its important note. As in Song XII it descends 
fit>m (f to a fourth below it, viz. a". It also contains the notes e" and f. 

Song XIV. The tones used in this song are S,, d", e", f^, g", 0° and the octave b°. 
It consists of two descents through a fifth, the first fixim b" to e" with an intermediate 
pause on g' ; the second, instead of starting from b", starts from a fourth below it, 
i.e. from ft to B^ with an intermediate pause on d". Thus in each case the descent 
through a fifth is divided into intervals of a major followed by a minor third'. 

The interval b" — e" has a ratio of 1'488 or 688 cents, which are subdivided into 
thirds of 379 and 309 cents respectively. 

The interval ft to 5„ has a ratio of 1525 or 730 cents. The pitch of d" could 
not be accurately determined. 

The octave appeals true; consequently the interval e° — fH may be deduced — 
218 cents. 

' The bnath QUik after the erotohet /"$ at first sight BDggeeta the onaljaiB of a flith followed bj % major 
third. Bat nheii I heard the aong, it oertaioly eoDvejed to me the oonstrnotion given aboTe. 

d by Google 


Song XV. The remaining songs are of more elaborate form than any of the 
preceding, and in some of them, as I have said (p. 239), we may possibly suspect 
foreign influence. Song XV is of this kind. It consists of two descents through a major 
sixth, the tune starting on <f and descending from e to g", and then rising a whole 
tone; whereupon precisely the same phrase is repeated a fourth lower, starting from a° 
and descending from 6° to if, ascending to e°. It then ascends from 0° through f 
to the fifth 6°, descending from <r to g'. The song, after alternately dwelling on g" 
and a" and descending to e", ends on the foirly well-marked tonic g". Hence the 
important notes in this song are g", a", V, tf , e', a scale which we have already met 
with in Songs IX, X and XI. 

The first major sixth has a ratio 1'675, equivalent to 893 cents; the second major 
sixth has a ratio 1-648, equivalent to 865 cents. The interval of 893 cents is made 
up of the four intervals of 207, 334, 155 and 197 cents. The f^ in the second major 
sixth could not be determined. The first and last of the four intervals making up the 
second major sixth measured 187 and 185 cents. Hence the average values of the 
intervals between the notes of the above scale appear to be: 

or starting from g' the values are 0, 197, 368, 702, 898 cents. 

The descent bam a" to e" near the close of the song amounts to 556 cents, the 
final descent fixim a' %q g" U> 181 cents. 

Song XVL To many this song, like the last, will appear to bear suspicious traces 
of European influence. Despite ita more complex form, however, it retains many of the 
leading characteristics observable in the majority of purely Miriam songs. The descent 
to a minor seventh fixtm e^ to f a the most striking feature of the tune. It is 
followed by a twice repeated descent through two (?) minor thirds. 

Song XVn. Here we meet with the same pentatonic scale as occurs in Songs IX, 
X, XI and XV. If the final note a" be accepted as a basis, the scale runs 

a" i" e'J «■ /'J 

with intemli of SOS SIS S78 903 oento 

These values reckoned from a" become 203, 416, 694, 896 cents. 

The important intervals used are ft — a°, having a ratio of 1*192 or 304 cents ; 
ft — c't, with ratio 1-516 or 720 cents; c't—f'S, with ratio 1-319 or 480 cents. 

The song starts by the ascent of a fourth to ft, whence the tune descends through 
e' to c', and thence a second phrase starts through (° and a° to the fifth below yS, 
ending on a". 

Song XVIII. This song has clearly e" as its tonic It employs the tones 

174 86S 202 168 oenta 

These values reckoned from the tonic as zero become 174, 527, 729, 891 cents. 

d by Google 

Hnsic 255 

The song also etarts by the twice repeated asceat of a fourth, from 6° to «', of 
607 cento, followed 1^ a return to 6° aad the notes e't, b'; whereupon a descent of a 
major sixth is song from c't through b", a", ft to »', which is thrice repeated. 

a. Dednotioiu from the AnolysM'. 

The range of the Malu songs is ill-defined. In most of these songs it amounts 
to an ap{Nroximate octave, but frequently it depends on the will of the singer or on 
the limitation of the compass of bis voice. 

The range of the remaining fifteen songs is shown in the following table : 


Song V. Tl. TIL YUl. IX. X. XI. xu. xm. xm *. 

Bulge min. 7Ui ooUtb miD. 7tb Sth nuj. 6th nutj. 6th maj. 6Ui min. Tth mln. Tth maj. etb 

" SBC0I.1B " SONOI. 


Bulge odUtb maj. IHb tnaj. 9th oatare ootaTs 

Hence the average range of the medieval songs lies between a major sixth and 
minor seventh, while the range of the modem songs is appreciably greater, averaging 
a minor ninth. 

Of the five Malu songs, numbers I, II and IV are in many respects similar. 
They are made up of a descending series of notes, each approximately a whole tone 
apart from its neighbour. In Songs I and II the object appears to be to descend 
until the song can be begun again on the same note as that with which it started 
originally. In this sense the initial note of the song may be styled the "tonic." We 
may then, perhaps, regard these songs as shewing a vety primitive effort in the direction 
of tonality. But except for the purpose of restarting the song the tonic appears to have 
little or no influence. The successive intervals appear to be formed re^idless of it. 
In one song the octave is divided into six, in the other into five intervals. Songs 
III and IV A are possibly of later date than the other three ; they are certainly some- 
what different in character. In III there is more distinct evidence of tonality, in IV a 
there is little or none, while in both songs there are ascending as well as descending 
intervals and well-marked glissando descents. Song IV a is noteworthy for the alternating 
ascents through a fourth and descents through a fifth, and for the twioe repeated phrase 
with which it opens. 

The following table gives the probable values in cents of the intervals actually 
sung in these five songs: 

■ I wish to eipreu my indebtedneM to Mr A H. Fox Stiangwajs, who ha* verj oaiefally read the proof- 
•hMta, for Tarioni crittoUniB, maa; of which I have gladlj availed mjself of. 

d by Google 


It will be DOted that the interval of 534 cents closely corresponds to three intervals 
of 172 cents, and that it is separated by 227 cents from the interval of 761 cents. 
We may perhaps suppose, therefore, that the important intervals sung in these songs 
are (i) a distinctly flat whole tone, (ii) a slightly flat whole tone, (iii) a distinctly sharp 
whole tone, (iv) a distinctly sharp fourth, (v) a veiy sharp fifth. No interval less 
than about 170 cents, i.e. no interval approaching the size of our semitone, occurs in 
these songs. 

Coming now to the heber and " secular " songs, I have calculated the total 
□umber of intervals used in each song of these two groups. I have omitted from this 
calculation the prime (i.e. no interval), rapid turns and unimportant grace notes. The 
following table gives the results: 

■er ATenge uoendiog desoending Mtio of mc 

ing intarTftli intaiTKlit to dew. int. 

19 41 1 : 3-3 

total unmb«i 
ol intervals 

In the 10 ktber aoagu 60 

Inthe5"8eoiiUr"BOiigB 7* 
In kU 16 Knga 184 

Hence there appears to be a distinct tendency in the more modem or "secular" 
music to increase the number of intervals in each song, and perhaps to increase the 
relative frequency of ascending intervals, although in both groups of songs the descending 
intervals preponderate — a feature which is, of course, still more marked iu the five 
Malu songs. 

Putsuing our analysis of the ascending (a) and descending (d) intervals in the above 
fifteen songs still more closely, we obtain the following table: 











a d 

a d 

a d 

a d 

a d 

In the 10 ktbtr fionga 



« 19 

1 8 


S 3 

Id the 6 '• aeanlBi " songs 

13 23 

2 4 

3 B 


6 S 

In aU 19 soDge 



20 42 

2 4 

3 13 


11 6 

9 3 2 1 

From this table' we conclude that a further tendency of the modem or "secular" 
music in Murray Island is to discard the use of fifths and greatly to fovour the use 
of thirds, whereas it eschews intervals approximating to a semitone. We also note that 
the only intervals which occur more fi«quently in ascent than in descent are the fourth 
and the octave. It will be remembered that of the Malu songs Song IV a alone* 
contains intervals appreciably greater than a whole tone, and that one of the striking 
features of this song consisted in its repeated ascents through fourths. 

We may now [ffoceed to compare the various scales which have come to light in 
our analyses of the records. The material, le. the notes of which the air of a given 
song is composed, when arranged in order of pitch, forms such a scale. As there is 
good reason to believe that the intervals of the scale were added to from above downwaid^ 

I Aa it wu impossible to determine the size of the seven thirds in Song XVJ, tbsy are omitted from (he 

* I here esolnde the indeterminate giitando at the close of Song UL 

d by Google 

MUSIC 257 

they are represented below in descending order. The following methods and signs have 
been found useful. 

The "basal" (usually the fundamental) note is transposed to c, and is indicated 
by the value ^. Notes having an importance almost equal to the basal note are 
written p. Notes of little or no importance are written T, P, or C. 

Five of the songs are evidently in the same scale: 

Song IX. Song X. Soag XI. 

The notes, constituting the material of these six songs, obviously form the fiuniliar 
pentatonic scale of c, d, e, g, a. The values of the intervab in cents for five of the 
above six songs are here given : 

SoDg e—d c—e e—g e—a 




766 899 



— 89B 




729 94( 




702 898 




CM 89$ 




728 SOS 

In other words, the distances between the successive intervals 

unonntto 201 191 361 IBfi 292 

It will be convenient to call this series of notes Scale I A. The notes c, d, g, a 
of this scale also occur in Song XVHI, but here / occurs instead of e. We may call 
this scale of c, d, f, g, a Scale I B ; it is based on a descent through two consecutive 
fourths c — g, /• — c : 

A further development occurs in Song XVT, which contains not only the six notes 
comprised in Scales I A, I B, but also the remaining note b, which is needed to make 
a scale corresponding to the Lydian mode' or to our major heptatonic scale. 

I In ^eakiDg of modM, I aX'njt emplo; the uident Oreek, not the eoeleriutloftl (Qregorian), tunnanobtton. 

It will be nnwmbered that th« Graek terms "Dorian," xphiTgiaii" uid "Ljdiui" ooRMpond ratptMydj to 
Uw eoolenwUckl "Fhrjffina," "Doriwi" and "Ionian." 

H. Vol. IT. 33 

d by Google 


This we xnay term Scale I: 

Let OB now turn to six of the remaining songs, viz. V, VI, VII, VILE, XIII and 
XIV. In VI and VIII we find a scale of the notea c, ^, /, g. In Xm the scale 
runs c, ff, a^, ifi, in V it runs c, cfi, e^, at>, 6l>, and in VII c. cfi, ^ (^ =■) g, l^. 

Combining these, we get the Dorian scale c, cSf, ^, f, g, a!>, 6^ (Scale 11), all of which 
notes (if we admit the once-occurring grace note dt*) occur in XIV. Of these seven notes the 
(rf* occura most rarely in the above six songs, while the commonest notes are c, e^, g, }P, 
the interval between successive pairs being a minor third. 

There remains Song XU, which makes use of the notes c, «[*, / 6^ of Scale U, 
but in which al{ takes the place of at', a change comparable to that from the Dorian 
to the Phrygian mode (Scale III): 


We have thus five scales on which these songs are constructed, the first two of 
which are contained in the third : 

BoKle I A. c <J « g a in Songi IX, X, XI, Xm a. XT, XTIL 

„ I B. c d / g a [d Song XTUI. 

.. l.cdefgab\li Stm% XTL 

„ n. eJfJf/gJ>f)fin Songa V, VI, VU, Vm, Xm. XIV. 
„IILe J> f a ^ ia Song XII. 

It will be noted that, with the exception of Song XIV which, I suspect, was 
once a (e6«r song, all the songs belonging to Scales II and IH are k^>er songs, 
while with this exception all the "secular" songs occur in Scale I (including I A, Ib). 
It would therefore appear that in Hurray Island the "major" scales^ are &r more 
often used in the secular songs than are the "minor" scales. At first sight this seems 
to contradict the conclusion reached on p. 266 that not one of the jbefr^r songs contains 

> I nsa the terms "major" mai "minor" seals for oonvenienoe liar* merely to denote tbe mkIm in which 
the third above the bual (luaaUysfandamental) tone ie nujor or minor. 

d by Google 

MUSIC 259 

the interval of the minor third, while this interval is quite common in the secular 
songs and the major third occurs in both groups of songs with equal frequency. 
A moment's considerotion, however, will make it evident that a song in the major 
scale may contain a great number of minor third intervals, while a song in the minor 
scale may contain a great number of major third intervals. Moreover, it must be 
remembered that we are applying the terms " major " and " minor " scales to songs in 
which no definite tonic is generally recognisable. 

Let us now determine the frequency with which these several notes (c, of, d, v, 
«. / 9^ o^i <*. ^< o) occur in the above fifteen scales derived from Songs V — XVIU. 
We find 

e oomnriiig in 15 Mslea 

» .. .. 

*f .. .. 

J>, d&, and e' (each) „ 

It is remarkable that the note h (forming the major seventh) does not occur in any 
of these fifteen scales *. Of the notes which lie below the basal c we find 

b and aI' each „ „ 1 „ 

Thus the order of frequency of the intervals in these scales appears to be (i) fifths, 
(ii) sixths, (iii) fourths ajid thirds, and (iv) minor sevenths. 

It must be remembered that this order does not express the frequency of the 
intervals actually sung in the Songs V — XVIII. It has been obtained by finding the 
material (i.e. the difierent notes) of which each song is composed, by transposing the 
"basal" note of each song to c, by massing all the notes thus obtained into a general 
scale, and by calculating the order of fi^uency with which the notes of this general 
scale, thus obtained, appear in the special scales of the individual songs. If, on the 
other hand, we wish to determine the fi^uency of the various intervals actually sung 
in these fourteen songs, information on this point is readily yielded by the table which 
has been given on p. 266. The order is as follows: major seconds, thirds, fourths, 
fifiihs and sixths, octaves. In other words, the fi^uency with which the various intervals 
occur varies directly with their size. The striking exceptions to this rule are the minor 
seconds, which lunk in order of frequency after the fourths. 

A point of considerable historical interest is affected by these calculations. Whether 
the fourth or the fifth is the earlier interval used by primitive man has always been 
a subject of keen controvert. Some writers have gone so far as to state that the 
descending fourth was the first of all intervals, and that it was subsequently divided 
and added to. Others have urged that inasmuch as the fifth is by &r the more 

d by Google 


conaonant interral, it must have preceded the fourth in date of evolution. So far as 
the Murray lalaoders are concerned, a direct answer may he given to this question. 

For a glance at the ancient Malu Songs I — IV A (p. 244) shews that the 
earliest tones to originate correspond approximately to oar major second, that the initial 
tone was throughout vaguely borne in mind, acting aa a rudimentary tonic, and that 
when the performer had suDg through an octave of descending tones he began the 
song once more. We can also see that the next interval (after the major second) to 
occur was a fourth, which probably was employed in descent earlier than in ascftnt; 
then came the fifth used in descent only. Turning now to the keber and "secular" 
songs we see the same order, viz. major seconds, fourths and fifths, with the important 
' interposition of the thirds between major seconds and fourths, the addition (in diminishing 
frequency of occurrence) of sixths and octaves, and the increasing frequency of ascending 

But, as we have seen on p. 256, the songs, regarded as a whole, shew the clear 
origin of their melody and their intervals in the natural fall or "cadence" of the voice. 
There is a descent from a high tone to one which, as time goes on, becomes more 
and more distinctly recognised as the tonic, and larger intervals first arise as the result 
of a fusion of smaller ones. Thus the fourth apparently arose as a " tritone " (le. as 
a synthesis of three approximately whole tones), at all events when sung as a descending 
interval. For the descending fourths sang in Songs IV a, XI, and XV measure 534, 
542, 556 cents respectively. On the other hand when the fourth begins to be sung 
from below upwards, it closely approaches our ordinary fourth, e.g. in Songs IX 
(504 cents), X (506 cents), XVIII (507 cents). Probably the fifth was subsequently hit 
upon in the same way, at first undetermined by any harmonic relations. In relation 
to this tonic, the tone which is a fourth above it comes to play a less prominent part 
than that which is a fifth above it (p. 259). 

We conclude, then, that there is good reason to believe that in Murray Island 
the use of the fourth preceded that of the fifth, but that with the development of 
the tonic, the note which is a fifth above it is more often used than that which is 
a fourth above it. Further, the tonic is almost invariably the lowest note in the 
melody and its upper octave rarely occurs. Hence it is impossible to consider the fifth 
to have arisen from the relation of this upper octave note to the note a fourth below 
it. We are bound to see here the rudiments of a harmony dependent on the greater 
consonance of fifths over fourths, despite the fact that the Murray Islanders never hear 
the intervals as literally "consonant" (i.e. sounding simultaneously) during their songs. 
They always sing in unisoa That they use the note a fifth above, more frequently 
than they use the note a fourth above, the tonic, can only be due to an incipient 
sensibility to the dictates of harmony. 

As a whole, the songs of the Murray Islanders are very primitive. There is little 
attempt at divisions into phrases, and scant feeling for tonality. The smallest intervfil 
sung is rarely much less than a whole tone. It is certain that quarter tones fail to 
find any place in Miriam music. There is moreover a striking absence of that regard 

d by Google 

MUSIC 261 

for and delij^ht in complexities of rhythm which are often well-marked features of the 
mnsic of many primitiTe peoples. The tempo is usually slow, and nearly all the tunes 
contain several veiy prolonged notes. Where, aa in the Malu songs, a rhythm is 
noticeable, it has invariably a very simple character. But generally there is no strict 
rhythm, the tempo is distinctly nAato, and the music has more the character of a 
ridtatif than of a melody. It is only in the secular songs that some advance from 
this condition has taken place. 

The Malu songs bear a general resemblance to several that have been recorded in 
Australia, especially as regards their lax tempo and their relatively considerable range. 
Their characteristic lies in a descent through a considerable range by an indeterminate 
number of approximately whole-tone intervals. It is not difficult to see how this feature 
has been derived from a prolonged cry or wait, the natural formless expression of sorrow. 
For this reason, doubtless, the mournful nature of the Malu songs is so prominent Such 
music is in striking contrast to that of the Veddaa^, who are also a people that have 
no musical instriimente. Here we find a stricter and more rapid tempo, complexities 
of rhythm shewing evidence of order and method, and a much more limited material, 
most of their songs being made up of two or three near-lying notes. 

But even in the Malu songs there is an attempt to deal musically with the material 
at the disposal of the singers. Even here there is often a certain contrast between an 
initial introductory figure and the mournful rather monotonous cadence of which the 
remainder of most of the songs consists. 

In the kAer songs the rhythmic element is even still more subordinate to the air, 
which, at times indeed, has rather the character of a ridtatif. The tonic (in the sense, 
at least, of a "central" note) comes to be felt with increasing strength. Thirds and 
sixths make their appearance. 

In the "secular" songs we find a wider range of notes, greater tunefulness, and a 
more obvious attempt at contrast and alternation of figures. A given phrase is, crudely 
enough, precisely repeated at another level of pitch. A figure of wide range of notes 
is followed by another' of very narrow range ; there has awakened an elemental but real 
desire for balance. The scale is almost always of a major character, commonly of the 
pentatonic form c, d, e, g, a. With the increasing material at hand, there goes, however, 
some increase of diffuseness and restlessness. Many of the tunes in this group are &r 
less compact than those of the k^er group. The air wanders on without presenting 
so definite a form or purpose. This diffuseness and restlessness are to be observed in 
a still higher degree in various songs of the western islands (p. 264) from which the 
words at least (cf pp. 241, 242) of the above songs have been derived. 

B. The Westebn Islands and Saibai. 
Uy anthropological work in the Torres Straits was confined to Murray Island. 
To Mr S. H. Bay, who visited other islands of the Straits, I am indebted for phono- 
graphic records of the music of Mabuiag, Yam and Saibai. He informs me that he 
knows nothing about the age of these songs, but from its title ("Waiat song — dance 
1 Cf. tb« writar*! ohftptoi on Teddk mDua in The Veddtu, b; C. O. u>d Bread* Seligniaiui (Cunbridge, 
1911), pp. Ul— SOS. 

d by Google 



of dead men") the Mabuiag Song XI is likely to be old, and Song XIII bears the 
title, "Sea and island — song from Ewoiam"; probably this, therefore, is also old. 
Mr Bay's transtatioD of the titles and wwds of the various songs of this group is given 
on p. 269. 

soNQS from: MABUIAQ'. 

Mabuiag III. 

' I faST 

the reoords 
of (he toueg 

retained the oiigiDal omnbering of the Mabuiag songs in the Cambridge ooUeotEoa. Sevnv] of 
were too faint foi the aongB to be tranecribed. The figures nudeniMtb the aotes give the pitoh 
Tibrations per eeoond ; the eigus + And - baie oonseqnaatl; been omitted. . 

d by Google 

J- 120. 

Mabuiag XIT. 

Yak I. 

Saibai I a. 

The two presumably oldest Songs XI and XIII are distinctly simpler in character 
than the others of the Habuiag group. Song XI bears a resemblance to the keber 
group of the Murray Island songs. Like Songs V, XI, XIII of the k^ter group, il 

d by Google 


ooQsiets for the greater part in ringing the changes on two notes a whole tone apart. 
The rhythm is strikingly irregular and there is little feeling for tonality. 

Song XIII has the characteristic gliaaando, first downward and then upward 
through about a fourth, with which we met in Song IV a of the Malu group. There 
is the same sequence of consecutive whole-tone intervals which characterises the Malu 

Of the four remaining Mabuiag songs, Song XIV is obviously different &om the rest 
in itB (to us) greater tunefulness and conciseness, its relatively regular rhythm, and its 
greater feeling for tonality. It bears a distinct resemblance to certain songs of the 
" secular " group of Murray Island songs, but has a range of compass (minor tenth) wider 
than any of those (cf p. 255). It exhibits, however, a feature characteristic of several 
of those Mabuiag songs in which tonality can be detected, namely, a^ desire to avoid 
resting on the tonic. Clearly the note d" would be the natural termination of the 
song, but it is only momentarily sung, the melody springing up a major third and 
then descending a fifth. 

The same avoidance of the tonic, accompanied nevertheless by a feeling for tonality, 
is shewn in the almost equally tuneful but less rhythmical and more diffuse Mabuiag 
Song IX. The main part of the air consists in a descent from 6° through a" and g" 
to e", the natural tonic. But the note e° seldom seems really final ; the melody 
either descends through a fourth to B^ or ascends a minor third or a fourth to g" 
or a". It will be noted that the phrase, a", g°, e", g", occurs also in the Mabuiag 
Song XIII. 

The remaining two Mabuiag Songs III and IV, which are dance songs, are likewise 
very diffuse and of irregular rhythm. If a feeling of tonality be admitted to exist in 
Song rV, the tonic would probably be f, the air being primarily built on the descending 
phrase a", g°, f. But /° provides no resting place for the melody. It at once ascends 
a minor third to a" or else ascends through a fifth to c'. 

So perhaps in Song III the natural tonic would perhaps be e". But it &ils to 
provide a resting place. 

In the first of the two songs from Yam there is a similar feeling of restlessness, 
partly owing to the inherent diffuseness of the melody, partly owing to the absence of 
any clear key-note. To us the song would perhaps end more comfortably on B^; yet 
the air never remains long on B^ but passes to the dominant /". In the second 
song the sequence of descending whole tones a", g", f obviously plays an important 
part, following a tuneful descent from /' through c', 6°^, g° to the lower octave and 
tonic /°. 

As a whole, then, the Mabui^ and Yam songs differ from those of Murray Island 
in greater diffuseness, restlessness and irregularity of rhythm, and in teas feeling for 
tonality. They are also characterised by a greater range of tones. The range of the 
A«ber songs of Murray Island averages, as we have seen, between six or seven tones, 
while the range of the "secular" songs averages between eight or nine tones. The old 
ceremonial Mabuiag Songs XI and XIII aver^;e only four tones, but the modem 
'Mabuiag "secular" songs average ten tones, and the Yam songs shew a range almost 
as wide. 

d by Google 

The average number of intervals ascending and descending in the Murray Island 
and Mabuiag songs may be compared by means of the following table: 

Average number Average number Average nnmbef Prequeno; ratio 

Soiigi of intervala ot uoending ot deMeadiug at anendiDg to 

per lODg interrala iatsrTals denending interraU 

Hoira; lalftnd "seoDlu" Ifi 9 9 1:1-6 

Mkbniftg "»eenlar" 21 8 IS 1:1-6 

A more detailed analysis of the intervals actually sung in the Mabuiag songs 
(comparable to the table already given on p. 266 for the Murray Island songs) 
yields the following data, the columns a and d referring to ascending and descending 

id. i". Sh" a: """» "*■ ■"«■■ "••■■"" "«"- 

ad ad ad ad ad ad ad ad ad 
UabDing "Moolftr" aongB 00 9 81 10 16 60 SB 21 10 00 00 

The Mabuiag, like the Murray Island, "secular" songs make no use of the minor 
second, and in both islands the fourth occurs more frequently in ascent than in descent. 
In Mabuiag, as in Murray Island, thirds and seconds occur iar more commonly in 
descent than in ascent. It is curious, however, to note that, in Mabuiag, the major 
third when it occurs (six times) is always an ascending interval, while in Murray 
Island it is always a descending interval. Of the wider intervab the sixth occurs but 
once; the octave is absent from the Mabuiag songs. 

The two songs frx>m Yam are characterised by the frequent use made of the 
interval of the descending fourth, and by the wide range of tones employed in the 
songs. The interval of a minor seventh occurs once. 

In Murray Island all four " secular " songs are in the major scale. In Mabuiag two 
of the four "secular" songs are in the major, two in the minor scale. In Murray Island, 
only the keit^ songs are in the minor scale; the Mabuiag Song XI, which we have 
already likened to the kd>er songs, lb likewise in the minor scale'. 

The three songs from Saibai stand in marked contrast to the Mabuiag and Yam 
songs we have been considering. They ftre as simple in construction as any of the 
Malu songs, the air consisting in the case of the first song of merely two tones g', f, 
and in the case of the second of three tones a°, ^°, /". The third song is hardly more 
complex, though the number of tones employed is increased to five— 3°, f", e", cf, c". 
In their simplicity and in their use solely of whole-tone intervals within the phrase, 
the Saibai songs resemble the Malu Songs I — IV of Murray Island. But in their 
general effect, particularly in their lack of rhythm and their r^citaitf character, they 
recall rather the keber group of songs. 

C. General Conclusions. 
We are able to divide the music of the Torres Straits into three main styles, 
which for convenience we may designate (i) the Malu, (ii) the ite&er, and (iii) the 
"secular" styles, 

(i) The Malu style is certainly the most ancient of the three in Murray Island. 

■ See tootDote to p. 2S8 tor the nee of the tenua '*mftjor" and "minor M«le" in this Seotion. 
H. Tol. IV. 34 

d by Google 


It pertains to the most sacred initiation ceremonies of the ialandera; and as the words 
to which these tunes are sung are so archaic that they have lost their meaning, the 
extreme antiquity of this style is unquestionable. Yet even here there is a differentiation 
into (a) songs (I — lY) consisting solely of whole-tone intervals^ and (6) a song (IYa) 
in which a pronounced gUssando and the use of fourths and fifths are characteristic 

Similar songs which contain only whole-tone intervals occur in SaibaL We have 
also met with a song (XIII), in which the characteristic gliasando ia a prominent feature, 
in Mabuiag; it bears the title "Song fi«m Kwoiam," and may therefore be considered 

(ii) The k^ter style doubtless originated at a later date in Murray Island than the 
Malu style. Both the Malu and the k^>er ceremonies appear to have been introduced 
fix>m the western islands. But the keber ceremonies undoubtedly ctune hiter, and the 
songs sung in connection with them ret^ the words of the western language. There 
is one old song (XI) from Mabuiag which distinctly recalls the icfrer style. The style 
is generally ricitatif, the notes are often very prolonged, and often much of the song 
consists in a pUy ^pon two or three near-lying tones. 

(iii) The "secular" style may be subdivided into two. Both are characterised by 
a greater liveliness, by a greater range of tones, by increasing complexity of structure 
and increasing feeling for tonality. In the one, however, there is considerably greater 
conciseness of form and (to our ears) greater tunefiilness and tonality than in the other, 
in which, on the contrary, diffuseness, an unwillingness to rest on the natural tonic, 
and the avoidance of large intervab, -are the distinguishing features. The former of 
these styles is (if we may judge from our few examples) characteristic rather of Murray 
Island, while the latter occurs in Mabuiag and Yam. Consequently, if the Murray 
Islanders have borrowed their " secular " music from the Western Islanders, as undoubtedly 
they have borrowed the words to which they sing these songs, they have selected or 
altered the style of the songs according to their own taste. As has already been shewn 
(p. 241) they have at times invented new songs to words of the western language. 
Their own compositions, even if based upon the songs which have at some time been 
introduced from other islands, appear to have distinctive character of their own, to 
which we have called attention in the course of this section. 

D. Appendix. Words ok the Sokgs». 

Worda of the Malu, kdter and " secular " songs obtained from Murray Island. 

Song I. Wan oka o adel Maluel e padnt entarer 
Yea why O holy one Main at the creek aways, 

Song II. IFau weluba o lewerleri>er o Jiteriba tamera 

Yea O pigeon's feather O food our Malu's club 

■ See seoond footnote to p. 266. 

' 1 am indebted to Mr S. U. Ray for the attempts U translation into English ; see alao these Eeportt, 
VI. Tip. 997—399. 

d by Google 

Song IIL Watt I ah ounotn', wau /nfts dwker ewatur 

Yea Irib ye two drink. Yea Idb he sioks it pulls him down. 

Song IV, lb' abara lewer korim aiara letoer 
Jaw his food head his food. 

Fdlowed in a whisper by the sacred words': 

Maiu okaaiti okaaokaok bamtilare tabatnaiiare 

Malu sorry sorry many are troubled many are troubled again 

batapilare tahatapUare bautakUare 

many gmmble many grumble again many cut themselves many cat themBelves 

Song IV A. IFau aka Malvet au adttd Uluti adud tereget 
Yea why Malu very bad man bad teeth 

Warbir^ naukarik UluH Warbir dereble tegwra 
Warbir haul me out men Warbir dug out play 

stand round. 
To the same air are sung : 

(a) Wan baurem kam 'oapa baurem t€ibametalam* 
Yea to fish spear child harpoon to fish spear t 

to fish spear. 

(b) Wau aka Maluet uzer taurameti Warbir' naukarik leluti 
Yea why Malu paddle sticks fast again Warbir haul me out men. 

{c) Wau dtgem kerem dtrapeida itetnadariei* 

Yea bird-of-paradise head is cut off two roll it up in mat, 

(d) Wau e/albol iaba taiatoa imadari SeU' padgege ni gedgege 

Yeu whales they apout here (lat) Seii in the valley fresh water in the place 


(e) Wau weduli gerth ke»ge olaiU Seii' padgege ni gtdgege 

Yea Malu's club 1 in the channel ) Seii in the valley fresh water in the 

place there. 

Followed in a whisper by the sacred words: 

Malu kopa itauado tiaukarik leluti iaaua dararager 
Malu buttocks smear haul me out meu smeared stick on. 

' i have little donbt tbat the irord Inwm is here intended. Imam made Wnrbir (or Warber) vbich U 
a walei-hole at Lu (these RejmTtM, Tol. n. pp. 7, 2S3, 297), bat probably tbe island Wanber is here conrneed 
with Warber. 

* For the dillereat versioiiB ot theu irords and their pouible meaning, eee these Reporu, Vol. vt. pp. 800, 
SOI (tootnoles). 

* This vae tTanaIal«d "all women are ready to cany." Mi Bay anggeetB in place of it taboo mttalam = 
go ont tttita honse. 

' This waa said to be the "HbId word" for Jt(irati = roll or told up. 

* Sdi was the " brother " of Bomai and is also the name of the channel in the reefs between Her and 


d by Google 


Song Y, Kodutba kodiaba moiaba dagata lagiaha aigapa 

To the ring to the ring to fire to place .to there 

»iga»i akamai a ... toaier babafnvla 
from afar ) 1 T 

Song VII. 

Wau htbi 
Yea dark 

M(t »a haibai ita... 
sleep now eyehrowa cover. 


in monotone 


Were tt 

jxre tepe were toaru' gedge toert ndar 
Haliotis BheU turtle on land Tellina shell 


1 tepe were baua gedge taere 
waves on land. 

Song VIII. 

meluba Dvdiie 

) along DaudaL 

Song XL 


gainati uir dimer 
pigeon ornament sew 
tie on. 

Song XII. 

to Daudai 

...Gtbariaba MuUeriaba taiarmavJee Amiaba 
to Gebar to Mukwa comes between to Yam. 

Song XIII. 

Pna jma..., 

»• etc, tokaiba namie^a (! ivtmiadaba) 

Song xni A 

dia...ina wara gi kalapudema tva toaia lana aba wali 

this other there put on back along cooonut they } fishing line 

Song xrv. 

Igia ba ba toaUika 0...icmuru 

name of plant for for a basket plait, 

(perhaps basket) 

Other versions ; 

Umuru Kiwein nmgaia 

along Kiwai along north. 

Idaba vKildHka' indba leriHka' itUUta 
t to basket go with 1 to basket, 

know how 

Song XV. 
Song XVI. 
Song XVII. A'aito 
Song XVIII. Iriboa 

Kolap nab ulai kalap pogaipa kclap nitM wagd (1 walgen) pogmpa 
linning top this go along top fails top yours after fails. 

Babira mena taUeda 
to father always brought back. 

Saiba ala mitge we mitge 
on lip on lip. 

kalaa inboa 
along N.W. 
Waru, il a Weetetn word, means (artle ; It nc 
Hal ni&7 be the Mabuiag mdlusnea. Atika i 
Ptobsbly ter=ree[, and atika=ga with. 

t ma; be Miriam, meaaing a na bird, 
IB "go with." 

d by Google 

Titles or words of the Songa obtained iront Mabtdag, Yam and Saibai. 

Mabuiog III. 

Mabniag IT. 

M&buiag IX. 

Mabuiag XI. 
Mabui>ig XIIL 
Mabuiag XIV. 

Yam L 

Ngata kaba nau puidaik 
I dance eong sing. 

Oana aoffttlau nau 
Oa's play's song. 

Korara ktoUiM puidaik 
t crocodile head sing. 

fFaialana na puidaik 
Waiat's ' song. 

Ur kawa 

Sea [and] island. 

(= charm's) 

Aivaia gulabwn 
pelican in cano 

I, puidaik 
g sing. 


Yam II. Yamaei barid 

along Yam Id. cuscus. 

Saibai L Matoa na puidam 
Mawit* song sing 


Saibai IL Afadub na puidam 

(= charm) song sung. 

' For Waiat see tbew Jteporlt, Vol. v. pp. 49 — 66. 

* A. iMremony for insaring rip« tiuii (of. these Reporti, Tol. v. pp. S48, S49). 

d by Google 


The sound-producing instrnmente — one cannot accurately describe most of them as 
musical — are decidedly Papuan in character. The absence of such instruments is a 
characteristic feature of Australian ethnography (N. W. Thomas, Natives of Australia, 
1906, p. 126). A good example of the manner in which an imperfect observation may 
give rise to an erroneous impression is afforded by the following remarks by Capt. Lewis 
concerning an experience in Erub. He says, "The mustoians, who were at least fifty 
in number, had squatted themselves at a little distance, singing and beating time by 
striking a piece of bamboo with a stick, and others by striking their hinder parts with 
their hands" {Naut. Mag. 1837, p. 760). Capt. Lewis does not refer to drums, of which 
there must have been many on the island ; if this were the only information we possessed 
we should be obliged to rank these natives at the same low level as the Australians so 
far aa sound-producing instruments are concerned. 

There are no stringed instruments, perhaps the flat string of a bow prevented it 
from developing into a musical bow. 

There was no native bell or gong. The bell is solely employed in connection with 
the churches and has been introduced by the missionaries; it is frequently called pat 
from the Lifiian pate, as the first "native teachers" came from the Lo^ty Islands. 
A bell is also called in Mer amulu ^perhaps, dugong rope thing). 

Stiidnlaton, Olappen and Rattlei. 

The kat or katak, green frog, is part of the stem or a flat piece of bamboo, 27 — 31 cm. 
long, in which several deep transverse grooves are cut &irly close together (fig. 226). A clam 

Fia. 321). Kat, U»biiiae; 29 cm. loog. Fra. 227. Marep, Mabuiag, 

d by Google 


ahell (Cyrena, fig. 59) ia Bcraped along the grooves, producing a noise which resembles 
the croaking of the green tree-frog, hence the name of the implement. I was informed - 
in Mabuiag that lads go round houses at night-time during the turtle season (p. 160) 
and make noises with the kat. 

A Btridulating noise was also made by scraping together two rough shells, such as 
clams (Tridacna) (vi. p. 311). 

At Mabuiag we obtained specimens of "bones," which consist usually of two slabs 
of bamboo from 146 to 149 mm. in length and 2 cm. broad (fig. 227). ITiey are simply 
called marep, bamboo, and have been introduced by Europeans. One pair we collected 
consiate of two portions of ribs from the carapace of a turtle. 

Fta. 338. Ktrktr kebtr, or bamboo clapper, Mer; length SO am. 
The kerker keber or kirkir keber is a clapper made of a section of pater, a kind of 
reed or small species of bamboo ; this is split longitudinally except at one end, and a 
portion of the proximal end of one half is out away so as to give that half free play 
upon the other (fig. 228). It was employed in Mer by a tain-maker to imitate thunder 
(vi. p. 199); from its name one would suspect a connection with funeral ceremonies, 
but of this we have no evidence. It varies in length from 44 to 66 cm. (17} to 26^ in.). 

' Fia. 339. SheU mtUe. Her; duuneter ol tlag 13 om. 
Various kinds of rattles were employed. Of these the shell rattles are worn in 
dances or used on ceremonial occasions. As an example of the former the shell fringe 

d by Google 


on belts worn in the aectilar dances (fig. 72) may be given; I am not aware that similar 

. rattles were worn on the legs. 

' The shell rattle shown in fig. 229 was held in the hand and shaken during certain 

ceremonies in which masks were employed. The instrument is named terpa, firom the 
Barbatia shells of which it is made. The shells are attached by a loop to a narrow 
plaited band, which is &stened to a stem of a plant tied into a circle; the whole is 
supported by two pieces of wood tied tt^ther crosswise. 

The objects most frequently used as rattles were the nut-shells, goa, gua (W.), gba 
(K), of the Pangium edule, which I believe are imported from New Guinea. A bunch 
of these nuta is frequently attached to a dugong harpoon, the stem staves of a canoe 
ornaments held in the hand when dancing, etc., or a bunch may be held in the hand 
when dancing; several single nuts are generally tied on to wantp drums and masks. 

Fio. 980. Rkttle tatie of gba DutR and bj Hagnr, Her. 

Fig. 230 illustrates a hand rattle which was used ceremonially by Miriam men who 
personated a spirit named Magur (VL p. 311). It consists of a twisted cane ring, 
18 X 20 cm. in diameter, supported by two crossed sticks bound round with string, 
there are about 85 nuts, each of which is suspended by a short string; most of the 
nuts are ruddled. A rattle of goa nuta, but more like - that of fig. 229, is used in 
beating the bush in pig hunts near the Eatau River {Album, ii. pL 322, No. 3). 

d by Google 


The gigantic pod (fig. 231) of the Queensland bean, kalapi, kolapi (W.), kolap (K) 
(Entada Bcandens), was used when dancing; it ia a very effective natural rattle. 

Fia. 3S1 Pod ot the QueenilHid bMli 

The padatrong or padatring (W.), lolo (E.) 
(fig. 232), is a rattle made of a split bamboo, in 
the hollow of which is inserted a bundle of long 
thin sticks. These are tied round their middle 
with a piece of string, which is further wound 
round and round the bundle and the end passed 
through a hole in the bamboo. When the end of 
the string is pulled suddenly it unwinds off the 
bundle, causing it to revolve rapidly with a loud 
rattling noise. The decoration of the rattle is 
subject to much variation. The ends of the bam- 
boo may be cut square, but usually the lower end 
is cut to a blunt point and the upper end pro- 
duced into a long spur, the tip of which may 
be enlarged and notched (Journ. Antk. Inst. XIX. 
pi. IX. fig. 7 ; Alfmm, i. pi. 341, fig. 16, this is 
61 cm. long); the rind of the bamboo may be 
engraved in the usual native fashion. Those with 
a projecting rod vary Irom 62*5 to 94 cm. (24J to 
37 in.), and those cut square vary from 31 to 57 cm. 
(12J to 22i in.). 

In Mabuiag the padatrong was used only by 
the men and solely during the turtle season. The 
young men used to go round the houses at night- 
time and startle the inmates by working these 
rattles. The rattles were kept on the agu, or 
turtle platform, in the hwod (v. p. 330), and after 
a successful turtling expedition men went round 
the agu clockwise in the daytime pulling the 
rattles (v. p. 333). As far as I could gather 

this instrument was used at night-time by the a b 

Miriam young men in order to fri^ten girla Fie. 283. Bamboo nttlee, Mer. A, to-t om. loDg, 

Doubtless it once had in Mer a cer«monial sig- "PP" P"^'*"" "■* '""^ «"* P*'""^ "^■ 

.~ ... 111% deoorat«d above and below with oassoTonr 

nificance which may have been for^^tten. feather* B, n cm. Where the Am of the 

bamboo is left it shewa yellow, the araas where 
' it haB beet) removed are darkened. 

H. Vol. IV. ». 

d by Google 



Jew'i harp. 

The widely spread Jew's harp, danAiri (W.), darobifri (E.), is found in Torres Straits 
(figs. 233, 241), where tt ia made of bamboo; those made for us in Uer were of large 
size and were decorated with typical punctate patterns. In properly made specimens 
the tongue does not reach the end of the slit ; a knotted string is passed through a hole 
in the handle and is tied by a fibre to the tongue at about one-fiflh irom its base, 
sometimes the &ee end of the string is provided with gba rattles. The narrow end of 
the harp is placed in front of the half-opened mouth and the string is repeatedly smartly 
pulled and released, the player at the same time breathing strongly. 



Fio. 233. Jew's baip, Met; 37 cm. long. 

Fto. 334. Cracking irhip, Mer; length of stlok 
M5 mm., Btring 335 mm., mid-rib S30 nun. 

There is a Jew's harp in the British Museum (C.C. 6523) which is supposed to come 
&om Torres Straits, it is about 355 mm. in length {AUmm, i. pi. 317, No. 6). A Jew's 
harp similar to that of fig. 233 was obtained in Mer and figured by Brockett (Narrative, 
1836, pi. UI. fig. 4); he says, "John Ireland told me that it was brought from New 
Guinea" (p. 22). A Jew's harp is called pslhipe at the mouth of the Fly River {Album, 
11. pi. 197, No. 4). 

OracUng whip. 

The pipedu (W.), lolo (E.) (fig. 234), is an instrument for making a cracking sound, 
which I have not seen elsewhere. It coosiets of a thjn, ^Apering piece of bamboo ranging 
from about 345 to 746 mm. in length, to the point of which a string is attached from 

d by Google 



345 to 650 mm. long, and this is &8tened to the end of a palm-leaf mid-rib which variea 
from 63 to 86 cm. in length. The bamboo is held in the left hand, the mid-rib is doubled 
up along the string and their junction ia held with the thumb and forefinger of the 
right hand. The right hand pulls the string so that the bamboo is bent as &r as it 
will go and then suddenly released and the lash makes a sharp crack. This appears to 
be simply a toy. The specimen figured in the Album (I. pi. 346, No. 8) was made 
by a Saibai boy in Mer in 1889. 


As bull-roarers occur on the mainland on both sides of Torres Straits (v. pp. 218, 221) 
it is natural that they should be found in the islands. Even in 1888 there was no 

Fio. 2S5. Uodels ot btgo naed t< 
(71. p. 197). 

Models of bigo aoed in turtle » 
raonieB, Hei (n. p. 31S). 

d by Google 


toDger, so &r as I could discover, od original bull-roarer to be found, but then and 
ten years later I was able to get a number made for me. 

There is a good deal of variation in the shape of bull-roarers as seen in figs. 235^237^ 
but they are generally of an elongated oval or lanceolate form ; one end is generally 
nicked or produced into a small knob, to the neck of which the string is tied, but in 
some specimens the end is perforated. They are often painted red, white and black as 
in fig. 237, or red, yellow and white as in fig. 236. The specimen from Moa shewn in 
pi. XII. fig. 8 is also brightly coloured in red and blue and is slightly carved. 

Fia. 387. A — D, models ot tiigu m uwd In turtle oeremonies in Mftbaiag, the vertioal ah»dins indioatos 
red paint and the crow-hatching black paint. One'Sighth natural eiie. E, wanei. 

It was only in Muralug that the bull-roarer, wanS« (pL XII. fig. 7), was employed 
in the initiation ceremonies, and it was only in this island that it was reckoned so 
sacred that no woman might see it (v. p. 217). 

In Kiwai (v. p. 218) the bull-roarer, madtibu, though employed in initiation ceremonies, 
ia at the same time (associated with good crops of yams, sweet-potatoes and bananas. 
With the exception noted above h, til-roarer in Torres Straits, so far as I am aware. 

d by Google 


18 solely associated with the procuring of food or of rain, which amounts to the same 
thing. In Mabuiag a small gardea shrine, an arch or booth, was erected within and 
outaide which bull-roarers, bigu, were hung. Uttle wooden human figures were placed 
in the ehrine, which at night-time became alive, took the bull-roarers and marched round 
the gardens singing as they went to make the ptante grow (v. p. 346). In Yam also 
the bigu "belonged" to sweet-potatoes and yams ns well as to turtle. Some Salbai boys 
in Mer in 1889 called a bull-roarer madub. 

In Mer bull-roarers were definitely connected with rain-making. The stone images 
of men, dmom, employed in rain-making ceremonies (Vi. pp. 194—198) were generally 
provided with one bull-roarer, bigo, or even two (vi. pi, VIII. fig. 1 ; pi. XI. fig. 3), in 
one specimen of the latter case they are blackened on one side and whitened on the 
other doubtless to represent rain clouds, they are respectively 103 and 107 mm. long, 
I obtained models of a small lanceolate type of bull-roarer, bigo (fig. 236), which I 
undeiBtood was used in connection with rain-making. They are made of hard wood, 
unperforated and undecorated, and are tied to a cord about 61 cm. long, the other end 
of which is fastened to a stick (about 91 to 107 cm. in length); the three bigo figured 
measure 125 x 37 mm, 180 x 35 mm., and 247 x 35 mm. 

So far as I know, only the Kauralaig employed bull-roarers for raising a wind 
(v. p. 352), In both Moa and Muralug " a big man who savvy " could raise a wind 
by very rapidly whirling a small thin bull-roarer, wanes, attached to a long siring 
(fig. 237 E). 

The employment of the bull-roarer in ceremonies connected with turtle-fishing is 
possibly merely an extension of the practices connected with horticulture. The turtle 
was a very important article of food for the Western Islanders, and, as we have seen, 
horticulture was relatively unimportant. There is therefore not much cause for surprise 
if a custom originally connected with increasing the supply of vegetable food has been 
transferred to increasing the supply of animal food, or the reverse may have been the case. 

Preparatory to starting out to catch the floating turtle the Mabuiag men took a 
bull-roarer from the agu, or turtle platform (v. p. 330), and swung it over the canoe 
(fig, 214); they also stood round the agu and whirled the large and small buU-roarers, 
bigu and vxmea (fig. 237), A performer whirled a bigu many times round his head, and a 
wanes was at first swung in the same manner, but after a few revolutions it was lashed 
backwards and forwards and was thus made to produce more than one kind of noise. 
When the canoes were returning, if the look-out man saw that they had been successful 
he whirled a wanes. On arrival the men first went round the agu clockwise swinging 
bull-roarers and pulling rattles; if they marched in the counter direction the turtle 
would swim away (v. p. 333). The bull-roarers were used solely during the turtle season, 
aurlal. One informant in Mabuiag said that swinging a bigu was lucky for a canoe when 
sailing, probably meaning by this that it enabled the crew to get turtle. 

When a turtle was caught in former days in Mer it was placed on its back on 
the beach and a number of men carrying large bull-roarets, bigo, walked round it three 
times " widdershins," or counter-clockwise (vi. p. 213), Models made for me of these bigo 
have a large elongated oval form (fig. 236) and are painted, as shewn, in red and yellow 
ocbre spotted with white or red; one is 355 x 115 mm, and the other 406 x 127 mm.; 

d by Google 



the Strings are about 107 cm. in length and the sticks about 137 cm. ; the latter are 
carved at the upper end into a series of roughly cut inverted cones, painted alternately 
red and white. These specimens were definitely stated to be similar to those that were 
employed in the turtle ceremonies. At the present day the Ingo is merely a child's toy, 
but it is not a common one. 

SuHpended from one of tiie stem poets of the Mabuiag c 

referred to on p. 215 I foand 

a perforated oblong slat of Tarnished black wood (Gg. 238) which must have been obtained 

from a foreign source. Its form and notched edges curiously resemble certain 

English bull-roarers. I was informed that it had no significance, but was 

put on the canoe for ornament, indeed Anu had picked it up. It is not 

a bull-roarer, and 1 refer to it merely to shew how easUy one may be led 

astray by the superficial resemblance of one object to another especially 

when a specimeo is simply collected without data. 

Illustrations of Western Papuan buU-roarere will be found in Vol. v. 
fig. 30, Album, II. pis. 200, 201. 


There are two kinds of drums in Torres Straits, the warup and 
the fmruburu (W.), boroboro (E.). Most were certainly made in Daudai, 
New Guinea, and imported into the islands', but I beheve that rarely 
certain islanders have made drums for themselves. I have seen one 
of these and it was undoubtedly a poor specimen. The drums are made 
out of a single piece of wood. The warup variety especially shews a 
high degree of skill in carving the form and in hollowing out, and 
the lines and general finish prove a fine artistic sense. It was more 
particularly characteristic of the western islands. 

One end is always open, gavjet (E.), the other is invariably circular and has a narrow 
rebate, round which the tympanum is &stened. The tympanum, pad (W.), is normally 
the skin of a Vananus, si gegur (K) (locally called "iguana" by the Europeans), but I 
have seen the skin of a globe-fish, korig (E.) (one of the Qymnodontes) used for this 
purposa The tympanum is cemented with beeswax (?) and further kept in position by 
a plaited or twisted band of various materials. Usually numerous pellets of beeswax are 
stuck on the tympanum, possibly to improve the quality of the sound. 

Fio. 238. Slst of 
wood lorniiDg part 
of the decoration 
of a oanoe ; 231 
mm. long. 

Fio. 3B9. Sketoh of a warup from Baibai, 
The wm-up (fig, 239; pi. XXVn. fig. 2) is a large houi^lass-shaped drum which 
has an average length of about I ra. (39^ in.) and an average diameter at the tympanic 
' For example, one I obtained at Nagir came from Mawata m'a Tatn. 

d by Google 



end of about 20 cm. (8 in.). The diameter of the central constriction is from about 
one-balf to one-third of the greatest diameter of the drum. The open end ia cut to 
represent an open mouth, the gape of which extends in some cases nearly to the middle 
of the drum ; the border of the upper jaw has a gentle upward curvature which ends in 
a blunt point, whereas the border of the lower jaw is cut in a straight line and ends 
in a square symphysis. Frequently an elongated narrow projection or shelf runs from 
the angle of the mouth some distance forward along the upper margin of the lower jaw 
on each side. 

The upper margin of the lower jaw, and the shelf when there is one, are decorated 
with cassowary feathers, Ovulum shells and often with rattle seeds, gba. The apex of 
the upper jaw is similarly adorned. Tufto of cassowary feathers may be inserted in the 
median line of the upper jaw, as in fig. 239. In a specimen in the Dresden Museum 
(6357) the band of the tympanum is ornamented with cassowary feathers. The patterns 
and designs cut on the drum are described in the section on Decorative Art. 

Fio. 340. Sketohw of thte« bunibtn-ii. A, &om DaudAi, toimetlj in 

B, Dretden (6400), 94 em. long. C, Cftmbridge, 96 

of the Ber. E. B. Savage, 

The bwvburu (fig. 240) is usually more or less cylindrical, with but a slight central 
constriction ; onlike the war-up it invariably has a handle. The tympanic end resembles 
thac of the warup, but the opposite end is simply circular. The average length is about 
95 CHL (37^ in.). 

I collected at Yam in 1889 a drum of the buruburu type, 74 cm. (29 in.) long, but 
the tympanic end is larger than usual and the handle is set quite close to the open 
and narrower end ; the decoration at this end is shewn in fig. 354 ; two cassowaries and 
daibau (p. 136) are engraved at the tympanic end. This drum ia in the British Museum. 

The sacred Malu drum "Wasikor" (vi. pp. 43, 296) is intermediate in form between 
the warup and buruburni types of drums (pi. XII. fig. 2). It has the general form of 
the latter but with a "jaw" at one end, and is without a handle; it also bears some 
resemblance to the Kiwai gama (fig. 242 c). A neatly plaited ratan band surrounds the 
constriction. It is 1'43 m. long and 20 cm. in diameter at the tympanum. 

d by Google 



Jukes gives an illuBtration (l. p. 176) of a dram (fig. 241) which he saw at Enib; 
aseuming the drawing to be correct, it may be regarded as a warup which bears a 
Blight resemblance to the Malu dram. 

Fio. 241. Drnrn sod jev'B harp from Erub ; tiom Jukes, t. p. 176. 

Several varieties of dram have been obtained from time to time at Kiwai island and 
naturally have been so labelled, but it is necessary to enquire whether an object collected in 
a place is actually made there. Wben at lasa on Kiwai island, I made enquiries on the 

Fio. 243. Drums froQi the moath ol tfae Fly Biver. A, Eiwai, 93 cm. (36} in.). B, EiwtJ, 103-3 cm. 
(40} in.), coUeoted by Be?. J. Chalmers, see fig. 360. C, uld to oome from Eiwai, bat probably 
from Dibiii, 81-6 em. (32 in.), see fig. 359 r. D, Dibiri, 96-S nm. (38 in.), ne fig. 360. 

subject and found that the indigenous drum (Kiwai gama) is cylindrical and decreases gently 
in diameter from each end towards the centre; there is no handle; the open end may have 
three or two small triangular jaws (fig. 242 a, b). The tympanum is frequently made of the 
skin of a snake, ior. These drums vary in length from 90 to 120 cm. 

Another type, the Dibiri gama, which is commonly stated to come from Eiwai, is cylindrical 
but decreases slightly in diameter to the neck ; the open end is provided with small slightly 

d by Google 


divergent jaws ; there is no handle. The tympanum is fastened on to the drum by meane of 
a band (fig. 242 c o). The average length of thia drum ia about 90 cm. {35} in.) and its 
average diameter about 11 cm. (i^ in.). The first drum of this type to reach this country 
was the specimen obtained by the "Fly" at "Pigville," which Macgregor {C.A. 1. 1892, p. 53) 
bfts shewn to be Bebea on the northern side of the eastern mouth of the Bamu River, It is 
now in the Brit. Mua. (CO. 8833) ; it ia 797 mm. {31| in.) long. I was informed at laaa 
that these drums carae from Dibiri at the northern mouth of the Fly River, and Chalmers 
collected an example 87 cm. long on Wabuda Island. Several drums of thia type have a 
tympanum made of the skin of a globe-iish {probably a Tetrodon). All the above are called 
tfoma. Their decoration is described later. 

About 380 miles upstream, in what I term the middle region of the Ply River (Dec. 
Art B.N.G. p. 76), at the " Villagio dei cocchi," d'AIbertis appears to have collected two "lai^ 

Fia. S4S. Fly Biver drum. 1-8 m. in length and 21 cm. in oiraiimferenoe at the tympsDnm. 
Home, 2STG (d'Albertii, u. p. 369). 

but very roughly made druma" (d'Atbertis, ii. p. 137) which are respectively 1 and 1-8 m. in 
length (fig. 243). Their decoration is referred to later (figs. 363, 364). Both are of the 
bunbtvnt type, but with the raised handle bands that characterise the drums from the 
Papuan Oulf. 

What appears to be a variant of the Dibiri gama was collected by d'AIbertis from an 
unrecorded spot on the Fly River (li. p. 269, Nos. 2, 4; Dee. Art B.N.6. pi. V. figs. 80, 81); 
in another spenmen the jawed mouth has become circular. 

The drum of the aggressive, warlike Tugeri, who live by the coast of Netherlands Kew 
Guinea near its eastern boundary, is similar to that shewn in fig. 243, but the handle ia much 
longer, One I measured was I'll m. long. Dlustrations of this drum will be found in 
Sir W. Macgregor'a British New Guinea, 1897, p. 69 (the drum to the left), and in a paper 
by Dr J. D. E. Schmelte, Inlemat. Arck. /. Elhrwgriiphie, xvi. 1904, pi. XI. fig. 6, pi. XV. 
fig. 2. 

In the FapD&n Oulf district the drums, aopa or apa, are slightly constricted towards the 
middle, and there is a central handle and well-marked toothed jaws. I have described several 
and figured one in the Dee. Art B.N.G. pp. 128—129, pi. VIII. figs. 133, 134. 

Whlstlei and pipefl. 

We obtained only three kinds of whistles of the " key whistle " type, but doubtless 
others occur. 

The persok persok' is the dried fruit of Lufla graveolene which ia used as a whistle 
in Mer by blowing across the orifice (fig. 244). The fruits vary in length from about 
55 to 75 tom. 

' A blae-bottle &j or a loctut Ji called pinok (B.) probabl; from the noise it makes. 

H. Vol. IV. og 

d by Google 


At Mer we collected a remarkable whistle, auper lu (fig. 245), which consists of a 
piece of driftwood from New Ouiaea in which there is a Datural tubular cavity. 

FtD. MS. WbiBtlt, auper lu, 113 x 165 mm., Mer. 

I do not know what the watu whistle of Mabuiag or the Miriam komelag is like. 

I was assured in Mer that pan-pipes were formerly made, but I have never seen 
an old specimen. It is worth noting that I was informed that one kind of pan-pipes 
was introduced many yeara ago by some Tanna men (New Hebrides) who landed at 
Ugar in Mer from a three-masted ship. We 
collected in Kiwai Island' five pan-pipes, 
pictgo, which have six pipes, the longest and 
shortest pipes of the largest and smallest 
specimens are 196, 112 mm. and 128, 83 mm. 
The pipes are bound together by a broad 
banana- leaf band, to which each pipe is 
lashed (fig. 24i6 a). We may take these as 
typical One Miriam specimen has six pipes 
which are bstened together by two narrow 
cane bands, the longest pipe (fig. 246 B) is 
135 mm. and the shortest 76 mm. The other 
has only four pipes, 105 mm. — 85 mm., and 
a single band ; but I lay do stress on this 

specimen aa perlmpe Ihe Tanna men intro- p„ ^ v^-^i^ A. Kl,^ B. ito. 

duced a new form of lashing. 

We obtained several flutes, pupui (W.), burar (E), ftwm Mer; these were made of 
bamboo or reed and range from 425 to 735 mm. in length. The tube is so cut as to 
have a node left at one end which is perforated, a short distance from it are two 
boles in the barrel; there is a V-shaped notch at the other end (fig. 247) on the 
same side as the holes. One specimen has no nick. 

lovm. 1910, p. 36t) flgarsB two pbn-pipes with lax pipes trom the Fly Biver uicl 
IB emitted by them. 

d by Google 


I was infonned at Mer that flutes were introduced by South Sea men. I must 
leave it an open question whether there was an indigenous instrument of this nature. 
At all events we collected &om the Cen- 
tral District of B.N.G. a similar flute 
without a nick from Horemore, Uaiva, one 
(iTietne) from laawi, Mekeo, decorated with Fra. 247. Plate, Mei ; 125 mm. long. 

characteristic burnt patterns, three from 

Barapada (N. of Port Moresby), and a decorated one from Port Moresby. There are 
also two specimens in the Cambridge Museum from Collingwood Bay with two holes 
in a square counter sinking, they have no nick. 


In sounding trumpets the lips of the blower are so disposed as to constitute reeds; 
the only instrument of this kind is the giant Fusus, &m (WO, maber (E.), or occasionally 
a large Triton. The Fusus is universally employed, and so far as I have seen the 
mouth-hole is always lateral (fig. 248). It was employed for conveying signals, but now 

Fin. 24S. Old Bbell trumpet, bit, from the hcod at Fala (v. p. 3) ; 41 am. long. 

at all events is most frequently blown when the natives are sailing, especially when 
going fast or racing. There is a Triton trumpet, u, in the British Museum from the 
mouth of the Fly River, which according to Chalmers is used for calling to arms and 
for frightening away the evil spirits of sickness from the village {Album, 11. pi. 201, No. 3). 

There are no instruments with a single beating reed or with a free reed, but there 
are two that may be classed as do^ible beating reed instruments. The first is simply 
a leaf of the karbi tree (W) which is doubled up and blown through between the lips 
in Mabuiag, but doubtless other leaves are employed there and elsewhere. 

The neabgir or burar is made by splitting a very small bamboo, the split end is 
put in the mouth and blown ; it was employed in one of the death ceremonies in Mer 
(vj. p. 139). The average length varies from 17"5 — 20 cm. (7 — 8 ins.), the extremities 
being 16'5 and 23 cm. 

d by Google 


Unfortunately our material k br too imperfect to deal adequately with the 
native songs aud, as elsewhere, in many cases some or all of the words of the song are 
either obsolete terms or borrowed words of which the singers did not know the meaning, 
this is especially the case among the Miriam. Mr Bruce says that "many of their songs 
are merely words to them which they cannot explain as they say the language is foreign. 
They look and act so seriously in singing them that I am inclined to think at times 
they must understand them, but whenever I try to analyse them they say ' Oh ! ite 
only words, but the tune is good." With the little they know of the Quraulaig (Western) 
language they make an attempt at adapting it to their own, but it is only guesswork 
and what they consider was intended by the bard in composing the song. They really 
do not consider the words as of any account. They will sing hymns in Samoan just 
as seriously and with as much gusto as if they understood the language; they are quite 
satisfied if the air pleases them." 

In my original account of the Western Islanders I expressed a belief that there 
were "certain clans whose more especial function it was to sing the chants at the 
dances, etc." (Joum. Anth. Inst. xix. p. 380); this was probably the case, and on Mer 
we found that at all events during the Bomai-Malu ceremonies the Zagareb le were the 
umrup le, drum men, who alone could beat drums, and who sang the songs and led the 
dances, hence they were sometimes called wed le, song men (vi. p. 287). Originally the 
Meaurem le and E6met le of the north-west of Mer (vi. p. 170) were the owners of 
the asaaem wed, hence they were called asawm gtz le, asaeem origin people, but they 
have from time to time given them to other groups so that now all use them. The 
aaasam wed are the laments sung at a funeral (vi. p. 130) in which anyone can join. 
The beating of the drums is strictly confined to the old men, whom Mr Bruce calls 
■meter le. The singing of the dirge or lament is kept on by relays during the whole 
night ; men can scarcely speak on the following day as their voices are quite worn out 
by the continuous strain. Mr Bruce says they are passionately food of their songs 
whether joyous or sorrowful. 

Songs may be accompanied by drum-beating on ceremonial or festive occasions, and 
may be sung, spoken in recitative in a humming tone, or even muttered. It is impossible 
to draw a bard and fost line between different classes of songs, with perhaps the esception 
of those composed and sung for personal delectation, and possibly the war songs. Singing 
in one form or another enters into nearly every ceremony; some are distinctly funeral 

d by Google 

SONGS 285 

or death songs, others are associated with definite cults such as the hero-cults, in which 
case they may be regarded as hymns, others again are what we should term magical. 
It is not easy to distinguish between u, magical song and a magical formula, wenewen 
(W.) (v. pp. 329, 350—352), zogo mer (E.) (vi. pp. 203, 229, 243, 264), or kog mer 
(vi. p. 221). The following references may assist the reader: war songs, v. p. 303 
death and ftineral songs, v. pp. 62, 74, VI. pp. 131, 134, 143, 150—153, 299—303 
ceremonial songs, v. pp. 66, 334, 340, 342. 346, 350, vi. pp. 214, 296—299, 302—303 
magical songs, v. pp. 16, 29, 72, 77, 217, vi. pp. 196, 198; various songs, V. pp: 58, 69, 
75, 94, 210, 332 ; see also pp: 266—269 of this volume. 

Western Songs. 

The following are examples of secular songs sung for personal delectation. They 
were given to me in Mabuiag in 1888: — 

Zanania dri uiidema gika dria 

At Zana dri broke foam along dri. 

The translation given to me was " The spray breaks on Zaua (Passage Islet) like the 
white feather head-dress (dri)." 

Baxi idi Uiga auxipa vhoha Bauia xdi laga umia 

Sea oil dwelling-place goes t along Bau oil place leaves 

I was told that this meant "There is plenty of sea near this village (of Bau, the 
village of Mabuiag), fine water where he stop." I think this must be a pun on the 
word bau, I suspect that the reference to oil is intended to convey the idea of a smooth 
oily sea. 

The following were sung to me by a Muralug man in 1888, he was a native tracker 
who spoke English fairly well. I took down his words verbatim: — 

"I can't pull the canoe round the point, the wind is too strong. I will have to 
stay here a year, for I can't get round the point. I don't yet know when it will be 
fine weather, so when I get fine weather I will go round the point. I want to see 
how the people are getting on there, then I come back again." 

"I got one fish on line, the one fish I got I lost, then I heave the sinker; every 
sinker I got I lost. I got ten sinkers and lost all besides my hooks; every hook I got 
I lost all the time, I could not get any more hooks than that." 

I once heard one of my informants, a Muralug man, humming to himself, and after 
much peiBuasion he told me he was singing ; — " See that man he got a cock (penis)," 
"I sing thing belong woman," "I kill man to-day." These were disconnected sentiments 
each of which was repeated a great number of times. Other Muralug songs are "Walk 
along bush to find, find," " Your hair look bad, you never clean it at all." 

The following six "secular" songs were collected on Mer by Dr Myers; the names 
of the islands whence they were said to have come are also given : — 

d by Google 


Antid or Maalg. 

Gubal tnanni-au (? maluia) -au urge nagua (1 nagi wa) 

Waterspouts alonf^ (interrog.) in sea look yes. 

From Kbulg, thence to Paremar and finally to Tutu. 

2, AloMrtoir i^ Akur itrwer) tamana baya wa lenainia 
thatch hot Same I have (t). 

3. Wait Italia gabueir. 

4. Gazo ka»i go mu puiam ini guda lamaia paradi paradi 

boy along blow penin hole or comes out pull poll 
there mouth 

paraguaa (1 ngapa uzar) mala garnda ini tamai gizv, tamai 

comes only smetllng penis comes out point comes out 

paradi paradi para uaa (1 ngapa uzar) gizu tavtai 
pull pull comes point comes out. 

G, Waier vsaier siiala ita kida magi halbalgiza taliptigaiba 

round little opposite thing scratch with finger-nul T 

Salbai (the words are Miriam). 

6. W«r wer gai maba netuna » 

sea-urchin many conch 

Mr J. Bruce haa sent me the following zera marlcai wed (cf. p. 240 and Vol. vi. 
p. 133). Thia is evidently the same Bong as that recorded by Dr Myeis (No. XI. p. 268). 

U boreta Qano tairo Dime dimey 
Bidoa bidoa eagapa SMi targaba 
Markai ! a\ la gi ai U ziba. 

Assuming these words to be in the Western language Mr Bay has transliterated 
and translated some of them as follows : borsa, bad, ganu-tai, send out smell, ditnidem, 
foolish, bid, petticoat, sagulpa, for a dance, seai-tai, show, gaba, club, markai loffia, markai 
along house, uzaripa (?), go. On the other hand the words collected by Dr Myers admit 
of a different translation if they are to be regarded as Miriam words, but as the kdier 
of the zera markai was introduced from the Western Islands into Mer (Vol. vi. pp. 128, 
133) it is probable that the words are really in the Western language though some 
may have subsequently become modified so as to approach in sound to Miriam words. 

d by Google 

The following four songB are copied from a MS. book written for Mr Ray by Pasi 
of Mer: — 

Smoking Song (Fasi's MS.). 

Abele mer peike eee le gogob iruvxir 
This word here when men to^Mcco smoke 

Abde Mib kabi lurika madera kabi turik moranagidana ttfu nagidana gaborono minn' 
TbiB pipe iron mat sleep nostrile true 

mtgvha tavgaratta' ibela* mtirana pi 

tobMOO a^es 

Sina eaemwia 
Enough it ends. 

Tobacco Song (Pasi's MS.)> 

AbeU mer ko* togoh irutvar 
Thin word for tobacco smoking 

Kematugvha kaTUUuguba inia' (W.)* taxiba (W.) inia taaiha nami abtrabfr tarema ba, 

along with the penia shoot out 

a bert aber tamaba iniini tekurttgu (W.) mada (W.) lapim (W.) mada tapima da, mora koror 
penis playing pudendum sting ray 

uzt node mora sat node mara koro milomilo mora koror pitai ear 
where your post where your your cracks (n.) shell. 

Tobacco Song (Pasi's MS.)- 

Kankere a kaaikere kmtabere a kanabere mudia kanaber modi kanaber urattra kakaper miti 

fiery spark lip 
uaruar (I iruunr) Babe woruar u^ keakeak ade mi rtade aga tararobi toga tarartAi 
marked drink N. to N.E. marked white we £. to 8. 

Mr Bay cannot make anything of this: mttdia and modi may be moder, mat, or 
along house (W.). 

> A W«atwn word. 

* gaborvno is th« Eiwfti word for noitrils, ftiid (an is Eiwai tot flniah. Taugaraiia ta*y b« Eiwai taugo, 
flniihsd, «rana, bum. , 

* Tbia ma; be iptder, U7 down, or the Kivki btda, wh&t? 

■ Tbia would M«m to ni»rk tbe Bong as s modem one, sa the coDStmation is th»t used in the miuiou 
books: h> rcall7=s^n, bat in tbe OoBpele it ii aiwd for the iofinitive: [or this use Mr IU7 ooald obtain no 
•npport from hia informsnls (Vol. ill. p. 73). ko logob truuar=lo smoke tobacco. 

* id is a snfOied particle, a locative of motion (Vol. m. p. 19). 

* Words marked (W.) are Western laogaage. 

d by Google 


A Song (Past'8 MS.). 

AbeU taer peike ese le tag dazirik ahde toed kega 
This word here when man liand draws back This song thus: 

Kobai kobai kerakera gamai a gaviai kerakera io6a tnitupa lagemaua uxdioal kakenipi 
spear-thrower taete 

paUmana mala koikoi banitona wa kale* dumu ne garba keke ideiba kekeo lutpi noka 

open only break yea * go along 

tarapa uaxar biibumola larapa uaz-ar kobaiai a kohatai kerakera gaviai a gaviai kerakera. 

Mr Bay says: "The first line is descriptive and is in Miriam. The song itself is 
unintelligible. It seema to contain words of the Western language, e.g. kobai, tnata, 
wu-ar — tuar (?), go, mitupa ^^ Toitapa {X), to the mouth, banitan, break." 

The Miriam have songs for special occasions such as the r(^ wed (see section on 
Various Social Customs), or when spinning tops (a top song in the Western language 
is given by Dr Myers, Song XV. p. 268), and songs, kamut wed, are sung softly or 
murmured when playing certain cats' cradles (see section on Games and Toys)! 

d by Google 


In ethDological nomenclature the term dancing is employed to describe a large 
range of movements on various occasions. The dances may be grouped for convenience 
as Ceremonial Dances, War Dances, and Secular Dances, 

Ceremonial Dance*. 

The general characteristics of the ceremonial dances were that they were held on 
stated occasions, and were for the most part of annual occurrence. Usually they took 
place in definite spots, for example in the western islands the kwod (v. p. 3) was the 
ceremonial dancing ground. It appears that at Mabuiag the kwod on the adjacent islet 
of Pnlu might be called the national kwod, whereas each of the Gumulaig clans had 
its ovm kwod on Mabuiag. Thus the great death dances took place on Pulu, but subse- 
quently there were local dances in the kwod at Widul and Gumu (fig. 249). There is 

Fio. 349. Drawing b; Snndfty of a death danoe, tarar markai, id the kvod ftt Qumn, Mabuiag. 

reason to believe that each Western clan had its special dances, which in modem technical 
terminology would be described aa being of a tnagico-religious nature. In the Murray 
Islands the analogous ceremonies were as a rule definitely associated with places. Another 
feature of the ceremonial dances was that the decoration of the person was of a defined 
character, and in most cases masks of some kind or another were worn. Sometimes the 
masks were merely lealy disguises, but on the majority of occasions very elaborate masks 
were employed, the general character of which will be described later. 

There is no need to repeat here the accounts of the dances given in Vols. v. and vi. 

H. Vol. IV. 37 

d by Google 


War Dancei. 

I regret to aay that I do not know whether a war dance waa performed before 
going out' to fight, in order to stimulate the warriors, but there is no doubt that war 
dances were held atier successful forays. Id Vol. v. pp. 302 — 4 I have described the 
only two Western war dances that I have seen ; a war dance is also described in 
Vol. VL pp. 274—277. 

Secular Dancei, 

The natives have always been very fond of dancii^, and as the dances are of a very 
energetic character they have considerable value in exercising the muscles of the body 
and limbs. They were important social events and gave opportunities for the girls to 
judge of the activity and stamina of the yoang men (v. p. 222). From statements in 
the folk-tales and from what occurs at present it is evident that visitors to an island 
took a pride in exhibiting their local dances, and there is reason to believe that the 
dances of other islands were occasionally adopted. 

I have designated aa secular the ordinary dance, kap, kab (W,), fcoA (E.), as it is 
now of a purely festive nature, there is however no evidence to shew whether this was 
always so. In some cases ordinary dances are either secularised ceremonial dances or 
they have been infiuenced by them. Dancing has been greatly discouraged by the 
missionaries although, so far as I have seen, the dances did not possess any objectionable 
features; but the "teachers," who are South Sea men, do not discoun^e the dances of 
the Polynesians and Melanesians who reside on or visit the islands, consequently one 
finds at the present day natives dancing alien dances, those of the Rotuma men being 
especially popular when we were in Mer. Dances are now usually called koppa koppa, 
but this is a jargon-English word which has been adopted by the natives. 

It is by no means an easy matter to describe the dance movements executed in 
the kap. Like all semi-realistic dances it is composed of numerous figures which are in 
fact so many separate dancea I do not think that there is any set order for these, 
and the performance may continue for an indefinite time. On all occasions on which I 
saw a kap danced by Western Islanders the pelican dance terminated the proceedings. 
The following are some of the figures, the names given to them are those of the 
Western Islanders : 

The whole company circles round and round the open space, two deep, with all sorts 
of gestures, cringing, swaying, tripping, leaping; the circling may be from left to right 
or vice versa. This figure is called gagai, " bow and arrow." These weapons were carried 
by the dancers, and the dance probably represents men on the war path. 

A man advances singly and dances with a stamping action. At Muralug this was 
called mot asimia, and mui viimi at Mabuiag, " put out the fire." 

In one dance the men continually stand on one leg rapidly moving the other up 
and down. 

A similar one to the foregoing is called n^rara ptmk, in this one leg is raised after 
the other. 

In the ngara taiemiin there is jumping with both legs. 

d by Google 


A mao wearing a f^ head-dress advances, crouches down and vibrates his head 
rapidly, this is called dri ffirer or dri movement 

One dance, karuma tapi (?), imitates the swimming movements of the large lizard 
(Varanus, wrongly known as " Iguana *'). 

In the tadu kap, crab dance, a man dances in a crouching attitude with the upper 
arms horizontal and the forearms vertical, thus representing the way in which the crab 
carries its nipping claws. 

In one figure all the men dance in a circle in single file, either from right to left or 
from left to right. During the pauses in the dancing every m|p performs some definite 
movements which illustrate an action in real life, such as agricultural, nautical or fishing 
employments; for example, a man would crouch and move his hands about as if he were 
planting yams, or he would pretend to look for pearl-shell at the bottom of the sea. 
These movements are well known to the spectators, though the foreign observer may not 
catch the alluBion. Probably most of these actions have become more or less conven- 
tionalised during innumerable dance representations, just as some of the adjuncts to the 
dance are degenerate representations of objects used in every-day life. These descriptive 
movements are the moat common of the figures danced in the Straits, probably the 
majority of the dances were originally imitative, but many have become conventionalised 
beyond recognition by the uninstructed. 

A more complicated figure which I saw at Mawata consisted in the men advancing 
in two lines, one up each side of, the dancing ground; the first pair of men who met 
retreated a little in the middle line, still facing the spectators; when the next two 
arrived the first pair separated to allow them to pass between and the newcomers took 
up their position behind them, and so on, until the last pair passed between the gradually 
extending avenue of standing men. 

As has been previously stated, the avmi kap, pelican dance, concludes a performance. 
The general body of the dancers stand together in the background ; from these two men 
step forward (sometimes one man only) and dance on the tips of their toes without 
advancing; as the drum-beats become more rapid the jumping is accelerated, their legs 
keeping time till with quickened music their feet become almost invisible from the rapidity 
of their movement. It looks as if a hole were being bored in the ground and the dust 
rises in clouds. Naturally this cannot last long and when tired the pair retire, their 
places being taken by another two, and this is repeated until all have displayed their 
skill, and a. splendid exhibition of activity and verve it is. The spirit of emulation is 
lai^ely evoked in this dance, and the onlookers admire and applaud the most vigorous 
and staying dancer of this particularly fatiguing step. 

Kab «n' is the general Miriam name for dancing. The several figures and par- 
ticuUr steps in the dance have their distinguishing names. The general arrangement 
of the dancing ground is shewn in the diagram. 

' I am indebted to Mr J, Brace for this aeooant of a kab eri. 

d by Google 



I 1 

The overture sung with a drum accompaniment before the dance begioa ia called 
babana segur. The music is very slow and ia the same aa that sung at a death, when 
it is called esasera or suna segur wed. 

The dance always begins with the dancers coming into the ground with their left 
bauds towards the centre, but as the dance proceeds they sometimes change and come 
in from the other side with their right hands towards 

the centre ; there seems to be no fixed rule except at oooooooo 

the beginning of the dance. When dancing in hori- 
zontal lines lacing the ^i^mmera each line of dancers 
retires with their left hands towards the centre. When 
the dancers are in one line only, they generally retire 
backwards facing the drummers. When the line 
separates in the centre, they retire to the right and 

Dancers who follow in a circular line are called 
ffir le (gir, boar's tusk), and those in a horizontal line 
&cing the drummers, jtuA ekwat (? to stand like a nose- 
stick); kep le (separate man) is the name for the one 
man or two men dancing singly or together at the 
finish of a figure. Teter Mag ia standing on one foot 
whilst quickly drawing the other foot up and down 
the calf of that leg, as a Highlander does in the " fling." 
Ber didgar ia resting on the baits of the toes with 
the feet apart and working the knees together quickly. 
Kab eupeviar is springing from the ground when dancing. 
Ser ame le (joy wonder people) are those who get up in the excitement of the dance 
and dance through joy beside their child or near relative and sprinkle him with water 
&om a coco-nut water vessel. 

The invariable final figure ia that called kimr or pap kerem, which also concludes 
all k^)er ceremonies (VI. p. 141); the performer dances on his toes, continually springing 
into the air. 

Drs MacDougall and Myers gave me the following information concerning a Icab 
which they saw in Mer. The dancing ground was an oblong space on the sand-beach. 
The drummers sat at the closed end near the sea and here were also the majority of 
the spectators who were mainly women, the remainder sat along each side of the ground 
gradually tailing off to the other end, where the dancers retired behind a fire into the 
shade of trees. There were several small Ares outside the ground, and a man in the 
open apace held handfuls of flaming long dry grass which he waved to and fi^ especially 
holding it below and behind the dancers to shew off their legs. The drummers with 
the singers generally struck up a song, but sometimes the dancers sang a refrain or 
called for a song by name. Each song seemed to be associated with its own particular 
dance and to be accompanied by some story or incident which was illustrated by the 
movements of the dancers. The words of one were said to be "Suppose he iail dowc 
and break him head." 

Diagnun of a Miriun drnDoing groand. 
00, cooo-nnt palms betuod which the 
dftooere prapaie and letire, nertu 
uub, rettiiif! pUee. ->, (he duioets, 
kab le. X X, bei U, men holding 
flaming torohea of di7 oooo-palm 
leaves tn light up the dancers. 
Ill, ths diaromeiB, vamp U, and 
singers, leedakirire U. • • ■, the «pea- 

d by Google 


When the music began the dancers ran into the open space generally along the 
one side, sometimes in pairs, sometimes one af^r the other, to form one long row. The 
dancers as they ran in joined loudly in the chorus, sometimes with other cries; they 
sang all through some dances, but were silent in others. Occasionally one man danced 
by himself. The dancers did not speak to other people, but stood or sat together while 
waiting their turn. All dancing was in a more or less stooping attitude and was done 
on the toes and fore part of the foot only ; the dancers did not assume the upright 
attitude until they returned to the shade. No two dances were exactly alike although 
there is much similarity and repetition. At the end of each dance the music ceased, but 
only for a moment. 

In the first dance, which was called " Pulling up grass," the line of seven dancers 
moved round the circle with the right arm bent and carried forwards and the palm 
extended upwards, the left elbow was drawn behind and the face uplifted to the right. 
Occasionally they crouched down, making clawing movements at the ground with both 
hands as though pulling grass, and holding the arms quite stiff. 

In the second dance, called "The breakeis on the surf," the men carried a aword- 
shaped leaf upright in their bands. 

In one dance the men ran in and formed a long row in pairs down one side 
crouching on the left foot and right knee with the arms bent to the sides, the head 
was thrown back and the face turned upwards with rhythmic movements turning from 
side to side, the rapid movements of the face to either side being accompanied by more 
rapid beating of the elbows on the chest. As the first two reached the end they faced 
the drums, crouched and then retired prancing backwards, sometimes they stopped to 
prance several times while retiring, often close to or over the fire, finally they ran into 
the shade. As each pair took up their position opposite the drums, the others rose 
into a stooping posture and pianced up two pUces, and so on till all bad finished, 

In a somewhat similar dance, which imitated the movements of pigeons, the dancers 
ran round three sides of the dancing ground, from time to time crouching and beating 
their elbows against their chests. Another dance illustrated a man stung by wasps. 

Hr Bruce informed me that there is a Miriam dance, seb ginar, ground dance, in 
which the performers sit on the ground and whilst the drums are beating go through 
all the attitudes of dancing. They are decorated in the same way as in ordinary dances. 

A characteristic feature of all the dances is that they are practically confined to 
the men. Only on one occasion did I see women dancing, and that was in Tutn in 
1888. They danced in a circle widdershins or counter clock-wise (if my memory serves 
me aright) without any violent movements; it may have been a repetition of part of a 
war dance. I have mentioned in Vol. T. p. 303, that it was only in connection with a 
war dance that women ever danced with men, and have there given all the information 
that I have on the subject. 

Dance Parapuernaua. 
In the section on Personal Ornaments and Clothing (pp. 33 — 62) I have given 
sufficient information about the decoration of the person for dances, and illustrations of 
men dressed for a dance will be found in pis. V. fig. 10, VIII., XXIX. fig. 1. It may 

d by Google 



be noted that the groin-shield (pp. 54, 60) is as o 
in front. 

Varioas objects may be held in the hand when 
dancing, sometimes, more particularly in ceremonial 
or war dances, these are efficient weapons or imple- 
ments, but usually they are made solely for these 
festive occasions and are perfectly useless otherwise. 
For eiample, the kap gagai is a bow of slight con- 
struction {Alhvmi, II, pi. 205, No. 6), but when arrows 
are carried they are always real ones. Stone-headed 
clubs may be imitated in wood (p. 194). An imita- 
tion beheading knife and aling <p. 199) were often 
worn suspended down the back (pi. VIII. fig. 5 ; 
Album, I. pi. 342, No. 5). The felse spare bow- 
strings carried in the bracer and the modifications 
which they exhibit are mentioned on p. 57. I have 
collected small models of the dugong harpoon, one 
from Mer {Dec. Art B.N.Q., pi. IV. fig. 55) is 1465 m. 
(4 ft. 9J in.) in length, the butt end of which is shewn 
in fig. 364 D; another in the Cambridge Museum 
measures 2*05 m. (6 ft. 8} in.) and has a dugong 
carved at the upper end, the butt end represents 
the usual head with painted eyes, and at the base 
of the shaft there is a raised beading carved with 
simple patterns. 

A great variety of decorated staves are often 
carried in the hand, some are simple sticks or poles 
variously painted and sometimes decorated with a 
tuft of feathers, others are carved, and though now 
carried in dances which are apparently of a purely 
secular character they have every appearance of 
being reminiscent of symbolic objects belonging to 
ceremonial dances. Good examples of these staves 
are the objects known as gvb in Mer (fig. 250). 
The staves A and D with their raised carved faces 
bear some resemblance to the carved boards, zogo 
haur, used in a turtle ceremony on Dauar (vi. p. 214, 
pi. XXL figs. 11, 12). A waterspout is called gvb, 
and in the kimiar baur or male board {l.c. pi. XVI. 
fig. 11) the lowermost face is connected by a bead- 
ing, haur gub, with a turtle ; in the kosher haur, or 
female board, there is a similar beading connected 
with a terpa, the coral-rock oyster. The baur is a 
spear that is used in catching turtle (p. 156). The 

: behind or on the hip as 

■ held when danoing. Mer. 
A., 1-125 m. long, QS mm. brood mt wideal 
poiDt ; plADo-oonvei, handle roand. B, 
1-66 m. Iouk ; prongs alternately red and 
black, ooDBB alternately red and blae. " 
1-1S6 m. long, diameter of knob 73 mm. 
■piral band is bamt on the handle. 

I-0S5 m. long, BTSrage width 66 mm. ; L 

and white. Vntioal iha^ngsred, bori. 
EOntal^blne, dobi = 7eIlow. 

d by Google 


Western Islanders believe that spirits employ waterspouts, baiu, as their spears for 
catching dugong and turtle (v. p. 359, figs. 75, 78), The zogo baur are analogous to 
the two baiu boards employed in Pulu for a turtle ceremony (v. p. 333), and the words 
of the aong in the Daaar ceremony are of Western origin. It is therefore not unreason- 
able to suppoe^ that these gub are secularised representatives of symbolic paraphernalia 
connected with spirits and waterspouts, and may have come originally from the western 
islands. Fig. B is evidently a conventionalised fish-spear. 

Some Nagir men who danced for me at Somerset, Cape York, in 1888 held a 
curious object (fig. 251) in their hands. The upper part was called a cloud, bSgai 
(p. 231), below this are eyes with feather eye- 
brows, baiib (which is abo the name for a rain- 
cloud)*, the crescentic portion held in the hand 
was called mulpal, the moon. 

The long pod of the Queensland bean was 
sometimes used as a rattle (fig. 231) when 
dancing, and it was imitated in wood in certain 
canoe ornaments, goted kalapi (W.), kolap peapes 
(E.) (fig. 213), which were often carried in the 
hand when dancing. 

I obtained at Tutu in 1888 some wheel- 
like ornaments, jrebiu-za, hand-thing (pi. XXXIII. 
fig. 3 A, b; Album, I. pi. 333, No. 2), which were 

held in each hand when dancing. They consist 

_ ^ij- I.-I.' j,i.i> Pio- 361- Object held in band when dftneiog, 

ot a central disc on which is carved the fore ^^^ 

part of a sucker-fish, gapu, or which is decorated 

with an eye made of nautilus nacre ; from this radiate a number of narrow spokes, 

coloured with red, yellow, white and black bands; the periphery is formed of a thin 

strip of wood into which numerous white feathers are inserted, most of which are cut 

in an elegant fashion. The ring of b is 38 cm. (15 in.) in diameter; in A there is an 

inner ring close to the disc. We obtained in Mer a wooden disc-shaped dance ornament, 

' samuru, carved on one side with a representation of a human face, 162 x 152 mm. in 

diameter. In the legend of Sid (v. p. 29) a dance ornament, sabigarigu, is mentioned 

which consisted of an oval piece of turtle-shell attached to a long handle, the edge of 

the disc being fringed with the black-tipped, white feathers of the Torres Straits pigeon. 

Fro. 262. D&noe aUie iepre«eiiLiri|{ n crocodile, ^ d.b., Mei. 

Models of various animals are sometimes held in the hand during dances, some of 
which are carved in a very life-like manner. For example, we collected at Mer excellent 
representations of cuttle-fiah, gole, crabs, dauma, various fishes, Torres Straits pigeon, etc 

' In 189B WftlUbf or Uurolug Mid that bigai waa the Moa, Yam, Mabuiag word bat panipan nag the 
Monln^ name ; he said that the baiib, ot baia, were augadau baiib and " belong Ewoiam " (t. p. S73). 

d by Google 


A small model of the dugong is carried by the Waier le and Areb le of Mer, it was 
described aa "belong flash," that is ornamental and not ceremonial. Also at Mer we 
collected s dance stave (Sg. 252) which represents a crocodile. 

Mrs F. W. Walker states in a letter that at Christmas, 1909, there was a dance of 
Pareraar and Waraber men at Badii ; each man and boy carried in one hand a very 
clever model of a large wooden fish, painted blue and white, and mounted on a stick ; 
. the dance and words were all connected with this fish, its life in the sea, the catching 
of it, etc. Some of the women and girls joined in the dance, forming rows on each side 
a little distance fi^im the men and boys. 

Mask! and CompoiitA Efflglei. 

Highly characteristic of Torres Straits are the numerous masks and effigies which 
have been obtained on the islands. This art extends to Daudai, and it is probable that 
some specimens labelled " Torres Straits " in museums have come from the mainland ; 
it is easy to determine whether a given anlabelled specimen came from the district 
generally, but not whether it belongs to the islands or to Daudai ; buA (v. p. 55) is the 
general Western name for a mask but the term most generally employed is krar or 
kara, which is also the name for turtle-shell ; kara-asi means to be pliable like krar 
when heated. The Castem name is le op, man's face; I do not know what kind of 
mask the nog or ttbg was (vi. p. 31). 

There is no doubt that certain masks at all events were regarded as sacred objects, 
probably they varied in this respect On more than one occasion I have known natives 
to refuse through fear to put on a turtle-shell mask, as it was not the right occasion 
so to do. Marau of Mer had a face disfigured it was said by kamer tnnar (vr. p. 226) 
because he put the sacred Malu mask, Zogo Malu, on his head. Dr Rivers found in 
Mabuiag that brothers-in-law, %mi, might wear each other's mask, krar, but a man would 
not put the mask on his own head, it was placed there by his imi (v, p. 149). 

Before describing masks which have a definite form allusion should be made to 
the leafy disguises or masks that were worn during the elaborate funeral ceremonies 
(V. pp. 252—258, figs. 36, 40, 41, pi. XIV.. p. 341, fig. 63; vi. pp. 131—133, 139, 142; 
see also Interrtot. Arch. /. Ethnographie, vi. 1893, p. 157, fig. 7 and pi. XIV.). 

Some masks are carved out of a single block of wood, others are elaborately 
constructed out of pieces of turtle-shell. Wooden boxes, pieces of imported boards and 
even kerosene tins have been employed in recent years in the manufacture of not a 
few masks, the foreign materials being used instead of turtle-shell. I doubt if wood 
were formerly also used in the construction of turtle-shell masks except to a veiy limited 
extent and then only in the form of supporting burs or rods. Although the plates of 
turtle-shell are not very large they form an admirable material for this purpose as they 
are thin, strong and light and can be readily cut, engraved, pierced and bent on applica- 
tion of damp heat. The material is also a very beautiful one, but in most cases the 
surface was more or less painted, sometimes entirely so. In museums many masks are 

d by Google 


plain, though probably when in use they were always profusely decorated with featheis, 
shells and rattles. Masks were sometimes made of other materials, as for example those 
made of Hibiscus bark worn by the alag and waiwa tag le of Mer during a ceremony 
following a good crop of enau fruJt (vi. figs. 23 — 25), or the enormous mask of the 
dogai keber of Mer, which was made of the decayed bark or the husk of the nuts of the 
coco-nut palm (vi. p. 143). In the Album, I. pi. 334, No. 5, a human face mask in 
the British Museum is figured, which measures 23 cm. each way and is made of the Bpathe 
of the coco-nut palm, mounted on wire. It is supposed to come from Torres Straits, 
but the wire coiled in a spiral round each eye-hole is a feature which I have never seen 
in this region. 

I have attempted to make a rough classification of these masks, but where the 
maker's &ncy has been allowed free play types are not adhered to. I know of masks 
in several museums which foil into the groups mentioned below, but I have not alluded 
to them unless there is some special feature of interest 

Frequently models of various animals, usually cut out of pieces of turtle-shell, are 
attached to masks; when they become detached they appear in museums as isolated 

Wooden Ma«ks. 

Three wooden masks in the British Museum from Saibai, carved to represent human 
faces, were employed in the mawa ceremony to ensure a good crop of viar fruit 
(V. pp. 348. 349, pis. XVII. figs. 1, 2, XVIII. fig. 1). They are respectively 46, 685 and 
53'5 cm. (18, 27, 21 in.) in length. The masks are hollowed out behind so as to fit 

Fio. 3£8. Man wearing a large maica mask, drawn by Saiiday o( Uaboiag (for the lace marking we p. 16). 

over the foce of the wearer, and a stick is lashed across about a third of the leugth 
from the chin, by means of which the mask was held in the mouth, in case boles are 
made to see through. They are variously painted, one is provided with an imitation 
noee-stick, the ears of two are decorated with Coix seeds, the hair is made of string. 
Three masks in the Dresden Museum were evidently collected at the same time by 
Dr S, MacFarlane, they are described and figured by Dr A. B. Meyer (" Masken von 
H. Vol. IV. ' 38 

d by Google 


Neu Guinea," etc, Konigliches EthnoffrapkiBchea Museum zu Dresden, vii. 1889, pp. 5, 6, 
pk IV. fig. 3, v.); they measure 35x38, 73x36, and 69 x 41 cm. There are two 
similar masks in Berlin and one in Cambridge, the last has a eoake carved on each 
side of the chin. A wooden mask in the British Museum from Mabuiag, 20 cnL high, 
is figured in the AUtum, i. pi. 261, Na 3, it is about 20 cm. high. Qizu said that the 
two mavm masks of Mabuiag (v. p. 349) were made of maivxi wood j hia unsatisfactory 
drawings of these masks shewed that they bad the same character as those of Saibai. 
In a better drawing by Sunday (fig. 263) the man wearing the mask is disproportionately 
small; there is evidently a fringe of Pangium seeds along the forehead. A somewhat 
similar wooden mask is shewn in an Elrub house by Macgillivray (il. p. 47). There is 
an interesting wooden mask with a crescentic ornament on the forehead in the 
Copenhagen Museum. 

Haiki which are or ihould have b«en mode of turtle-thell. 

The turtle-shell masks fall into several categories: — 

1. Small masks which cover only the upper part of the face. These visors or 
ventails were worn in connection with the meket airiam zogo of Mer (VI. figs. 54, 66). 
Possibly they may be in some way connected with Kwoiam's augud (p. 203; 
V. pp. 372, 373). 

2. Masks representing a human face. A good example of a turtle-shell mask of 
this type from Mer is shewn in pi. XXXVL fig. 4 which was identified as a pop le op, 
which was worn in one of the funeral ceremonies (vi, p. 135) ; it is 38 cm. (16 in.) high, 
and is sketched in the Alhum, I. pi. 328, No. 3. One from 
Erub, now in the British Museum, is figured by Jukes (i. p. 178) 
and resembles another fit)m Erub figured on pi. 186 of I'AUaa 
ptitoresgue by Dumont-D'Urville. A somewhat similar mask in 
the Berlin Museum bears as a crest on the head a Iret-work 
cassowaty couchant. Sketches by natives of analogous masks are 
given in Vol. V. figs. 62, 71, 72, and other illustrations in Vol. v. 
pi. XX. fig. 1, and Vol. vi. figs. 56. 59, 60. pis. XVIII. figs. 3, 4, 
XXII. fig. 7. XXVIII. fig. 6, XXIX., XXX., where the occasions 
are mentioned on which the masks were worn ; a number are 
also engraved on a pipe in the Oxford Museum (fig. 256 D). 
Dr Meyer figures a large simple mask, 138 cm. long and 71 cm. 
broad (Lc. pi. I.), and a small one 28 x 22 cm. {l.c. pi. IV. fig. 1), 
both fi^)m Mabuiag. At the tmidu kap of that island (v. p. 339) 
masks of this kind were worn. Fig. 264 represents a performer 
who has put on his shoulders a ring of wood or cane* on which 
he rests the mask, krar, which is evidently of large size. He i 
leaf petticoat, tu, and is canying a bow, arrow and a coco-nut leaf flag, dadu. 

In the ethnographical museum in Copenhagen there is a large turtle-shell mask 
from Dauar, Murray Islands, which is ruddled all over and profusely decorated with 

' So it woa desoribed to me, but ver; likely [t jaa,j be the oyliadrical sapport figured b; Dr Uejer, indeed 
hii ipeoimen mof well be « murfu kap mtak. 

Fio. 3H. Hui dreased for 
the taadti kap, dnwn by 
Weiia of MabniKg. 

i dressed in a coco-nut 

d by Google 


Ovalnm shells and rattles, a Fusub shell is &stened vertically on the forehead, and a 
large Murez is attached to each side ; it is 83 cm. long (K. Bahnson, Etnografien, Vol. l. 
1900, fig. 92, p. 191, and L. Frobeniua, The Childhood of Man, 1909. fig. 205). This 
mask, which came to the museum in 1864 from the Sydney Museum, has a general 
resemblance to the Aurid mask. 

The Aurid mask, which almost certainly represented the hero Kulka (v. p. 378), 
is of especial interest as the skulls of the murdered passengers and crew of the Charlet 
Baton were attached to it; a surviving boy said that the natives held a corrobery 
over the figure on feast days. On plate XXXVI. I have reproduced two drawings of 
this mask, one (fig. 2) fix)m Brockett's Narrative and the other (fig. 1) from the 
Missionary Mag. and Chronicle, May, 1837. Both accounts describe the mask as follows: 
" The body of the figure, it seems, was composed of tortoise-shell and smeared over with 
a red colour, and measured between four and five feet by about two-and-a-half. A semi- 
circular projection stands out from the forehead, made also of tortoise-shell fejicifully 
cut, and when taken fiY>m the island was ornamented with feathers. In the centre of 
the figure, fi-ora the projection upwards, is a small bundle of broken arrows bound 
together. [This may have been the tally of the number of people murdered.] The 
eyes are detached and formed with a silvery shell, something like what is called the 
mutton fish, and the lace is surrounded with shells arranged with method." 

Dr Meyer describes and figures a more complex mask, 80 x 55 cm., from Mabuiag 
(^c. pi. II.). The face of this fine mask is bordered by the pattern shewn in fig, 299 A, 
and the forehead by a key-pattern resembling that of fig, 90 but less elongated This 
mask is intermediate between a simple face and a complex mask, as in front of the 
face is a model of the saw of a saw-fish (Pristis), 

In the Pitt-Rivers Museum at Famham is a mask with a beard and whiskers of 
fret-work, above the head is a large model of a frigate-bird. On the front border of 
each wing is a perforated turtle-shell disc with a central disc of pearl-shell. One disc 
has a ring of half-a-dozen holes, the other is ao cut as to present a ring of V- or 
Y-shaped bars each with a circle cut out of its widest part. The motive is thus 
an&lf^us to but simpler than that of the discs figured on pt. X. 

3. Not a few masks represent a complete animal combined or not with a 
human face. 

(a) Without a human face: — 

One in the British Museum from Mabuiag is an excellent model of a shovel-nosed 
skate, Rhinobatas, kaigas, which was a prominent totem in that island (v. pp. 154 — 164). 
It is 1*22 m. (4 fl.) long. On each side of the mask is a small turtle-shell fish. It 
is supported on two rods the free ends of which, as is frequently the case, are carved 
to represent a bird's head {Album, I. pi. 329, No. 1). The laige crocodile, about 2*1 m. 
(7 ft.) long, in the same museum was made, I believe, for Dr MacFarlane in Mabuiag ; 
it is a very clever piece of work (see Joum. Aidk. Inst xvii, 1888, p. 87, and Album, 
I. pi. 328, No. 1). 

Dr A. B. Meyer describes and figures a good model of a hammer-headed shark, 
Zygtena, kurai, 1'26 m. in length (I.C. p. 4, pi. III.) ; the small ray at the snout is evidently 
a puhai (v. p. 374). I believe it may be identified with the mask referred to in a 


d by Google 



folk-tale (v. p. 54), see also fig. 266 c. The importance of this shark in the ceremonial Ufe 
of the people ia alluded to in Vols. v. pp. 64—66, 154, 155, 374—377 ; vi. pp. 290—292. 
In the Homiman Museum is a mask (pi. XXXVIII. fig. 1) representing what 
seems to be a king-fish, Cybium commersoni, dabor, or possibly a mugarir, Cybiam sp. 
(fig. 315). The body of this fish has a pattern somewhat similar to that of pi. XXXV. 
fig. 2. The mask painted on a Miriam top (pi. XXXVII. fig. 4) seems to be simitar 
to the foregoing, the shai^'s tail-fin is probably an error. These masks evidently belong 
to the type illustrated in fig. 255 a 

Fill. 366. Sketohes ot mMbi referred to ia ft folk-tale (v. p. G4). A, baidam (Btegoatanu tigrinam). B, dabor 
or dtbu (Cjfaiimi oommersoni). 0, kurti (Zfgsnft). Those in the upper nnr were drava by Oiza *nd those 
in the lower row by Sanda;. 

In the museum of the late Sir Cuthbert Peek at Rousdon, Lyme Begis, is a 
beautifully executed mask, probably a dabor, which may well be the original of fig. 265 b. 
The mask (pi. XXXVIII. fig. 2)' is 1-35 m. (4 ft. 5 in.) long, on each side of the tail is 
an engraved design somewhat similar to that of pi. XXXV. fig. 2, and it is decorated 
in the usual manner with feathers and shells. The most characteristic feature is the 
decoration of the cylindrical support into which the head of the wearer was inserted ; 
the ornamentation (pi. XXXVIIL fig. 3) consists of two bands each containing engravings 
of three human heads and upraised arms, the palm of each hand is marked with a 
white spot, in the upper band each figure is separate while in the lower band the arms 
cross each other (this attitude may be compared with the drawing by Fast given in 
Vol. VI. pi, XXVII. fig. 1); there is a single arm at the end of each band. Stars are 
represented in all the spaces. The pattern of the central band is, I believe, unique. 
The whole mask is ruddled and the intaglio portions are filled in with lime. 

I obtained at Yam in 1888 a small mask made by Wadai, who had long been 
dead, it represented the head of a mugarir, or barracuda. 
(6) With a human &oe. 

Representations of masks of this type are found on some pipes (fig. 256, B was 
identified on Mer as a dabor, c from its heterocercal tail is probably some kind of shark). 

On shewing at Mabuiag a photograph of a mask in the British Museum 
(pi. XXXV. fig. 2 and Album. I. pi. 328, No. 2) collected by Dr MacFarlane, I was told 
I Mr C. Orover, o( the BoaidoQ Obseiratorj, tor the photi^nphs korl for •dditioaal 

d by Google 


that it represented the h6ad of a hawk and the body of a fish. It was dreamt of 
by an old man named Pedia and made by Wigi and Anaii of Mabuiag (v. y. 345). 
It is 1'27 m. (50 in,) long, the pattern along the body was called ktitikuti minor, mark 
of the ktdi (a kind of shark). 

no. 266. Tnoing ot muks engraTed on bkroboo tobkooo pipes. A, B, BritiBb Mnwam (6580). 
C, D, Oiltord MDMna. \ n.s. 

Dr Meyer describes {Ic. p. 4) a mask from Mabuiag' representing a kaigaa. It is 
1*06 m. long and is provided with human arms as in fig. 266 b, c; so also was the 
Main mask (vi, figs. 61 — 63) which represented a hammer-headed shark. 

I collected in Nagir in ISSS a mask, krar, made by Gizu of that island, which I 
gave to the Cambridge Museum. The head is 65 cm. long, and was stated to be that of 
a crocodile, kodal, above is a human fitce with bands. Originally it was provided with a 
fish's tail. Behind the head were two vertical triangles of turtle-shell with an apical 
swelling, which were called h!ige or b!igai (p. 296). 

A variety of this type, in which the human tace is replaced by a projecting human 
figure, is supplied by the labur mask of which we have two drawings made independently 
by two natives of Mabuiag (figs. 267, 268); it is said to have been made by Kitulkula 
of Moa, who sold it to the Mabuiag people. The mask consisted of a turtle-shell model 
of a fish, baiig Urn, with long jaws, it is almost certainly a gar-pike (Belonidie), the 
fins preclude its being a king-fish. The projections on the snout are the variegated 
feathers of the poapu bird, and at the tip are feathers of the bird-of-paradise ; feathers 
are also attached to the guy-lines, there are five kaikai (thin sticks to which white 
feathers are ftstened, v. p. 336), and below the jaw are seed rattles. A man named 
labur is placed on top. In Joaoi's drawing a figure called Malu is placed on the fish's 
nose; a screen, kai, is shewn in the background behind which the dancers retired. This 
mask was probably worn in a ceremony which had for its object to ensure a good 
fishing season (v. p. 343). 

4. Masks which represent an animal's head combined or not with a human face : — 
(a) Without a human &ce. 

On pi, 186 of Dumont-D'Urville's Atlas pittoreeque is figured a turtle-shell mask 
from Erub, which consisto of an animal's head without a human iace. 

< This muk Mid the looi meDtioned on pp. 296, 299 were obteined from Dr S. HaoFarlane, thef were 
taken from a "aacred bonse" in Mabniag. 

d by Google 


(b) With a human foce. 
A common type of mask is that of which the lower portion consiuta of an animal's 

d by Google 


into an animal's head, and two upwardly slanting sticks decorated with feathers and 
rattles. The whole mask is profusely decorated with various kinds of feathers (mainly 
cassowary, Torres Straits pigeon and reef-heron) and FangiuTo seed rattles. Maino told 
me in 1898 that the mask was sometimes worn during the marhai ceremonies (v. p. 257), 
it was worn in the kai) (dance), and he asserted that it bad nothing to do with augud 
(totem) ceremonies. A similar mask is described by Dr Meyer (l.c. p. 3). 

(c) Masks representing an animal's head combined with human bee and sur- 
mounted by an animal. 

The four masks of the Saw-fish Dance (v. p. 342) were all of the same pattern 
though differing in minor details. The lower portion was an animal's head with numerous 
small teeth, eyes were variously indicated, on one or two masks curved bands to represent 
gills were painted, at the tip of the snout was an Ovulum shell. Behind the head 
was a fringe of vegetable fibre dyed russet red and brown, and below the snout a fiinge 
of shreddy leaves. Surmountmg the bead was an obliquely placed human face which 
was surrounded by the characteristic open-work border. Along the sides of the face 
and down to the end of the snout were inserted black-tipped white feathers of the 
Torres Straits pigeon which were mounted on thin sticks, about the middle of the mask 
these were very long and bore three feathers, kaikai (a very usual decoration of ceremonial 
head-dresses). Above this was a representation of a saw-fish (Pristis) about 1*37 m. 
(4 ft. 6 in.) in length, its long snout was not only provided with the appropriate paired 
series of horizontal teeth but a double row of similar teeth was inserted so as to depend 
from the under side of the saw. (The origin of this supernumerary row is obvious; 
formerly when the snout was made of turtle-shell the teeth were formed by lateral cuts, 
every alternate tooth thus formed was not removed but bent at right angles to the plane 
of the snout. As this was the traditional method of representing a saw-fish's rostrum 
the double series has been retained, although in the masks now described it entailed 
double the amount of labour.) On the hinder part of the body of the saw-fish were 
represented three dorfial fins and a heterocercal tail; small tufi» of cassowary feathers 
were inserted along the whole upper edge of the fish. Towering above the centre of 
the fish, which consisted only of snout and tail, was a high narrow triangle covered 
with turkey-red twill and flanked with white feathers. The whole erection was steadied 
by guy-lines which extended in front fixim the apex of the triangle to the tip of the 
saw-fish's snout and thence to the fiwnt end of the mask, and behind to near the end 
of the fish's tail and thence to the back of the mask. The upper guy-lines were 
decorated with white feathers and pieces of calico. The total height of the mask was 
about I'37 m. Behind and below the eye can be seen the end of a bar of wood which 
passed through the body of the mask ; this bar was held by the teeth of the wearer, 
the hands were not used at all to support the mask, nor did it even rest on the 
shoulders of the dancers. Various kinds of feathers were employed in the decoration 
of the masks. The masks were painted with red, black, white and a little blue (intro- 
duced) pigment In former days such a mask would be almost entirely constructed of 
turtle-shell, but these were made out of old wooden boxes and kerosene tins (Vol. v. 
pis. XVm. fig. 2, XIX. fig. 1. and Intemat. Arch. f. Ethno^. vi. 1893, pi. XIII.). 

The mask shewn in fig. 256 A belongs to this type, it is surmounted by a hammer- 

d by Google 


headed shark. Qizu drew ibr us a similar mask but Bunnounted by a dugoDg, it has 
three warka, feathers of the Torres Straits pigeon, on the head and tail, and on the 
back four kaikai. 

It is not clear what animal's head is reproduced on these ma^ks. At one time I 
thought it was always a crocodile's, but I believe that in some cases it may be a king- 
fish's or something of that sort, which has become conventionalised in the direction of a 
crocodile's head. 

5. One type of mask consists of a box into which the head of the wearer is inserted, 
surmounted by an animal. The bos, which is made of small boards obtained from 
Europeans, probably replaces a cylinder of turtle-shell shewn in fig. 255 and pi. XXXVIII. 
figs. 2, 3. 

A good example of this type is to be found in a mask made by Euduma of 
Muralug in Moa and taken by him to Kagir, where I bought it in 1888; it has been 
given to the British Museum (pis. XXXV. fig. 1, XXXVL fig. S). On the fiwnt of the 
wooden box there is a tin face, 245 mm. in height, surmounted by a wooden baiib, 
decorated with cassowary feathers, from the middle of the baiib a wire projects which 
supports a fish cut out of turtle-shell {zugawad, it was described as a fish of about 
30 cm. in length that jumps out of the water). On the back of the box is carved a 
kaimai (fig. 27), and on the sides are three eyes; all the designs are variously painted in 
red, black and white. Surmounting the box is a well executed model in turtle-shell of 
a gaigai, bonito, Thynnus ; it is 71 cm. (28 in.) in length. Dr Qunthw in his Study of 
Fi^es describes the genus Thynnus us having "a longitudinal keel on each side of the 
tail" (p. 458), a feature which is well shewn in the model. 

In the Museum of Victoria, Melbourne, there is a mask the box of which is 
O-shape in plan, there being two holes for the eyes in the converging sides; there 
is on one side of the box a human fiice 8un<ounded by typical fret-work. On the top is 
a cleverly made representation of a monkey stretching out its paws over the front of the 
mask, and holding in its arms the usual haiU}. Someone must have brought a monkey 
to Torres Straits which so struck the fancy of some native that he made a model of it. 

Composite Bfllgl«s. 

We may well suppose that simple &ce masks were firat made, gradually as the 
natives became skilled in working turtle-shell the more elaborate masks were constructed 
and eventually human and other effigies were attempted. These and the complex masks 
exhibit considerable dexterity and constructive skill. 

Jukes (I. p. 193) gives a figure of a human effigy, about 91 cm. (3 ft.) high, which 
was obtained at Erub, a fish depends from the chin and the object on the head may 
be a fiigate-bird. For an unprofitable discussion concerning this effigy the reader is 
referred to the Journ. and Proc Roy. Soc. JVjS. Wales, xliii. 1909, p. 50, xuv. 1910, 
pp. 81 — 83. In the Waiet zogo of Waier (vi. 277) that personage was represented by 
a turtle-shell image about 92 to 122 cm. in height. An image flx>m Mabuiag in the 
British Museum {Album, I. pL 329. No. 4) is crudely made, it is about 58 cm. high. 

In Vol. V. pp. 374, 375 I have described the effigies of the two great totems of 

d by Google 


Yam, the crocodile and hammer-headed shark, without doubt the latter was very similar 
to certain masks (p. 299). In the Truro Uuseum are two remarkable turtle-shell objects 
from Mer which were presented to that museum about 1840 by Lieut. O. B. Eempthonie ; 
one is a highly decorated crab 33 cm. long and 15 cm. broad, which must have been a 
wonderful object when perfect The other (pi. XXXIX. figs. 3, 4) is 76 cm. long and 
resembles the body of an insect, the head bears two enormous eyes, the thorax is 
barrel-shaped, the abdomen is an elongated cone. It is highly improbable that we shall 
ever know its significance, but I would venture the suggestion that it was connected 
with the lag zogo (vi. 218) which was concerned with the control of mosquitos. The 
shrine was situated behind the beach on which visitors would be most Ukely to land 
on account of the safe anchorage in its propinquity. 

There is a very remarkable turtle-shell mask-like object in the Berlin Uuseum, 
which is somewhat triangular in plan and in vertical section, the anterior end is curved 
from aide to side, the posterior end is pointed and terminates in a long spine. The 
dorsal keel and the latero-ventral edges are provided with a double series of serrated 
spines, serrated and simple spines project from the anterior surface, on the upper portion 
of which is affixed a four-rayed star also of turtle-shell. The form of this object is 
somewhat similar to, but less flattened than, that of a king-crab (Limulus) while the 
serrated spines recall the fringed labial processes of the carpet-shark (Croesorhinus 
tentaculatus) or those of a Lophius, but I am not aware that this fishing fi^g or angler 
occurs in these waters. 

Jukes (L p. 168) describes and figures a "curious ornament" which was obtained 
at Msaig. "It consisted of two rudely carved figures of fish, about two feet long, 
connected together by cross pieces, about one foot long [the ends of which were carved 
to represent bird's heads], over which fi^me was a large figure of a bird, with an 
immense toothed bill, the eyes and some other parts cut out of mother-of-pearl, neatly 
inlaid. It was altogether two feet high, and by no means badly designed or executed." 
Jukes correctly conjectured that the bird represented a hombilL This curious object 
may have been a portion of a mask of which the front part was wanting, or it may 
have been a ceremonial object of a kind which has not been described. It was either 
imported firom New Guinea or made by an Islander who had seen a hombill in New 

d by Google 


Qbeetinos and Saldtations. 

By S. H. Rat. 


Thbodohout the islands of Torres Straits the old form of greetiDg was to bead 
slightly the fingers of the right hand, hook them with those of the person greeted, 
and then draw them away so as to scratch the palm of the hand ; this was repeated 
several timea. This movement was called in the Western language get-pudai^, lit. hand- 
digging, get being the word for hand and pwLai referring to the scooping out or hooking 
motion. In the Eastern language the term used was tag-augwat, hapd-plucking* {tag- 
degwat, to pluck hands), tag meaning hand, and augwat being the root form of the verbs 
degwati to pluck, egwatumur to pull. 

In the Western tribe kissing combined with embracing the head sometimes took 
place. This was termed gyd-wiai, mouth-touching, gud-wtdi, mouth-approaching, or paru' 
nudai, lace-rubbing^ The modem term for kissing is gud-tapamai; the meaning of 
tapamai is uncertain. 

In the Eastern language eskos, lit. stick out, is used for kiss, but there is no evidence 
as to the custom being indigenoua 

The modem sentiment of kindness of feeling in a salutation is shewn by the names 
now in use. In the Western language mbu-anai {sihu-tvanai), pity, love, lit liver-putting 
or leaving, is used for greet, salute. In the Eastern langu^e the term is omare, liver- 
leaving or naa, to be sorry. 

' In a folk'tsle told b; Maino of Tutu, whan Kebra ODd Waier met Naga, it vas itatad that palai noin 
get pvdan, they two plnoked haada (with) him [v. p. 49). 

■ Tag-augwat was a promineDt element in the Miriam marriag« oeromoaiM (ti. pp. 113, 114). 

' Dr WUtod {Sarr. Voy. round the World, IBSfi, p. BIO) sayi that a friendly natiye "touched my dom 
with his." Dr Haddon wai iaformad in 1869 that among the Miriam after a long separation or when friend* 
retaroed after they had been given np for dead, they would embraoe heada and rnb oheeha together, once tor 
each cheek. He was informed that they neither mbbed noses nor kissed in former days. 

d by Google 



The words used as Balutatioos ore imperatives, aometimeB combined with a demon- 

In the western islands the visitor on arrival says Sangapat to which those visited 
reply Wa! On leaving, the visitor says Iowa! or lawakai! and the same is said in 
rettuii. Nipel meaning you two, or nita, you, is prefixed to indicate two or more persons. 
If the journey is to be short, the word magi, little, is prefixed to latva. 

A person passing another says Sauki! on approaching, and Siaupa! on leaving. 
Those passed say 8aiva! 

Sangapa, sauki, naupa and sowoa, appear to be derivatives from the demonstrative «i, there, 
that, with tbe particle wa indicating visibility, the directive endings pa, ki, and the nominal 
or local ending nga. Tbua eangapa for si-wa-nga-pa, to that place, latiki for ei-toa-ki, along 
there, tiaapa for ai-toa-pa, towards there, touxt, there, /aura is from the verb imoai, to journey, 
and kai is the sign of tbe present tense. Wa means yes. 

In the eastern islands when two or more persons meet, both sides say Maiem! 

A person passing or leaving says Nawa! to one person remaining, Dawam! to two, 
Dawadariivam ! to three, and Uridwa! to a company. The person or persons left reply 
Bakeam ! to one, Bakeamvlam ! to two, Bakeaumdare ! to three, and Bakeaware / to a 

Maiem appears to be formed from the pronoun ma, thou, with the dative suffix -em, but 
this doee not explain its use in address to several persona But em may be the suffix some- 
times added to personal names to indicate a man's oompanions (lit. pp. 56, 57), and hence 
maiem may mean your people. 

Nawa, damam, dawadariuiam and uridtoa are the imperatives of the verb to stay, stop. 
Bakeam, bakeamtdam, bakeautmdare, baketxwxtre, are the imperatives of the verb to go. 

In both languages of Torres Straits the Samoan Taiofa! introduced by the misaion 
teachers is often now used. This is the usual Samoan greeting, an abbreviation of tH, I, 
alofa, love. 

The waving of boughs of trees appears to have been the general method of demon- 
strating peace and friendship. 

There are no definite records of the forms of salutation in use in New Quinea in the 
immediate neighbourhood of Torres Straits. In E.iwu Tauitoo ! was used for Good-bye [ but 
this appears to be the Western Ishmders word lawal perhaps adopted from tbe Torres Straits 
teacher. At Toaripi in the east of the Papuan Glulf, the person saluted says A reha I or, 
Koti la reha I literally you (are) there, or came to you there. The person saluted says Ara 
mehal or Koti ta mehal I (am) here or came here. 

Tbe Fittapitta natives in North-west-centrsJ Queensland use expressions similar to the 
Miriam when parting company. The person leaving says NUngkdmd ! Stay ! sit down I (from 
verb n&ngka, to sit, stay). The person remaining says KUnddnipA I Go away I (from verb 
kOnda, to go'). When passing at a distance Hoo ! is shouted, and on nearer approach SouA I 
Come here!*. 

The Ealkadoon use Naioo I to attract some one's attention at a distance*. 

< W. E. Both, EHmological Studim amtrng the North-weet-eentrat Qtuentland Aborigine* (1697), p. 20. 

' ib. p. 8. 

d by Google 


Vabiods Social CusroMa 

By A. C. Haddon and J. Bruce. 

Bob, Serehadino fartub. 
ToTUAM, Fannino game. 
Taua, Ezchanoc or food. 
AiBWBB OR Mam, Exohanoe of food. 

A pleaaiog feature of the Miriam was the number of purely social gatherings vhich 
were held in addition to &mily gatherings aod the meeting of friends at marriages, death 
ceremonies, and the ntimerous magical and religious ceremonies. When one remembera 
that there was a population of some four hundred persons (388 in 1894) on an island only 
2-79 kilometres (just over IJ miles) long and 1-65 km. (less than one mile) wide, the 
&cilities for casual intercourse were great; but in addition provision was definitely made 
for friendly gatherings, thus proving the sociable cbwacter of the people. These frequent 
social gatherings may also be regarded as an evidence of the immunity of the Islanders 
from sudden attack by foreign foes. 

Rob, Ssrenadlng parUet. 

About the time of the beginning of the south-east season, the married and 
unmarried men and the unmarried girls of a certain village arrange a party to serenade 
other villages on particular nights. 

The serenaders wear a girdle of the yellow, sprouting leaves of the coco palm, 
kupi, bind a band of leaves round their heads, and insert a feather of the Torres 
Straits pigeon in their hair. They carry two short pieces of wood in their bands, which 
they use to beat time for their singing. The villages which they intend to serenade 
usually manage to hear beforehand of the proposed visit and so preparation is made. 

The villagers are sitting round their fire when the visitors arrive ; the latter march 
in a column, the two sexes being mixed indiscriminately, singing and beating time with 
their sticks. They circle round* the seated villagers and break into a kind of dancing 
movement to the time of the musia After continuing this for about an hour they 
cease and the residents give their visitors small presents of tobacco, bananas and so 
forth. The serenading party then goes on to the next village, it may take two or 
three nights to complete the round. 

The villages which have been visited arrange to return the compUment, and they 
serenade the other village and so get presents in their turn. 

In another form of rob the serenaders start at a later hour in order to give a 
surprise visit. They all sit down in a semicircle round the door of a house, whose 
inmates are supposed to be asleep. After the rob wed has been aung, the bead of the 
house comes out and presents his visitors with tobacco if he has any, but if he has 

> Qsnenll? thej maroh or danea round the flre with the left hand towards tbe centre, bnt od lonu OMuiODe 
Mr Braoe bM eeea them begio in this way and then reverse their direotioD without stopping the maaio. 

d by Google 


DODe he gives them what food he may hare in the house. Then they visit another 
village for a similar purpose. 

The following are the words of the usual rob wed, but their meaning is uuknown, 
nor have the singers the sUgbtest idea aa to what the song refers to. They say this 
rob toed was introduced to tbeir forebears at the veiy beginning of the settlement of 
the island, so how could they tell what the words mean [ 

Samo pamrana 

Parap paaarana a 

Kela a kezvb ia 

Tero ia atamke 

A neide kopas utirida 

A neide kda kda 

Bali bali kela a 

Bali bali kela a. 
Qiaz, one of the Sebeg te, composed the following rob wed : 

Kawair wair dorigile 

Wair wair dorigile 

Wair wair doriffile. 
Kawair wair is the name of a sea-fowl; these birds generally fly in pairs at night, 
and as they fly along the beach they cry out tbeir own name "Kawair wair I" Lu dorigile 
is the noise made by wind blowing through the branches of a tree or through a clamp 
of bamboos, and dorigile here refers to the soughing noise made by the birds aa they 
fly. This effusion of Qiaz' is considered rather good, and is quite a favourite rob wed. 

Bob ia coDsidered to be one of the very oldest forma of amusement. The natives 
say that it was introduced to Mer by the people of Dugong Island (Asub le) aod 
of Half-way Island (Zogared le), but they do not know whence these obtained it; aa 
these islanders and the Waraber le were the middle men who traded between the 
Uuralug le and the Miriam it is possible that they brought it from Prince of Wales 
Island, or some other western island. In any case the Kiriam are very fond of the 
custom and it helps to while away many evenings pleasantly ; but, as with all their 
forms of amusement, they do it to excess, not knowing when to leave off, the conse- 
qaence is that it generally leads to a rupture of good feeling* and is stopped. Recourse 
is then bad to some other diversion, which in its turn shares the same Eate. 

Totnam, Fanning pune. 

A rather nice custom, called Totuam, ia carried on by the Miriam during the 
mosquito season, that is during the months of the north-west monsoon, and especially 
in December. The Dauar le, for example, would go one night to Sebeg and &n the 
people of that village with leaves of the abal tree, Pandanus; if the following night 
waa fine they would go to Babud and so on until they had visited all the principal 
villages. Then those who had been visited would return the compliment, so that every 

> For HI exunple ot tbU of. qnamls amongat th«iiiMl*M, Vol. n. p. IfiO. 

d by Google 


one in tarn had the opportunity of enjoying the luzuiy of being &nned by another 
person. The people who are being fanned sit or lie down in groups in a circle whilst 
the &nners remain standing plying their lans amidst all kinds of chaffing and laughter. 
After the &nning is over the visitors get presents of food and tobacco from those whom 
they have been fanning. Men, women and children all take part in this custom, which 
is, however, only carried on in fine weather. Each village knows beforehand when it is 
to be visited, in order that presents may be held in readiness. It is a pleasant social 
custom, but unfortunately, as is the case with many of their customs, it generally ends 
up with a row, especially if it is carried on too long. Something is sure to happen to 
somebody in the community during the month or two of the ceremony, or the inhabitants 
of some village think that they did not receive so large a present as they themselves 
hod given; then the quarrelling begins. Some one has been seen to make the 20*70 
gesture*, or heard to repeat the eogo mer^, in order to try and injure some person, so 
the fanning does not always end quite so sociably as it began ; still there is never any 
serious quarrelling, it mostly ends in talk. 

When totvam is finished they hie to the lag togo le and beseech him to send the 
moequitos away. They are not very particular as to time, but when the mosquitos 
eventually depart the lag zogo le gets the credit of having banished them (vi. p. 218). 

Tama, Esohanga of fbod. 

A man who intends to make a tama gets together a great deal of food, then he 
goes into the bush, cute down a small tree, clears away the smaller branches, and strips 
off all the leaves. The denuded tree is erected in a deared spot, and the area encloeed 
with a fence. Bananas, coco-nuts, yams and other food are bung on the tree and when 
all is ready a large fire is made in the centre of the enclosure in the evening. 

The invited guests arrive, bringing food with them and dead coco-palm leaves for 
fuel, and they sit in a circle round the fire. The host, or master of the ceremonies, 
lu kem le (" the owner of the things "), goes round the circle, standing before each 
person in turn, who holds up the object he has brought saying, "Tama" and mentioning 
at the same time what it is he has brought. The "master" asks, "Where did you get 
this?" or "Who gave it to you?" and the visitor may make any answer that he can 
think of — the more humorous the better, and the present, which is usually food, is 
exchanged for some of the food on the tree. 

When some one makes his appearance for the first time, all the people present call 
out " Kerkar le tahakeamu ! " (" New man come "). He is invited to sit on a small mat 
near the fire, and he puts the present of food which he has brought on the mat in 
fiwnt of himself. The other men get up, take away his food and give him other food 
in exchange. The lu kem le stands in &ont of the new man, and pointing a fire-brand 
towards him says, " You see this fire-stick ! Tou go home and look after your wife. 
When you come again another night, you must bring firewood and food with you. If you 
have not brought kindling and food the next time I ask you, I will duck you in the 
a 1 had not heard o( them previoiuly to thii Meonnt 

d by Google 


sea. Now, you take your food and go outside." Excbaoge is made only of the same 
kind of food and to an equivalent amount. 

After this preliminary ceremony a sort of auction or mart takes place to which 
reference is made in Vol vi. p. 188. Two occasions on which we were at a tama, first 
when one of us was a kerkar le, and then when he brought his firewood and food, 
are described in Head-Huntera : Black, White, and Brown, pp. 58, 59, 62. 

Aliwer or Mam, Ezohanffe of fbod. 

Aiswer, or Ma/>n, are names given to the custom of a man giving a present of food 
to another, or of one village making a similar present to another village. Some time 
afterwards a return gift is made, as my informant expressed it, "like race," that is, in 
a spirit of friendly rivalry, the one party trying to outvie the other and thereby exhibiting 
at one and the same time its wealth of food and its generosity. This is not a case 
of buying or selling nor even of a definite exchange of food. I once saw this custom 
in operation in 1889. It was the return present, and the people from the distant vilU^ 
were sitting chatting, laughing and eating with their hosts; round about were bamboo 
poles on which bunches of bananas were hanging and other food was piled close t^. 
Mr Ray (llL p. 165) gives wetpur aa "a native exchange of presents" and regards it as 
equivalent to tama, "a ceremonial exchange of presents." I am not quite sure whether 
aiswer, mam, and wetpur may not be other names for the same custom as tama, which 
Mr Ray regards as "perhaps an introduced word." 

d by Google 


It IB not an easy matter to clasairy the games played by the adntte and children 
of various peoples, nor can one draw a hard and fast line between games played without 
apparatus, those played with objects which may be either actual tools or implements or 
toy imitations of them, or playing with toys. The following enumeration is not intended 
as a classification but as a simple method of grouping to bcilitate reference. 

Game, fiin, play is termed aagul (W.), segur (E.). 

1. Ctemea of movement that develop and exeroUe the bodl^ powen. 

The game that combines agility, endurance, skill and emulation to the greatest 
degree is that of dancing, but for the sake of convenieDce dances are dealt with in a 
special section (pp. 289 — 293). 

Among the games of agility may be mentioned skipping with a rope which is 
played by children in Mer, where it was stated to be an indigenous pastime. In Habuiag 
I heard of young people swinging on the aerial roots of the kdbi tree, which I believe is 
a kind of banyan (Ficus) (c£ Holmes, p. 287). 

The most energetic of these games is a kind of hockey, shinny or shinty, which 
is played everywhere. It is called kokan in Mabuiag, which is also the name of the 
ball itself. The ball is made of wood and varies from about 66 to 60 mm. in diameter 
and 3^ to 4 oz. in weight, the lai^;est being 78 mm. and 10 oz., it is struck with a 
roughly made bat or club, (awain, dabi (W.), which is usually a piece of bamboo, vaiying 
from 60 to 86 cm. in length, on which a grip is cut The game is played over a long 
stretch of the sand-beach, there are two sides and each player has a stick, but so &r 
as I could discover there were no goals and no rules. The game is very " &at " and 
causes intense excitement and a tremendous noise; it is not without an element of 
danger as the heavy ball is hit with extreme vigour. We witnessed in Mabuiag one 
great match between married and single men (c£ Holmes, p. 282). 

I In Bulletin No. i (IMil} ot his North Queauland Ethnograpkg Dr Walter Both giTM a oUuifloBtion and 
daMTiptioD ot the "Gamea, Sporta and AmiiBeiiieDta " of raiioDa tribes ot North Qneenalaud. Tbon tntetMtcd 
in thia mbjoet ahould eoDanlt (hia Talnable paper, and alao the following in the Jaunt. Boy. AiUh. Imt, xxx f iu . 
190S: "CbildreD's Qamea in Britiah New Oainea," b; Capt. F. R. Barton, pp. 269— S7S; " Introdnotoir Notaa 
OD the Taya and Oainea ot Elema, Papnan Onlf," by Bev. J. H. Holmea, pp. 380—388; "Notea on Children'* 
Oamaa in Britiah New Qainsa," b; A. 0. Haddon, pp. 380—397. 

d by Google 


2. Ghun«a of dexterity and iklll. 

In the Miriam ball game the players stAod in a circle and sing the following kai 
wed, ball song: 

Kai tapitari Ball hit. 

Kai tapitari Ball strike. 

Abu kak kai o! Fall not ball! 

Atimed kak kai o ! Throw not ball ! 
As soon as they begin to sing one player strikes up the ball with his hand towards 
another player, who in his turn hits up the ball, and so on, keeping time to the rhythm 
of the music. The song is repeated as long as the game lasts. Should anyone let 
the ball &U to the ground, he is jeered at. According to one account the game is 
properly played by two sides. 

Fio. !1E9. Frnit of th« kai tree lued m » ball, Fio. 260. P«lm-Ieaf ball introdnced b; 

68 mm. long, Mer. South B«a men. 

Formerly they used for this game the thick, oval, deep red fruit of the kai tree 
(fig. 259), which is quite light when dry ; this fruit, which has a tough rind, varies from 
about 6 — 7 cm. (2J — 2J in.) in length. At the present time they generally use a hollow 
cubical ball (fig, 260) made of Pandanus or coco-nut palm leaves. This ball was introduced 
by South Sea men and is a common Polynesian toy; the names kokan (W.) and kai 
(E.) prove that it is a loan object. It vanes in size from about 35 to 53 mm. in 
diameter, an average size being 45 mm. ; one oblong example measures 55 x 95 mm. (nL 
Roth, p. 17; Holmes, p. 280; and Barton, p. 279, and for a bladder game, p. 264). 

Mabuiag children play a catching game called vdai (wadat) or damadiai; the former 
is the red flat bean of a Mucuna, the latter is a hard fruit that comes from New 
Guinea. Boys and girls go in pairs into the sea, a boy tries to throw a bean to another 
boy which his partner attempts to intercept; should she succeed she in her turn throws 
it to another girl and her partner tries to forestall her. 

In the north-west season the Miriam men play a game for which I obtained only 
the name kolap. Two mats are laid at a distance of about 15 m. Two men sit behind 
each mat, those lacing each other obliquely being partners. Each man has four kolap beans 
which he tries to throw on to the further mat; a score of twenty finishes the game. 
I believe there is a similar game in which beans are thrown at a mark on the ground. 

Game of hide and seek : — Mr J. Bruce has given me particulars of a game in Mer 
called nem deraimer, " louse searching." It is played by men and women. Two sides 
are arranged, and they all kneel in a circle on the sand or round a bare spot. One 

H. Vol. IV. 40 

d by Google 



person, who is blindfolded, pute bis head down — or if not blindfolded the &ce ib placed 
so close to the ground that he cannot see what is being done. One of the other side 
then picks from his hair a louse, which is hidden tn the sand in the centre of the 
circle. All the players begin to sing and beat the ground with their hands, except the 
man from whose head the louse was taken, whose business it now is to search for it. 
If he succeeds in finding it, one from the other side has to put his head down and 
do the Bearching. Should the man fail to find the louse, one of the other side shows 
it to him and his side has to remais in until someone has been successful in the 
search. The one who finds the louse eats it. 

A variant of this game is called pone deraimer, or " eye searching," the crystalline 
lens of a fish being hidden in the sand instead of a louse. Roth (p. 17) found this 
game among several tribes of North Queensland. 

Both sexes in Mer play a game by the light of the full moon, in which food is 
bidden in the bush or gardens for others to find (cf. Holmes, p. 286). 

I do not know whether there is a hide and seek game in which players hide 
themselves, as in New Guinea (cC Barton, p. 267 ; Holmes, p. 285). 

Dr Uacferlane informed me that they played a guessing game in Her which 
consisted of giving two syllables of a name fi^m which the whole had to be guessed, 
thus: "ia?" "Elia." 

Children also play guessing the number of small objects held in the closed hand. 

Games with string exhibit great dexterity of the fingers and are distinctly games 
of skill For the sake of convenience these will be treated separately at the end of 
this section. 

3. Chune* of emulation. 

Various kinds of spinning tops are to be found in Torres Straits. The most general 
is a top (fig. 261 A) made of a Queensland bean (Entada scandens)* ; 
in Mer both bean and top have the same name kolap, but in Mabuiag 
the top is called vxma and the bean k&lapi or kolapi. The fiat, 
chestnut-coloured bean is perforated and into it inserted a thin stick, 
kolap pea (K) (usually the mid-rib of the coco-nut leaf), with an 
average length of about 14—15 cm., the beans averaging about 45 mm. 
in diameter. In a Mabuiag top the stick was called ful, probably 
because it was made of tulu wood; in Mer it is also called teter, or 
leg. The pewer kolap of Mer (fig. 261 b) is made from the fruit of 
the pewer which has a diameter of about 20 — 25 mm., the sticks 
being 10 — 12 cm. long. A similar top is made of the dried fruit of 
the eom tree (Thespesia populnea). 

These tops are spun with the fingers and resemble in this respect 
the tee-totum, but strictly speaking the latter is a four-sided or faceted 
top used in games of chance. These tops were simple toys for children 
and never attained the importance of the Miriam stone tops. Small 
cone-shells are spun by children by twirling them in the usual way 
between the thumb and finger. 


Fra. 361. Seed topi, 
Mer ; k«lap and 
pewer fcol^. 

■ Tbe pUnt u oklled tirip or lireb and the top is oonsaqneotly Bometimee oftUed ririp kotap. 

d by Google 


The use of Btone tops, kolap, is and always has been confined, so tar as our know- 
ledge goes, to the Miriam, for it is only in the Murray Islands and to a very limited 
extent in Erub that the fine-grained volcanic ash occurs of which they are made. The 
top is shaped like a split pea, the upper surface being occasionally slightly convex. 
Those in the collection vary from 11 to 19 cm. in diameter, 14 — 15 cm, being a common 
measurement ; they are generally about 45 mm. thick, and usually weigh about 2 lbs. 
The top shewn in pi. XXXVII. fig. 4 is 19 cm. in diameter, 5 cm. thick, and weighs 
4 lbs. ; for further dimensions see figs. 380 — 385. A top is seen in pi. XXI. fig. 3. 

The tops are carefully made and smoothed down, and although not absolute circles 
tbey are made as true as possible in order that they may balance when spinning. 
A hole is drilled through the centre, which is larger above than below, and into this 
a stick, kolap pea, is inserted. The shortest we have is 15 cm., the longest 41 cm., but 
the usual length is about 25 — 30 cm. The sticks are generally made of the heavy 
wooden heads of arrows (generally palm-wood) whittled down, but what was most preferred 
was the hard wood called dab, which was obtained from the Queensland natives through 
the intermediacy of the natives of Waraber and Half-way Island ; sometimes a piece of an 
old dugong harpoon would be employed as it was made of hard wood. The stick tapers 
delicately to it^ upper end and is smooth and neatly made. 

The upper surface may be plain, but it is usually decorated in various ways as 
described in the section on Decorative Art (figs. 378—383, pi. XXXVII. figs. 3, 4). 

Very great care is taken of these tope and the better ones are kept in round 
baskets specially made for them, which are often lined with calico as a further protection 
for the top. A basket made of coiled basketry is described on p. 82; a polygonal 
basket is seen in pi. XXVIII. fig. 1, but I believe this to be of foreign make. It is 
very amusing to see elderly men carrying with both hands a top ensconced in its basket 
with the greatest care ; the tops are also very carefully handled, and some are so much 
valued that we could not induce their owners to part with them. On the other hand, 
tops could be taken by certain relatives (vi. pp. 100, 101). The stick of his top was 
hung over the corpse of a deceased man (vi. p. 130), 

A top is spun by repeated slow, steady, sliding movements of the outstretched palms 
and fingers. Formerly they were spun on pieces of melon shell (Album, i. pL 344, 
No. 4), now pieces of broken crockery or the under surfaces of cups and saucers are 

Kolap toed, top songs, are sung during top spinning, kolap omen': one of these songs 
is given on pp. 268, 288. 

Stone top spinning was very prevalent during part of our stay in Mer, indeed at 
one time the people played so assiduously every week-day that they had no time to 
attend to their gardens, and on Saturday they did not bring in enough food to last 
till Monday. The Puritan Sunday is in full force, and none would dream of breaking 
it by gathering food, consequently numbers of children came to school on Monday morning 
without having had any breakfast. This made them peevish and inattentive, so Mr Bruce 
had to complain to the mamoose, and an edict was issued prohibiting kolap matches 

■ Tha spinning moTemene of a top is omen, or omen-omen when ipinning fait When a top is spinning 
BtcAdily, or as ve sa; "going to sleep," it is called koiap ku», and the Btopping of the lerolationa is called tiri. 


d by Google 


on Saturday, and the men were told to go to their gardeDs as heretofore. On one 
occasion there were thirty tops spinning at the same time (pi. XXVIII. fig. 1). The 
men sang songs and there was great cheering on of slackening tops, and shouting and 
jeering when one stopped. At the critical time when one was "dying" great care was 
taken to shelter it from the wind so as to prolong its " life " a few seconds longer. 
At one match we timed the four best tops, and found that they spun for 27|, 26{, 
26i and 24 minutes respectively. We have seen men of all ^^es (but no women or 
girls) engaged in these matches, the grizzled taking as much interest in the performance 
of their tope as the young men. In the larger competitions one section or side of the 
island is pitted against another. 

Dr C. H. Read {Joum. Anth. Inst, xvii, 1887, p. 85) was the first to deacribe these tops. 
He says (p. 88); "I do not think it very probable, though it is, of course, possible, that 
the natives of the Torres Straits islands invented spinning tops for themselves. It is far 
more likely that they received the idea from a more cultured and ingenious race ; for, apart 
from the rarity of the occurrence of this toy among savage tribes', it is evident that the 
notion of a spinning top, a very complex toy [!], would be little likely to spring ready made 
into the mind of a people of the mental calibre of the Papuan. We must, therefore, look 
elsewhere than among the races of New Quinea for the origin of the toy." Spinning tops, 
such as those described above, can scarcely be considered "very complex" toys, but at the 
time wheu Dr Read wrote his paper so little was known about the Torres Straits Islanders 
that he naturally underrated their mental calibres Two objects may very well have suggested 
the making of tops, namely the disc-shaped stone-headed club and the pump-drill (p. 128), 
but we do not know for certain that they possessed this instrument, though a pumpdrill with 
a circular disc as a fly-wheel occurs in the Gulf District. I see no reason however why the 
seed tops should not have been directly invented and the stone-headed club may have su^ested 
the stone top, though its shape is never like a split pea except in one specimen from Mer 
noted on p. 192. Tops spun with the hand occur in North Queensland (Roth, p. 18), and in 
the Papuan Gulf (Holmes, p. 281). 

Fio. 262. Peg-tops, Mer. A, 61 mm. long, weight a} oi. B, 90 nun. long, weight i\ at. 
C, SO mm. long, weight f ai. 

Two kinds of top spun by means of string were collected by us, the most common 
ng the peg-top (fig, 262), which is also called kolap. The form is variable, being 

1 The absence at Bpaoimens of partionlKr objeotB, eapeaUJly a trivUi object like a toy, in museums or the 
from the ooooonts of trsTsUers is nsgative eridenoe of eitremelf little value. 

d by Google 



either conical or biconical, the apex in the latter case is frequently truncate ; the point 
is knob-like. It is made of enoa (Mimusops) wood. Three other apecimens measure 
70 X 40 mm., 77 x 39 mm., and 84 x 46 mm. Those which we obtained in Mer were 
introduced from Mabuiag where they were said to be native, but this is I think 
improbable. The form of some of them is like that of European peg-tops, others 
resemble the common form of Malay top (gasing) which is found in the Malay Peninsula 
and throughout the East Indian Archipelago, but as the kolap is spun in the European 
way I think it must have been derived from European and not from Malayan sourcea 

The second kind of top spun by a string is without doubt of foreign design, though 
it appears to have been locally made. It consists of a disc of lead 4 cm. in diameter, 
in a hole in the centre is fastened a cane tube within which is inserted a 
wooden stick, 148 mm. long, with a swollen conical point (fig. 263). I obtained 
it at Mabuiag where it was called wana; I saw only this one specimen. 

I did not come across any whip-tops, but they occur in the Mekeo 
and Kabadi Districts of British New Guinea, where they certainly seem 
to be indigenous (c£ Head-Suntera : Black, White, and Brown, 1901, 
pp. 272, 273). 

The Islanders had a pastime of throwing sticks along the sand-beach 
when walking. This is alluded to in a folk-tale (v. p. 45), where Bia 
threw "a duhin, a simple toy spear made of the hard dukun wood, but 
he only played where there was a sand-beach, and not where there were 
plenty of stones." The toy consists of a thin rod of hard wood with a 
swollen fusiform end, something like a miniature dugong harpoon. In Mer 
we obtained two objects that were used for this mildly competitive game (fig. 264); 
the rods of our specimens have probably been broken, but if so the fracture was of 
ancient date. They are called omaiter or aipersi lu, sliding thing; the word omaiter is 
used also in another sense (figs. 367 — 370). Similar sticks, commonly known as wit-vnt, 
or " kangaroo rat," are used in Australia for the same purpose (cf. K. W. Thomas, Natives 
of Australia, 1906, p. 140, and Roth, p. 18). 

Fio. DM. Toy thiow-itiakE, omaittr, Mer. A, head end 11 om. '. 
B, head end U — 16 om. \oag, 39 mm. di&m. ; 

mm. diam. ; total length 1 
lotkl leogtb ai-a cm. 

The men sometimes had shooting matches with bows and arrows; in the western 
islands javelin hurling competitions were held, I have also heard of sham fights with 
blunt arrows. 

d by Google 


4. OameB of ImltaUon. 

Here as elsewhere children delight in imitating the occupations of their elders, and 
this mimicry forms a not unimportant part of their education. That this was so in 
the past is evident from the injunction of lads at initiation in Tutu to abstain in future 
&om playing with play canoes and toy spears. Not only are models of canoes still made 
for boys to play with (Album, I. pi. 346, No. 9), but I have seen in Mer AiUy rigged 
models of luggers and schooners with which the young men amused themselves, and 
the spirit of emulation was gratified by racing one against another (c£ Holmes, 
p. 283). 

Toy bows and arrows of eaeae grass stems were made by the boys. Dr Wilson 
{Voy. 1835, p. 311) noticed boys in Mer, "some of them very young, amusing themselves, 
shooting with bows and arrows, suited to their strength." 

On Yam I came across two small heaps of ruddled clam and other shells among 
some bushes where the boys played at "attgud" or "ktaod." The great totem shrines in 
the kwod of that island are described in Vol. v. pp. 373 — 378. The ceremonies connected 
with these now obsolete shrines were the moat sacred of their religious ceremonies. In 
this case the boys cannot be said to have exactly mimicked their elders, as they did 
not know what the real augud was like nor how the ceremonies had been conducted, 
but they " made believe " to their own satisfaction. Heaps of shells are a ■ 
feature in the kwod of the Western Islanders. 

5. Oame of divination. 

The only divinatory game known to me is that called koko in Mer, which is played 
by girls only. A number of girls bind their heads with bands of leaves or vines and 
flowers like a garland, or they make a rough sort of leaf basket which they put on 
their heads, or wear a fillet of a strip of palm leaf to which are fastened vertical or 
horizontal palm-leaf rings — in fact they adopt as fantastic an erection of leaves and 
flowers as they can devise. They then walk into the sea until only their heads and 
shoulders are above the water, form into a line each placing her hand on the shoulder of 
the one in front of her, and repeatedly sing "Koko, koko, kaiep xoagA," keeping time to 
the music by bobbing their heads up and down. Later they resort to the sand-beach 
and sitting down in a ring or semi-circle sing " Kegu-a bama-a gared-gep," at the same 
time they push their hands backwards and forwards palms downwards in the sand (this 
is called tag dittari, hand shoving), and mntteT " Lar-teregu tarasawem, kuxneru taraeamem, 
kegu teraaawem" (with fish teeth rub me, with bamboo knife rub me, with charcoal rub 
me). Finally they examine their hands to see whether they have been cut or whether 
the bit of charcoal held by each one has made one or two streaks on their palms 
(pi. XXVIIL figs. 3, 4). Should there be two marks they cry out, " Ah ! keg has killed 
a man," and begin again. Koko literally means, according to Mr Bruce, to cany on the 
back as a mother carries her child, it also is the name of a fine weather omen bird 

d by Google 


(vi. p. 260), kokokoko is a wood used to make fire-aticks, and kogkog or koko is marital 
intercourse ; kaiep, probably the same as kaip, a small shell used for scraping food ; 
vmg^, a Cyrena shell used for the same purpose; kegu or keg, charcoal (cf. VI. p. 146). 
Mr Bruce states that bamu, gared, gep are in this instance only sounds with no meaning ; 
it may be noted however that bam is turmeric and gared, south. This game is alluded 
to in the folk-tale "Markep and Sarkep" (vi. p. 54). 

I think I can state definitely that the following games did not occur in Torres 
Straits : chuck-stones or dice of any kind, gambling games, kite flying. 

6. Various Toys. 

A "pea-shooter" is raade in Mer out of a small bamboo, Abnis seeds ("crabs' eyes") 
being used as pellets. 

Fia. 26S. Toj bird mode ot board, Mst. Tail, wing, ring roond neok, and eje red ; 
leoglh 62 cm., breadth 2S cm. 

Fid. 366. To; rat, Mer; 82 om. long. 

Toys for making a noise are described under Sound-producing Instruments. 

I obtained in Mer a piece of water- worn board carved to represeat a bird 

d by Google 


(fig. 265). A hole was burnt io the board through which, so I was informed, a string 

waa threaded and knotted on the other side. The bird, eiwr, 

was put in the sea and the player holding one end of the 

string ran along the beach. There is a similar specimen with 

a small central hole, also from Mer, in the Vienna Museum 


A root which had been slightly touched up so as to make 
it resemble a rat (fig. 266) was a plaything with some children 
on Mer. It is quite possible that simple carvings were some- 
times made to amuse children. In the British Museum there 
is a model of a crocodile's head carved in wood, painted red, 
white and blue, and &stened to a piece of spiral wire, which 
is evidently a toy. I also obtained in Mer a piece of wood 
carved to represent a human face with a very protuberant nose 
(fig, 267), it was merely a toy. 

Miriam children play in the sea with the spathes of the 
leaves of coco-nut palms (fig. 208), pretending they are small 
canoea Toy canoes were frequently made for children throughout 
the islands ; I collected one at Mabuiag in 1888 (A&um, i. pi. S46, 
No. 9) which is 615 mm. long and is decorated with simple 
patterns by charring the wood. 

Via. 367. UrotMqne head u 
a toj, 26'S em. long, Her. 

7. String Flgurei and Trick*. 

String figures, wome (W.), kamut (E.), allied to our cat's-cradle, are universally 
played by the children and sometimes by adults, but it seems to be dying oat'. 
Usually one person plays it alone, in some cases using the toes as well . as the fingers, 
and often bringing the mouth into requisition. The patterns are very varied, and many 
are extremely complicated in manipulation although the final result may be simple. 
They are all intended to be realistic; in some cases the object represented is obvious, 
in others the imagination must be called into play, but other natives invariably recognise 
them and difierent islanders make the same figures. There are a large number of 
undescribed figures in addition to those described below, among which may be men- 
tioned: one child; two children; a woman micturating; coition; a dog; crow korkor 

1 I first drew attention to this pastime in 1890 (J.A.I, m. p. 861). In 1B98 Dr Bivers and I deTiaed a 
B;«tein ol QOmenolBtnre for recording the movementa which me pabtiahed in Man [n. 1903, No. 109, p. 147); 
thii paper gave a BtimaluB to the enbject. Sevenl iaveBtigatorB have adopted the neeleis plan of pnbliBhing 
drawings ol the completed figures withont any indication of how the; are formed. For farther information 
the reader ia referred to String Figures b; Mra Jajne (1906) and to Cat'i CradUt from nuuy Landt by Kathleen 
Haddon (1911), both of whioh deal with the diatribntioD of the varioaa flgnrea and tricka. I am indebted to 
the oonrteaj o( Ueaara Longmaiu Oreen and Co. for the loan of figa. 368—376, 278— 2S4, 294 from the la*t- 
mentioned book. Tlie string Ggorea and trioka were ooHeoted bj Dra Bivers, MaoDongall, Ur Baj and myselt. 
I have to thank m; daughter for the final form ol man; of them and Mr W. Innea Pooook tor re-writing 
Nob. 19, 24, and 39 and for other aaaiatance. 

d by Google 


(W.); the pearhu fish; a Bmall fish, termoi (W.), which accompanies sharks; a crayfish, 
haiar (W.); the larva of the aDt-lion, gobai (W.); a mouth, gud (W.); liana or other 
climber, vgal ngal (W.). The names of the various islaads are given where we obtained 
the figures, but doubtless they occur eveiywhere in the Straits. 

Various movements appropriate to the object represented are also made, thus 
swinging movements are given to the limbs of the crayfish, other moving figures are 
mentioaed below. In some cases the figures are accompanied with songs, kamvt wed (K), 
which are sung in a low tone. For the translation of these I am indebted, as usufd, 
to Mr Bay, but the words are frequently obscure and cannot be translated with certainty. 

The term "string figure" is employed in those cases in which it is intended to 
rejiresent certain objects or operations. The cat's-oradle of our childhood belongs to 
this category. "Tricks" are generally knots or complicated arrangemeats of the string 
which run out freely when pulled. Sometimes it is difficult to decide which name 
should be applied. 

A piece of smooth, pliable string should be selected which is not liable to kink. 
A length of about 2 m. (6 ft. 6 in.) is usually the most convenient; the ends should 
be tied in a reef knot and then trimmed, or sewn together with cotton, or, best of all, 

A string passed over a digit is termed a loop. A loop consists of two strings. 
Anatomically, anything on the thumb aspect of the hand is termed "radial," and 
anything on the little-finger side is called "ulnar," thus every loop is composed of a 
radial string and an ulnar string. By employing ' the terms thumb, index, middle-finger, 
ring-finger, little-finger, and right and left, it is possible to designate any one of the 
twenty strings that may extend between the two hands. 

A string lying across the front of the hand is a palmar string, and one lying 
across the back of the hand is a dorsal string. 

Sometimes there are two loops od a digit, one of which is nearer the finger-tip 
than the other. Anatomically, that which is nearer to the point of attachment is 
"proximal," that which is nearer the free end is "distal" Thus of two loops on a 
digit the one which is nearer the hand is the proximal loop, that which is nearer 
the tip of the digit is the distal loop; similarly we can speak of a proximal string and 
a distal string. 

In all cases various parts of the string figures are transferred from one digit or 
set of digits to another or others. This is done by inserting a digit (or digits) into 
certain loops of the figure and then restoring the digit {or digits) to the original 
position so that they bring with it (or them) one string or both strings of the loop. 
This operation will be described as follows : " Pass the digit into such and such a loop, 
take up such and such a string, and return." In rare cases a string is taken up 
between thumb and index. A digit may be inserted into a ■ loop from the proximal 
or distal side, and in passing to a given loop the digit may pass to the distal or 
proximal side of other loops. We use these expressions as a general rule instead of 
"over" and "under," "above" and "below," because the applicability of the latter terms 
depends on the way in which the figures are held. If the figures are held horizontally, 
"over" and "above" will correspond as a general rule to the distal side, while "under" 

H. VoL IV. 4X 

d by Google 


and " below " will correspond to the proximal side In some cases, when there is no 
possibility of confusion, we have used the shorter terminology. 

A given string may be taken np by a digit so that it lies on the front or palmar 
aspect of the finger, or so that it lies on the back or dorsal aspect. In nearly all 
cases it will be found that when a string is taken up by inserting the digit from the 
distal side into a loop, the string will have been taken up by the palmar aspect, and 
that the insertion from the proximal side into the loop involves taking up the string 
by the dorsal aspect of the digit. 

Other operations are those of transferring strings fivm one digit to another and 
dropping or releasing the strings from a given digit or digits. 

The manipulation consists of a series of movements, after each of which the figure 
should be extended by drawing the hands apart and separating the digits. In some 
coses, when this would interfere with the formation of the figure, a special instruction 
will be given that the figure is not to be extended. Usually it is advisable to keep 
the loops as near the tips of the digits as possible. 

There are certain opening positions and movements which are commtm to many 
figures. To save trouble these may receive conventional names; the use of these will 
soon be apparent, but it is better to repeat descriptions than to run any risk of obscurity. 

Fia. 368. PMition I. 

Position I. — This name may be applied to the position in which the string is placed 
on the hands when beginning the great majority of the figurea 

Place the string over the thumbs and little-fingers of both hands so that on each 
hand the string passes from the ulnar side of the hand round the back of the little- 
finger, then between the little- and ring-fingers and across the palm; then between the 
index and thumb and round the back of the thumb to the radial side of the hand. 
When the hands are drawn apart the result is a single radial thumb string and a single 
ulnar little-finger string on each hand with a string lying across the palm. 

This position difToTB fi-om the opening position of the EInglish cat's-cradle in which 
the string is wound round the hand so that one string lies across the palm and two 
across the back of the itand with a single radial index string and a single ulnar little- 
finger string. 

Opening A. — This name may be applied to the manipulation which forms the most 
frequent starting point of the various figures. Place string on hands in Position I. 
With the back of the index of the right hand take up from proximal side (or from 
below) the left palmar string and return. There will now be a loop on the right 

d by Google 


index, formed by strings passing from the radial side of the little-finger and the ulnar 
side of the thumb of the left hand, i.e. the radial little-finger strings and the ulnar 
thamb strings respectively. 

Fia. 360. Openiag A. 

With the back of the index of left hand take up fix)m proximal side (or from 
below) the right palmar string and return, keeping the index with the right index loop 
all the time so that the strings now joining the loop on the left index lie within the 
right index loop. 

The figure now consists of six loops on the thumb, index, and little-finger of the 
two hands. The radial little-finger string of each hand crosses in the centre of the 
figure to form the ulnar index strings of the other hand, and similarly the ulnar thumb 
string of one hand crosses and becomes the radial index string of the other hand. 

The places where the strings cross in the centre of the figure may be termed the 
crosses of Opening A- 

In some finished figures if the strings are pulled apart carelessly a hopeless tangle 
is the result. To avoid this take the top and bottom straight strings of the figure 
and pull them apart, and the string will usually resolve itself into a simple loop. 

String figurei. 

1. Baur, a fish-spear with several prongs, Mer. 

Position I. 

Take up with the right index the transverse string on the left palm from its 
proximal side, give it one twist and returiL Pass the left index through the right index 
loop fiom the distal side, and take up the transverse string of the right hand frY)m 
the proximal side and return through the loop. 

Drop the thumb and little-finger loops of the right hand and draw the hands apart. 

Fio. 270. Fiah^pear. 

d by Google 



2. Dab, spear*, Men 

Opening A. (Left palmar etring must be taken up first) 

Transfer right index loop to left index, and the original left index lo<^ to the right 
index, passing it over the one just transferred. 

Release right index and the spear Hies to the left; by bringing the right thumb 
and little-finger close together the handle of the spear appears. 

Pick up on right index the string just dropped, uid release left index; the spear 
then flies to the right. 

3. U. Her, urab, Maboiag. The coco-nut palm. 

Pass fingers &om the distal side into thumb loops aad close hands. 

Fio. 372. CkMo- 

Fio. 973. Tem, 

> The atme given for tluB figare wu dab, which aeema to be a in[Bnomer, u the dab U & limple ipeur 
(p. lOfi), vhereu the Qgore reprewnta & pioDgsd Bpesr. "Throwing the flib.Hpeftr" would be a more &ppropil*te 
Dune. The Mine figure ooonn at PrineeBH Charlotte Ba; and middle Palmer River, N. Queensland, where It 
id oaUed " Dnok in flight." Walter E. Both, N. Queaula»d Eth. BM. So. i, 1902, pi. V. fig. 6. 

d by Google 



Put toe' from the distal side into thumb loops, drawing radial thumb string over 
all other atrings, and holding it down. 

Exchange loope on little-fingeis, the right passing over the left. 

Repeat with indices. 

Draw tight and work the strings up to form the crown at the head of the tree. 

4, Sirar, the tern (Sterna Bergii), Men 

Opening A. 

Hold ulnar side of little>finger loop with toe. 

With little-fingers take up ulnar strings of index loops from the proximal side, 
returning jnviximal to the ulnar strings of the little-Snger loop. 

Hold radial thumb string with the mouth. 

With thumbs take up from the proximal side the radial strings of the index loops 
and return proximal to the radial strings of the thumb loop. 

Release indices and mouth. 

Move the hands inward and outward, and the strings will imitate the movemeate 
of the tern's wings. 

Sing: Sirara lalxUvba airara liibalvba neidgt kari-gedge doali dofjoaiU. 
Tern feathers on rock on my land. 

6. Le sik, the bed, Mer. 

Opening A. 

Put thumbs proximal to index loops and into little-finger loope frttm the proximal 
side; take up on the backs of thumbs the radial strings and return under index loops. 

Pass little-fingers through the index loops from the distal side and into the thumb 
loops from the proximal side ; with backs of little-fingers pick up ulnar thumb string 
and return through index loops. 

Release indices. 

Sing: le sHoge, le sikge, ut-eidi, ut'Oidi, sik erapei. 

man on a bed, man on a bed, asleep lies, asleep lies, bed breaks. 

At the word " erapei " release little-fingers and the figure disappears. 

Fia. 274. Bed. 

■ Tbe Dfttive method of inaiiipniation i» given in eMh OMe, bnt altbongh & foot maj freqaently be Doed, t( 
is often more Mtie&otoi? to get the help ot another person, or book the etrisg on soma object. 

d by Google 


6. Tup, a amall fiah (p. 166), Mer. 

Hold part of the striDg between the thumbs and indices, the hands being about 
eix inches apart; make a small loop by bringing the right hand towards you and to 
the left. Hold the loop between the thumbs and indices so that both the loops hang 
down, and pass both indices towards you through both loops. Draw the hands apart 
and turn indices up. 

There should now be two loops on each index, with the two radial strings r unni ng 
straight across, while the two ulnar strings cross. 

Pass thumbs into the proximal index loop from the distal side, and with backs of 
thumbs pick up the proximal ulnar index string. 

Pass thumbs into the distal index loop from the distal side, and with backs of 
thumbs pick up the distal ulnar index string. 

Pass little-fingers distal to the distal radial index string and proximal to the 
prosimal radial index string; with backs of little-fingers take up this string and return. 

Each httle-finger is now in a triangle. Pass the indices irom the distal side into 
this triangle, and by turning them up towards you, pick np on their tips the slanting 
string, i.e. the distal radial index string. 

Release thumbs and extend, by turning the palms away from you. 

Sing: Tup igoli umi Waierge. Water keage, Waxerge Waier 

Tup swim round to Waier, Waier in the channel, to Waier Waier 

in the channel. 

Fio. 375. Tup. 

7. Oeigi, king fish, Mer (fig. 314; cf. vr. p. 18); Dangal, dugong, Mabuiag. 

Opening A. 

Release right index and draw out; bend left index into its own loop, thus holding 
down to palm the string running from left thumb to little-finger. 

Release left thumb and little-finger and draw tight. 

Put string over left band aa in Position I. 

Pass left index over the transverse string of the right hand, and return, twisting 
the index towards you and up. 

Pass right index into right thumb loop from the distal aide, and turning the finger 
up away from yon, pick up the ulnar thumb string. 

Pass right index into right little-finger loop from above, and by bending it towaids 
you and up, pick up the radial little-finger string, allowing the string just picked up 
from the thumb to slip ofi*. 

Pass right little-finger towards you into the triangle just formed, and hook down 

d by Google 



against the palm the ulnar thumb etring, allowing the original little-finger loop to 
slip off. 

Similarly, with the left little-finger hook down the left ulnar index string. Release 
thumbs and extend. 

Another person puta a hand into the central diamond. If the manipulator leaves 
go with the left band and pulls with the right, the fish will be caught; but if he 
leaves go with the right hand and pulls with the left, the fish will escape. 

Sing: Qaigi usar perkori karem-lar ko-dittdare. 
deep-sea fish. 

Each word is repeated twice, the last word was said to meao, He poke yonr fin 
(VI. p. 16). 

Fia. 276. King fish. 

8. Laiplaip neur, girl with large ears, Mer. Make geigi. 

The figure consists of three lozenges, the outer strings of each lateral lozenge may 
be described as the upper and lower lozenge strings respectively. 

Pass thumbs of each hand above the upper lozenge strings and, drawing these 
strings backwards, pass thumbs below the lower lozenge strings and take these up and 

In middle of each half of figure there is now an axial string passing to point of 
junction of thumb, index, and little-finger loops. 

With hack of little-fingers from above take up the asial strings and drop indices. 

A new figure is produced also consisting of three lozenges. 

Pass little-fingers on distal side of lower lozenge strings, take up these strings from 
above and return. 

Sing: laipknp near tarabuli urpi U a kolam 

big ear girt both come down fire-ash person and through sexual intercourse 

both come down. 

Fta. 277. Oirl with big ear*. 

9. Nor, canoe with two masts, Mer. 

Opening A. 

Another person must pass his hand distal to the ulnar string, and proximal to the 

d by Google 


ulnar pair of crossed atringa, and take up from above the radial pair of crossed strings 
at their point of junction, and draw them well out. (The natives perform this action 
with their right big toe ; the loop ia therefore called the " toe loop.") 

Bend down the right middle-finger through the loop on the right index, and take 
ap the uhiar thumb string on its dorsal surface and return. 

Repeat with left middle-finger. 

Release thumb, index, and little-finger of each hand. 

Draw out laige the loop remaining on the middle-fingers and with this go through 
Opening A. 

Pass middle-fingers distal to the little-finger loops and into the toe loops from 
the proximal side. Then pass them distal to all the transverse atringa except the radial 
thumb string; take up this string on their doraal aspect, releasing thumbs, and return 
through toe loopa. 

Release toe loops and indices and draw tight. 

Fro. 378. Canoe with two mtala. 

10. Pagi, Mer, ger, Mabuiag, a sea-snake. 

Opening A. 

Pass the right hand round the left; hand so that all the strings cross the back 
of the left hand from the ulnar to the radial side. 

Pass the left hand and its strings from the distal side into the right index loop 
and bring it out proximal to the ulnar index string. Release right index. 

Unwind the left hand, bringii^ the right hand back to ita usual position. Release 
left index. 

There is now a single transverse string on the right palm, and a single transverse 
string on the back of the left hand. 

With left index take up from the proximal side the transverse string on the' 
right palm. 

Transfer the string from the back of the left hand to its palm and draw tight. 

Release left thumb, transfer the left index loop to the left thumb. 

Put each index into its little-finger loop from the diatal aide and take up the 
ulnar string with the back of the index. 

Hold the hands pointing away fixim the body with the index fingers uppermost. 
Withdraw left thumb, and with it gently press down the radial little>finger string 
until the "snake" appears. Gently draw out the right hand and the snake will swim. 

d by Google 


It 13 interesting to note that instead of the pointed tail characteriBtic of laud- 
snakes, Pagi faaa the broad flat tail peculiar to sea-anakes. 

Sing: Pagia mat nagedim upi etauerida kai amarem' pekem. 

Sea-snake you to where tail strikes I to side. 

Fio. 279. 8e»-»n»ke. 

11. Ti meta, the nest of the ti (the sun bird, Nectarinia australis, vi. p. 8), Mer. 
Old, canoe, Mabuiag. 

Opening A. 

Insert each index into the little-finger loop bora the distal side; bend it towards 
you and pass it to the proximal side of the radial little-finger string, and bring it 
back to its original position by passing it between the nlnar thumb string and the 
' radial index string. Release little-fingers. 

There are now two loops on each index and a large loop passing round both 
thumbs. Insert the little-fingers firom the distal side into the index loops and pull 
down the two ulnar index strings. (End of Ti meta opening.) 

Let go both thumbs gently and insert them into the same loop in the opposite 
direction to which they had been previously (i.e. change the direction of the thumbs 
in their loops). 

With the dorsal aspect of the thumbs take up from the palmar side the strings 
passing obliquely from the radial side of the indices to the ulnar little-finger strings, 
and extend. The inverted pyramid in the centre represents the nest. 

Sing: Ti ti ti mari keaa diteredi kari kesa diteredi. 

This is very obscure, it may mean TV thee property choose, me property choose. 

Fio. 380. Neit of ths aon bird. 

12. Nageg, the trigger-fish or leather-jacket (Monacanthus, pi. XXXVIII. fig. 4; see 
VI. p. 19), Mer. 

Ti meta opening. 

Drop right thumb loop without pulling tight, and pass right thumb into the upper 

I 7 CRormoJi, roUa kboDt. 

d by Google 


central triangle, and press the two strings of the loop just dropped by the thumb 
towards the right. 

Take up with the thumb, from the proximal aide, the oblique radial index string 
and return, letting the two original loops slide off the thumb. 

Take right thumb out of its loop and insert again in the opposite direction. 

With dorsal aspect of thumb take up the two ulnar index strings and bring them 
through the thumb loop. 

Take out the right little-finger from its loop and place it in the right thumb 
loop from the proximal side, withdrawing thumb. 

Take up with the right thumb from below, and close to the index, the radial index 
string that passes across to the radial side of the left index. Withdraw index from 
both loops. (End of Nageg opening.) 

The loop released by the index will form part of the head of the fish, and the 
short loop above it is the dorsal apine. 

Drop left thumb string without drawing tight. 

A big loop is now left which will form the tail of the fish. 

Press down with the left thumb, from above, the oblique string from the radial 
side of the left index till it is below the two straight strings connecting the figure. 

Release thumb, and pass it above the straight strings and take up from the far 
side of the two strings, and fiT)m below, the string just pressed down, and extend, 
keeping the left thumb string in the middle line of the figure. 

This string represents the row of spines on the fish's tail 

Sing: Nageg upi aeker dike, abele lar upige aeker dike. 

Nageg tail comb' it is here that fish on the tail comb it is there. 

Another version is Nagegera {Nageg's) erakai upige (on tail) aeker (comb). 

Fio. 381. Trigger-fisb. 

13. Saper, the flying-fox (Pteropus), Mer. 

Repeat the previous figure Nageg to the end of the Nageg opening, only using 
both bands all through ; the figure is then symmetrical. 

Extend by passing each index into its thumb loop firom the distal side, and picking 
up on its tip the radial string. Release thumbs. 

Say: a au, which is the noise made by the flying-fox. 

1 This has referanoe to the Beriei of biiibII apiues at the bwe of the tail ot the nageg fish; in the folk-tale 
Sageg is tbe mother ot Otigi. 

d by Google 



14. Ltm baraiffida, the aetting sao, Mer; Dbgai, a eUr, Mabuiag. 
Opening A, 

Pass little-fiDgeis distal to index loop and insert them into the thamb loops from 
the distal side. With backs of little-fingers take up the radial thumb string and return. 
i thumbs. 

Fio. 282. Fl;ing-foi 

Paas thumbs proximal to the index loops and into the little-finger loops from the 
proximal side. With backs of thumbs take up the two radial little-finger strings and 
re tuna. Release little-fingers. By this movement the little-finger loops have been 
transferred to the thumbe. 

Pass little-fingers distal to the index loops and into the thumb loops fix>m the 
proximal side. With backs of little-fingers take up the two ulnar thumb strings and 
return. (End of Lem opening.) 

Transfer loop of left index to right index and loop of right index to left index, 
passing it over the loop just transferred. 

Pass middle-fingers from the distal side through the index loop and take up from 
the proximal side the two ulnar thumb strings and return through index loops. 

Release thumbs and indices. 

Pass the thumbs fiY>m the proximal side into the middle-finger loops and withdraw 
middle-fingers, thus transferring the middle-finger loops to the thumbs. 

Fra. 338. 8«ttii)g ann. 

Extend the figure with the thumbs towards you; there will then be a St Andrew's 
cross in the centre of the figure. Insert the indices from the distal side into the 
Uteral spaces of the cross, and into the inverted triangle (the one farthest bom you) 


d by Google 


from the proximal side. With backs of indices take up the respective arms of the 
«roas and return. 

Pass middle-fingers through the index Loops irom the distal side and take up &om 
the proximal side the two ulnar thumb strings and return through the index loops. 

Release thumbs and indices, and with the thumbs manipulate the figure so as to 
make an approximate semicirole with four diverging loops (rays). 

Drop middle-fingers and draw out gently and the sun will set. 

Sing: Lem a lem a gair lager-lager tag a lager-la^er kokeaa kokeaa. 
Sun and many ropes hand and ropes. 

Lager, rope, lager-Utger may mean stringy (rays of the setting sun); the last three 
words may be lagelag, wishing, kogiz kogiz, much sexual intercourse. 

15. Ares, fight (two men fighting), Mer. 

Lem opening. 

There is now a triangle in the centre of the figure; into this insert the indices 
from the proximal side, and with the back of each index take np its respective side 
(the radial thumb strings). 

Pass the proximal index loop of both hands over the two distal loops on to its 
palmar aspect (in other words, Navaho the proximal index Loop, c£ Ca^a Cradles from 
many Lands, p. 5). 

Release thumbs, twist the index loops three times to make the "men," and release indices. 

Insert the four fingers into the little-finger loops and draw slowly apart. After 
the two men meet in the centre only the left string should be pulled, until this becomes 
free; the remaining man may then be pulled to the right. 

This figure represents a Murray Island man and a Dauar man who meet and begin 
to fight, and they "fight, fight, fight" (which the performer repeats) until the Murray 
Island man kills the Dauar man (when the left loop M\a), and being a head-hunter, 
he cuts off his enemy's bead and runs home with it (the hindermost loop (fig. 284 B) 
representing the head). 

Fia. 384. Tvo men fighting, Hei. 

d by Google 


16. Lar gvle, cuttle-fieh, Mer. 

Opening A. 

Pass fingers from above into the thumb loops and close hands. 

Put toe fi«m above into the thumb loops drawing the radial thumb string over the 
other strings. 

Pass thumbs under the index and little-finger strings and drav back the uln&r little- 
finger strings on the backs of the thumbs, returning through the thumb loops. 

Release the little-fingers. 

Pia. ass. CDttlB fiata, 

17. Epei, a basket, or Kanaw, a Shell (? kanai, the mitre shell, Mitra), Mer. 

Opening A. 

Pass thumbs proximal to the index and little-finger loops and take up the ulnar 
strings of the little-finger loops and return. 

Take up the straight string between the thumbs in the teeth, and lift it over 
the tips of both thumbs. 

Release little-fingers. 

Pass httle-fingers proximal to the index loops into the thumb loops, bend them 
towards you over the radial thumb strings and return, thus picking up these strings. 

Release string held by the teeth. 

Release indices and the figure disappears. 

Sing: Ktmaurede kaiuturede epei tttepeti kerisor topaidili (? lapaiteredili). 
basket shell spill again. 

Fio. £M, Baakst. 

d by Google 


18. Omadcer, children, and Gazir and Kiam, Mer. 

Opening A. 

Pass thumbs proximal to the index and Uttle-fioger loops, hook bade botii little-finger 
strings on their dorsal aspect and return. 

Pass each little finger over its index loop, take up the ulnar thumb string from 
its proximal dde and return. 

Release all thumb strings. 

Pass thumbs over index loops, and insert proximally into proximal little-finger loops, 
hook up ulnar strings of distal little-finger loops and return through the ^oximal loop. 

Release little-finger loops without extending the figure, (uid passing little-fingers over 
the index loops insert them in the thumb loops in the same direction as the thumbs. 

Stretch thumbs and little-fingers widely apart and alternately raise and lower each 
hand. The row of " children " will then dance. 

Sing: Omaaker aegur batuglei segur ki ban wdba dada wahawaba waipeda 
Children play two go round play we you middle yourselves go round 7 

utimdeda* tva mo wa ma sigesima sigazema* via mo ma sigeeima. 
throw away ? yes. 

This was explained as, Children play and sing and run away and hide. 

Fia. 387. A, ChiIdr«D. B, Ouii and Eiam. 

Drop the index loops, one a little before the other, and draw the hands apart 
Two rings will be left on the strings, one of which is larger than the other. The 
larger one is Qazir, the smaller, Kiam, these are the sacred grounds in Mer at which 
the Bomai-Malu masks were exhibited (vi. p. 284); the ceremonies performed at Gazir 
and Kiam were of a similar nature, but the former were the more important. 

' FeTlwpg itimdeda, shoot. * Peihaps tiifa,ii (W.) from afar, or nian'ma (W.) ooma up. 

d by Google 


Becite: Kiam kebi gaire Oazir au gaire, 
Kiam small lot Gazir large lot. 

19. Tamer atkanur le. Tamer snatcbii^ man, Mer, 

Extend loop with both hands and place the middle of both strings ronnd the big toe. 

Keeping the strings parallel, pass the right loop of the string through the left, and 
the one now on the right through the other. 

Let go strings and pick up through the loops the portion of the loop that was 
left OD the toe and play. 

Sing: Sua! Bua! Bua! (or Bub! Bub! Bub!) He! He! He! 

The figure was described as " two legs and no body." It ie intended to represent 
the passing of one of the sacred Malu clubs (Camera) (Ti. p. 296) from one Beizam 
boat to another during Bomai-Malu ceremony at Las (vi. p. 310). 

Fig. 288. Htut pauing the Main olab tamtr. 

Fio. 289. Oongb. 
20. Kobek, cough, Mer. 

Stret-ch the string and place it behind the neck, bringing the ends in front. 
Opening A. 

Pass middle-fingers through the index loops from above, take up the palmar strings 
and return through the index loops. 

Release all except middle-fingers, and make a coughing noise. 
Sing: Kobek dawiSna* kapumita* dawStia kobek idid lu aabsab lu. 
Cough greasy thing sour thing. 

' T ada-mttaa from adaka teay (W.), Mod away. 

■ 8aid to be • Mog-word with no maaning, but pertwpa (W.) iopu, good, injla, taate. 

d by Google 



Fio. 990. Ibegota. 

21, Kvper, maggots, Mer. 
Opening A. 

Release right index and draw tight. 

Pass middle-finger of left hand under the index loop, hook fix»n above the radial 
thumb string and return. 

Release thumb and little-finger of left band. 

Insert left thumb proximally into the index loop of the left hand. 

Pass left index distally into the right little-finger loop, hook up from below the radial 
little-finger string and return. 

Pass left thumb distally into the right thnmb loop, hook up 
firom below the ulnar thumb string and return. 

Pass proximal loop on left index and thumb over the tipe 
of both digits. 

Apply the point of the left thumb to that of the index 
and transfer the index loop to the thumb. 

Insert right thumb in its loop in the opposite direction. 

Take right thumb and little-finger loops on all the fingers 
of that hand. 

Replace left middle-finger by left little-finger in the opposite direction. 

Take out left thumb fiom its loops, pull with right hand, and the maggot goes 
along to the right 

Slue round right hand pulling smartly and the mf^got disappears. 

Sing: Kupera zans' zariz upi eupamada. 
Maggots go along jump up. 

22. Kai, ball (playing ball), Mer. 
Opening A. 

Release thumb loops. 

Twist index of right hand. 

Twist little-finger of right band. 

Transfer little-finger loop of right hand to index, and take up both loops in the 
four fingers of that hand. 

Insert left index from distal side into the little-finger loop, and transfer it to index. 
Take up both loops in the four fingers of that hand. 

Holding the strings taut, work the figure up and down by moving the hands. 

Sing: Kai tapiiari kai tapitari abvrkai kai o aUmsd-kak kai o. 
Ball they hit back ball not falling ball not throwing ball 

Fia. 291. Kai. 
> Eyidentl; the WMtarn utarit, gMi along. Tha Bong mw esplmioad U " Frait ha atiiik, nuggot jump iuide." 

d by Google 


23. Atr^ l«, Quui going backwards, Men 

Position L 

Pass left index over the ulnar string, take it ap from below on tip of finger and 

Let go right hand. 

Pass little-finger of right hand under the oblique string from the little-fii^r to the 
index of lefl hand. Hook down ulnar index string and draw out. 

Retaining loop on right little-finger pass right index distally into left index loop, 
[Hck up on dorsal surbce the palmar string from thumb to little-finger and draw out. 

Release indices and little-fingers of both hands gently. 

Insert left little-finger into thumb loop. 

With right hand draw the figure along by the central stringa. 

Sing: Lokoi dirmeda, Lohai itimeda. 

(a man's name) Lokoi shoots. 

Fio. 393. Hftu going baokwarda. Fia. 99S. Sinaw. 

24. Sinaur, Mer. 
Opening A. 

Pass the right foot between the right thumb and index loops and pull down the 
radial index string with the big toe. 

Release right index, stretch the toe string, and drop the other right hand loops, 
letting them hang loose. 

Insert right thumb and index into the left index loop and pick up the ulnar thumb 
and little-finger radial strings and return. 

Drop the index loop and put all the left hand fingers through the thumb and little- 
finger loops. 

Work the loop up and down the strings. This gives a sawing movement, the hands 
approach as bhey rise and are drawn apart as they ^nk. 

25. Kokowa, a land crab, Saguane, Eiwai Island. 
(This figure was collected by Mr S. H. Ray.) 
Hake Ti meta (No. 11). 

Put the little-fingers from the proximal side into the thumb loops. Release thumbs. 

Pass thumbs away from you through little-finger loops and to the pahnar side of 
the double strings running troia index to little-finger. With backs of thumbs take up 
these strings, returning through little-finger loops. Release little-fingers. 

H. v<d. IV. 48 

d by Google 


Pass tittle-fingere from the proximal side into the thumb loope, and release thumbs. 

A straight string passes from index to index. Take ap this string from the 
proximal side, close to the indices, with the thnmbs. Release indicee. 

Put indices into thumb loops towards you and withdraw thumbs. 

A loop passes from the centre of each palmar string to the outer angle of the 
central lozenges; take up with the thumbs finm the proximal side the string of this 
loop that lies nearest to you. 

Brii^ thumbs together, tip to tip, and exchange the loops, the left passing under 
the right. 

Pass the middle-fingers distal to the index loops and take up the ulnar thumb 
string from the proximal side. 

Release thumbs and pass them into the middle-finger loop from the distal side, and 
take up the ulnar middle-finger string from the proximal side. Release middle-fingers. 

By these two movements the thumb loops are token off the thumbs, twisted once, 
and replaced. 

With the thumbs take up &x>m the proximal side the radial index strings, and 
return through the thumb loops, allowing original thumb loops to slip off. Release 

Pass indices from the proximal side into the thumb loops and withdraw thumbe. 

One of the two radial little-finger strings of each hand goes across the figure and 
crosses the corresponding string from the other little-finger in the middle within a central 
triangle. (If not apparent this triangle will become so by a slight manipulation.) 

Take up these strings firom the proximal side at the point at which they cross 
the triangle with both thumbs, so that there is a double string running from thumb 
to thumb. 

With the thumbs, from the proximal aide, take ap the radial index strings and 
return through the thumb loops, allowing original thumb loops to slip off. 

Release indices and extend. 

This figure represents a land crab with its nippers held up. 

Flu. 294. Land orab. 

d by Google 



26. Mbnan, a, lizard, Mer; Maita, inteatines of a turtle, Mabuiag. 
Hold the string in the left haad bo that the loop hanga down from it 

PasB the right hand through the loop away from you, then turn the fiugera down- 
wards and pass them round the right string towards you ; pass the hand between the 
hanging strings and your body, and bring it forward to the left of the left string; 
turn the fingers up and bring It back towards you between the two strings. 
Pull the hands apari> and the right band is released. 
Sing: Miman bapitili Pethriem enau enau aroem. 

Lizard rolls to Peibri a fruit (p. 133) for eating. 

27. Keln mokeis, the mouse, Mer. 

Hold the left hand with the thumb uppermost and the fingers directed to the 
front. Put the whole left hand through the string letting the loop foil down its dorsal 
and palmar aspects from the radial side of the thumb. There will then be a pendant 
palmar and dorsal string on the left hand. 

Pass index of right hand beneath the palmar string and between the thumb and 
index of the left band, then hook it over the dorsal pendant string, bringing it out 
between the thumb and index of the left hand. Give the loop thus made a twist 
clockwise and place it over the left index. Pult tight the pendant strings. 

Again pass right index beneath the pendant palmar string and between the index 
and middle-fingeis ; hook it over the dorsal string aa before; bring this string out, 
twist the loop clockwise and put it over the middle-finger. Pull tight. 

Repeat so as to make similar loops over the ring- and little-fingers. Pull all the 
strings tight. 

Remove the loop from the left thumb and put it between left thumb and index. 

This loop represents the ear of the mouse appearing through a crack. Make a 
squeaking noise, and, when another person (the cat) attempts to catch the mouse, pull 
the palmar string with the right hand and make the mouse disappear suddenly. 

28. Au mokeia, the rat (Uromys cervinipes, Gould), Mer. 

Hold one end of the loop with the right hand and the other end down with the 
wrist of the left hand, palm downwards. 

Pass both strings between index and middle-finger of the left hand. 

Release strings of right hand, and, keeping strings parallel, take up the transverse 
string of the palm through the others and pull tight. 

Pass loop with right hand over the whole left hand, and pull tight 

Bring back both strings between index and middle-finger. 

Pass both strings between thumb and index and round thumb to palm, and back 
between index and middle-finger, keeping radial string uppermoet all the time. 

Bring together index and middle-finger to hold the strings in place, and bring the 
loop forward over the whole hand, passing the radial string to the radial side of the 
hand. Draw tight. 


d by Google 


from distal to proxim&l side beooath the tranavene palmar string, 
pass the loop backwatde over the whole hand. 

two distal loops of the thumb, and doubling them back, bold the 
D index and middle-finger. 

the loose string on the back of the hand, squeak and puU tight, 
held between the thumb and index. The string will run out. 

he pilot fish (Naucrates) that accompanies a shark, Mabuiag. 
on the lefl middle-finger, the strings lying midway on the back of 

and of the loop over the left 
it twisting the loop, so that 
ra the same radial string, 
ble-finger loop over across the 
lond between the thumb and 
r side, drawing it tight. Bring 
loop over the middle-finger to 
) that the dorsal cross strings 
mple loop on either side of the 

m towards you. Take up a 
middle-finger ulnar string and 
through the little-finger loop, 
lop thus formed take up a small Fio- ns. Pilot Otb. 

'finger radial string. 

ively with the distal and proximal strings which pass between thumb 
he last loop on the thumb and pull the free strings till the figure 

rith the left hand and sing: 

;ermoi koeia vxtra daka nguti wara daka nguti aigitaian. 

nes on each side of the &ce and head and also the top of the head. 

umb, keep on pulling and sing while the figure dissolves: 

m, e Wawu ntnu gvba, e Naigai ninu guba, e Kvki mnu gnba. 

td. East thy wind. North thy wind. West thy wind. 

off the little-finger and gently pull out the two strings which pass 
• the back of the hand and sing: 
ibva wada gudia sagu gudia iangeta mata mit. 

fish (p. 157) mouth octopus mouth, 
ring from middle-finger, keeping it in its cup-like loop, put the chin 
and bobbing it up and down sing : 

bua waruna i&ua boina ibua dangalau ibua kaitena ibua. 
turtle dugong. 

d by Google 


30. Lnoer, food, Mer; Ai, food, Mabuiag. 
Position L 

Pass indices over the little-finger atnngs and take them up from below. Return, 
bringing the part raised in an oblique line acroea the fingera 

With thumbs take up this oblique string from below and return below thumb string. 

Pass little-fingers over the ulnar index string, and take up from below with the 
backs of the Uttle-fingers the radial string of the oblong. 

With thumbs take up fr«m below the remaining string of the oblong (now become 
the radial string) and return; 

Release index loop of left hand, letting it lie looeely on the palm. 

Offer it to another person and say "Will you have a yam?" when he saya, "Have 
you any food for me 7 " pull the strings and the yam disappears, and say at the same 
time, "I hav'nt any." 

Repeat with right hand. 

31. Buli, a fly, Mabuiag. 

Hold the string between the index and thumb of each band about 6 inches apart. 
Make a small circle by bringing the right hand towards you and to the left, and place 
the string it has been holding between the left index and thumb to the near side of 
the string already held. 

Put this double string between the teeth with the small loop hanging down and 
hold the long loop straight out with the left hand. 

Put the right index from below into the long loop, then bending it towards you, 
hook it over the small ring, the tip pointing downwards. 

Turn the finger up towards you and to the right until it points upwards, then 
bring ib between the two strings of the long loop from below and put the tip on 
your noee. 

Release the strings held in the mouth, at the same time pulling the long loop and 
protruding the tongue. The string should come off the right index. 

d by Google 


The decorative art of the Islandere is so closely coonected with that of the natives 
of the adjacent maialand of New Guinea (Daudai) and of the islands at the mouth of 
the Ely River, especially Eiwai, that it is impossible to study them apart. As far as 
possible I shall state where a given specimen comes from, but it must be remembered 
that many decorated objects have been imported from New Guinea to Torres Straits 
and possibly a few have passed in the reverse direction. In the case of many museum 
specimens no provenience is given. 

The artistic sense of a people is exhibited alike in the forms of objects, the designs 
employed, and the scheme of decoration, but for the sake of clearness I have thought 
it better to keep these last two subjects distinct. On looking at a large number of 
artifiuits it is evident that the natives have a decided feeling for form. It must be 
remembered that a hundred years ago iron was unknown, and carving was done solely 
with stone and shell tools. There is every reason to suppose that all the forms and 
all the characteristic patterns were established before the coming of Europeans. The 
technical skill and sense of contour and [HY>portion exhibited by soch objects as the 
dugong harpoon and the drums are of no mean order when we bear in mind the tools 
with which they were executed. The following subjects are dealt with in this section: 


Dbookativb and Pictobial Motives 

Patterns axid designs 


of scenery and naturtd phc 

BepreeeatatJODs of pUnts. 
RepresentatiooH of aDimsls .... 
BeprcoentatioDS of human beings . 
Hepreeentations of various objects . 
Thi DacoRATioM OF Special Objects 


Dugong liarpooua 

Tobacco pipes 


Names and Sionificancb of Pattbbnb Ain> Desioks 
Treathbht of Dbgobated Subfacbb 


Carviho in thb Bound 

d by Google 

decorative, pictorial and glyptic art 343 

The omamentatioQ of Sat sur&ces is usually accomplifibed by one of two methods: 
L By the patterns being carved or cut in the wood or turtle-shell (tortoieeshell) 
of which the object is made. This carving is typically in low relief and the designs 
are simple in character. The intaglio portions are frequently filled in with lime, bat 
other pigments may be employed (figs. 298—300 A, 348—350, 354, 369, 360, pis. X. 
figs. 4, 7, 11, 13, XI. fig. 7, XXX.— XXXIL, XXXV. fig. I). 

n. By drawings or patterns composed of finely cut, scratched or punctate lines. 
Of these there are three main varieties: (A) Simple incised lines, such as the animals 
engraved on drums (figs. 351 — 353, 355, 356) and engravings on canoes (figs. 209 — 211) 
or on objects made of shell (figs. 60, 204, pi. IX. fig. 21 ; Vol. v. pi. XI. fig. 1); they ara 
also cut on turtle-ehell (fig. 300 b, c, pi. XI. figs. 9, 10, 15, 16) and rarely on bamboo 
(figs. 43. 372). 

Fio. 29S. Bubbinga ol ligits line* on bamboo tobacco pipw. Nat. iiie. A, remarkably mioDte zigzigged 
liaea on a pipe in tlie Brituh Mnsenm, aaid to be trtym Cape York, bat oertainly of Struts worknianiihip. 
B, Doaner design on a pipe obtained b; Jukes from Bmb (British Hnseoni 46. 7—81). C, from a pipe 
bom the Fly Biver (Dtesden 4189). 

(B) Fine zigzagged lines are very characteristic of the islands and the adjacent 
portion of New Guinea. The zigzags may be extremely small and close together 
(fig. 296 a), or they may be boldly rendered (fig. 296 b) ; all gradations between these 
two occur in Torres Straits. At the mouth of the Fly River the zigzags are usually 
coarse and may be very large (fig. 296 c). 

(C) Even more characteristic are the designs engraved in punctate lines. Although 
objects decorated with punctate lines have been brought from Daudai (fig. 297), this 
method appears to be pre-eminently characteristic of the islands. 

d by Google 



Finely engraved zigzag and ponctate lines are employed in the decoration of many 
objects which have a smooth polished surface, mainly on 
those tnade of bamboo; they are rather rare on objects 
made of turtle-shell (pi. XI, fig& 5, S, 14). This technique 
is never employed on wooden objects, neither does pearl- 
shell seem to be adapted for it, as simple incised lines alone 
are used for its decoration (figa, 74, 308, pi. IX, fig. 21). 

In some cases, as in fig. 300 B, c, the incised lines 
constitute the pattern, but more frequently the cut out 
portion of these comparatively coarse patterns is filled in 
with lime, and as the carved sur&ce is usually blackened 
the designs stand out in bold relief, as in figs. 300 A, 
349, 354. This form of decoration is applied to wooden 
objects such as combs (p. 361), drums (p. 364), dugong 
harpoons (p. 373) and the like. Portions of masks and 
other objects made of turtle-shell are often ornamented g 
with patterns of this nature (pis. XXXV. fig. 1, VII. fig. 4, 
XI. fig. 7). 

The pattern very rarely varies in the same band as 
in figs. 300 B, 356 B. On comparing the band shewn in 

fig. 300 B with that on the left side of the mask, which Fio. 397. Puaotata linea on {dps*, 
represents a human &ce (A. B. Meyer, Maaken von Nm ^m^ m u^fe^ttma Uanu 

Ovinsa, eta, Taf ii.), it will be seen that the two sides ^ S" °ru'*i„",°*!J^u *""'■ 

. ' ' B, from the bowl of the mm 

are practically symmetncal with each other ; it is probable pipe. Nrt. mm. 

BabbingB of th« huidleB of two wooden ooinba liom lira ; ] nftt. size. 
Pig. 398, in aothor'a ooUeotioD. Fig. 299, in British Muwam (see 
pL X. fig. 11). 

Fio. SOO. Bnbbinga of patterns oat in 
tDrtle-BliellobjeatB,^. A,froRiBids 
□f a face on a mask from Mabniag 
(Dreaden, 0896). B, oentnl part 
of a (abajfomr I (BritiBh MuwDin). 
C, npper end of a Ur. Britiah 
Uoeenm (pi. XL Dg. 18). 

> In Mer the oomb-like inarkingB were called mi Urtg war, clam-Bhell (Trid&onft) tooth pattern. 

d by Google 



that the zigzag line which surroundB the chin is intended to indicate a beard. The 
lozengee of this figure were called piau in Mer (p. 36) and thus probably repreeent a 
frontlet of nautilus nacre. This mask is referred to on p. 299. 

Dbcobativb and Pictoeial Motives. 
Pattern! and X>«sigi». 

By &r the greater number of patterns consist of straight lines, zigzags, triangles, 
lozenges and chevrons, all of which are shewn in fig. 298. 
The Islanders share the common Papuan custom of inscribing 
their patterns within parallel lines, theae are first drawn and 
then the pattern is delineated. 

Patterns carved in low relief or cut with broad or fine 
lines are always of a simple character. The commonest 
pattern is the ubiquitous zigzag, which is nearly always cut 
in low relief, the intaglio triangles usually being filled in with 
powdered lime, which further emphasises the dark zigzag band. 
Occasionally the alternate triangles are painted in colours, in 
which an alternation, say of red and blue, is not uncommon. 
It is obvious that when several rows of zigzags are drawn, aa 
in fig. 299, the base of a triangle of one row will so coincide 
with that of a contiguous row as to form a diamond or 
lozenge. I am inclined to believe that this is one way in 
which this simple form was discovered in the district. Even 
now, after generations upon generations of designers carving Pia, soi. 
the same simple patterns, the lozenge is most frequently made 
by drawing a median horizontal line parallel to the boundary 
lines and then cutting a more or less symmetrical triangle on 
each side of it (figs. 298, 299, 370). 

on » bunbcn tobfteeo box 
Mid to oome from O^pe 
Tork but ocrttinlj Bude 
b; a ToiTM Bti^ta Idsndei 
(Brit. Htu. 6IK0, KuumH 
CoL); radoaadbrt. 

If the BnggeatioQ of Profeesor J. L. Myres {Ihport, Bril. Aiaoeiation, 1907, p^ 665) be 
adopted it may be stated that the essential feature of the patterns of the Torres Straits 
Islanders is that they are limited by transverse or eacircling bands. Each band typically ia 
filled by a zigzag Hue forming a single series of triangles which are alternately enhanced 
internally with boricontal hatches, rarely are they enhanced internally from the left or right 
and in no case is this oblique batching parallel with one of the sides of the triangle. 
Triaugles or lozenges may be enhanced by iutemal repetition (figs. 328, 356 a). 

We also find what Prof. Myres terms "between parallel lines, a convergent series of 
recurrent opposite triangles," in rare cases the apices may not touch (fig. 296 a), or they may 
be confluent (fig. 376), but most frequently the apices are separated by a line (as in the 
lowest design but one of fig. 298). 

Rarely can a pattern be described as "between parallel lines, a oonvei^nt series of 
recurrent alternate tiiangles," thus forming an intermediate zigzag spaoe. That is, in Torres 
Strula zigzags are always intentional and not incidental. 

B. Vol. IT. 44 

d by Google 


BepreientaUotM of Scenery and Natural Phenomena. 

I know of only two objects decorated with a landscape, botb of which are pipes. 
One in the Oxford Museum, which I obtained at Mabuiag, is evidently a representation 
of that hilly island. The provenience of the second (fig. 302) is nnknown, but I have 
little doubt that the island of Mer is here intended, on account of the shape of the 
hill and the {U'esence of dome-shaped structures which I take to he the bee-hive huts 

Fia. SOS. Tiuiiig ot » landaoap* engnved on ft tobkooo pipe in Berlin (n. 860). A Uttla 1< 
ttuD i orig. dse (1S8 mm.). 

Fia. SOS. Bkeloh of H«r from Uu ■outh- 

i, ahewing the hill Qelun, 

which characterise the Eastern Islanders. I add, for comparison, a rough sketch (fig. 303) 
which I took of this island as seen fixim the south-west by west. The natives have 
a legend that this hill, Qelam, was originally a dugong (Vols. III. p. 248, v. p. 38, 
VI. p. 23). The eye in the native drawing indicates the prominent block of rock which 
forme what the natives term the eye of Qelam, the projection at the extreme left of 
the drawing is evidently intended for the small jutting rocky escarpment at the QeUim 
pit, Gelam's noee, at the south end of the island (see map, VI. p. 170). The break in 
the ground, below the foremost bird, probably indicates the hill Korkor, which occurs 
at the beginning of the tail of the dugong in my sketch and marks the further side 
of the original crater. The part extending beyond this is the lava-dow which forms 
the north-eastern half of the island. The vegetation is su^ested in a veiy perfunctory 
manner. I do not know what the lines that stream from the apex of the hill are 
intended for, one informant called them atwer, smoke, probably they represent a cloud. 
I should add that to make it approximately accurate the native picture should be 
reversed, assuming my identification to be correct. The view is suggestive but it is an 
impossible one bom any one standpoint, and it appears to me that this is characteristic 
of a great deal of the pictorial art of savages. 

I have quite a considerable number of native drawings of islands, and maps of 
ishuids and coral reefs drawn by natives (p. 229). The map of Mabuiag by Waria in 
Vol. 7. p. 163 may be compared with that on p. 7, and so may the profile of that 
island on p. 60, where also a plan is given of a series of coral reefe. 

d by Google 


Alluaioii has previously been made to the representatioD of waves (fig. 338), ripples 
(fig. 305 a) and footprints (figs. 326, 332). 

At our request various natives made drawings of constellattons and other heavenly 
bodies with variable success (figs. 215 — 223), and Qizu drew for us rain felling from 
a dark cloud and waterspouts (v. figs. 75, 78). 

Kepraaentatioiif of Plant*. 

It rarely happened that a native spontaneously 
made representations of plants, a notable exception 
0CCUI8 on the pipe figured on pi. XXXVII. fig. 2 
and examples on two other pipes have just been men- 
tioned. In some of the drawings made for us trees 
or other vegetation are represented (v. figs. 6, 47, 
66). Fig. 304 may represent a plant form, but this 
is doubtful, it occurs on the same pipe as fig. 302. 

Kepres«ntatfoiu of AnlmaU. 

In the following account I shall enumerate in a systematic manner the animal 
forms depicted by the Islandeis and shall only refer incidentally to the objects on which 
they occur; further information on this point will be found in the explanations of the 

Owing to the absence of indigenous mammals in the islands (with the exception 
of rats and bats) and the paucity of birds, it is not surprising that these are poorly 
represented in decorative art. Almost the whole of the animal fixtd of the natives is 
obtained bom the sea, and consequently they have a considerable acquaintance with 
marine animala This satis&ctorily explains the preponderance of marine forms that are 
represented, a parallel to which can be found in E^rly ^gean and Cretan art and the 
ceramic decoration of the ancient Yungas of the coast of Peru. 

Not many invertebrate animals are depicted; the following is a &irly complete list 
of those I have come across. On a pipe from Muralug (British Museum 6521 ; Album, 
L pi. 318, No. 3) two objects are engraved, one appears to be a jelly-fish, one of the 
Leptomeduaee (fig. 306 A), and the other a sea-urchin or sun-star (fig. 305 fi). Especially 
common on the reeGs of Mer are the large blue starfish, Linckia Isevigata, titui-titm 
(W.), tauri-sauri (E.), which unlike other starfish very fi-equently have only four instead 
of five arms. As these are very noticeable objects at low tides, it is not surprising 
that they have been copied; a [npe formerly in the Museum of the London Missionaiy 
Society is covered with a number of these starfish, nearly all of which have four arms; 
they also occur on a pipe fivm " Cape York " in the British Museum. A starfish, one 
of the Asteroidea, is drawn on another pipe in the British Museum firom Parama 
(fig. 306 c). 


d by Google 



On the large Saibai drum (p. 367) there are two representations of a man holding 
in one hand a. fish, which was identified in Mabuiag as a gaigai (p. 352), and in the 
other a spiny lobster or crayfish (Palinurus), kaiar; at the tail end of each crayfish is 
a Tosette-Uke figure, which was said to be a gangar, or bote in a rock, from which the 

Fia, S06. BngnTingi of fniin«.l» and men od bamboo Mid wooden □bjeota, rednoMl abt. ). A, jeUj-fiah, vtd 
B, Btm-aUr, Honing, 0, aeft-Drahin or starfiah, Pannift. D, tiger ahaik. E, buiuiier-heMled aliark, Paramft. 
F, ating-iay, Parama. O, king- or qneen-QBh, Mnralug. H, ran-flih, Mer. I, sea make, Farama. J, E, men 
with dngong harpoon, Parama. L, man with tortla, Saibai. All an on pipei eioept H whioh ia on a oomb 
and L on a dram. Bridih MDaenm. 

man is pulling the crustacean. Qizu drew for me a kaiar (fig. 306) in which the 
binunons antennales, the large spiny antennn and the five pairs of ' ambulatory 
appendages are correctly indicated, the great nippers characteristic of true lobsters have 
been correctly omitted. 

On learning that a design on a Mabuii^ mat (p. 66, pi. Xm. fig. 2) was called 
baidaanau ipi (or ipitnga), shark's wife, I asked Qizu to draw one for me, and he made 

d by Google 



two drawings (fig. 307) of what is evidently an Isopod which is ecto-parasitio on sharks; 
the design appears to be concerned only with the feet of the crustacean. The upper- 
most design on the mat might very well have been mistaken for the sucker of a gapu 
(figs. 316, 317, 351). 

Fio. 807. Dnwing b; Oiia ot ui Isopod 
pftmitia OD iharki, ). 

Via. 306. Dnwing ot a kaiar by Oiin, rednoed } . 

So br as I am aware the centipede (Scolopendra), aag (W.), isi (&.), is the only 
terrestrial invertebrate that is delineated. It occurs on two or three omaiter or dagong 
harpoons (figs. 365, 367) that are employed in nefarious magic (ti. p. 224). It was also 
employed by women as a scarification design in Mer (fig. 15) and Daudai (fig. 42), 

Gizu drew for us several invertebrates such as a spider in its web, a cuttle-fish, 
squid, etc 

Fio. sob. Two abatka (7 Charourodon rondeletii) uid a tntUe engraved on ■ peul-ahell, 
BerUa Haeeain (n. 667), aee flg- 7^; i- 

Amongst the vertebrates fish are very frequently represented, especially sharks and 
rays. Two sharks (fig. 308) are engraved in plain lines on a pearl-shell. I believe that 
the shark, baidam (W.), b«ieam (R), which plays a considerable part in the magico- 
religious life of the people is the handsome striped or spotted " zebra- " or " tiger- 
shark" Stegostoma tigrinum', at all events this species is drawn three times on a pipe 

1 Bat ths Bhukc of fig. 806 « 
not ooii&n«d to any one apeoies. 

a alM identified b; natives m baidam or bdlan, ao paAapi tUa n 

d by Google 


in the Britdsh Museum (fig. 305 d) and twice on a pipe in the Dover Uusenm (fig. 374) ; 
it was also represented by masks (fig. 255 a). 

Fia, 809. Babbing ol m ating-n; (Fteroplatea) and two hunmer.headed Bhulw o 
Cambridge Maeenm (0. m. 9^, 1- 

I bKmiuni dnmi. 

The hammer-headed shark (Zyg%na), Arum (W.), irvjapap (E.), from its stnmge 
appearance naturally excites interest, and 
probably for this reason it enters largely 
into the magico-religious system of the 
natives. Drawings of it occur on pipes in 
the British Museum (figs. 305 e) and Cam- 
bridge Museum (pi. XXXVII. fig, 1); two 
are engraved on a drum in the latter 
museum (fig. 309). I also give a drawing 
by Gizu (% 310). 

Fio. 310. Huiimsr-headed abuk dnwD bj Oini, \, 

d by Google 



Tida ahark fonned the main feature of certain Western and Eastern masks (figs. 25S c, 
256 A, c£ also Vols. v. pp. 64, 374, pi. XXU.; vi. p. 291, pis. XXIX., XXX.; A. B. Meyer, 
Mcuknt, etc. p. 4, TaC iii.). 

Skates and rays are frequently represented. The sborel-nosed skate (figs. 60, 311, 812) 
(Bhinobatas), kaigaa (W.), werpirupiru (E.), 
was a totem in the west (v. pp. 164, 164, 
305). I know of several drawings of the 
ating-tay (Trygon), gtoiar or gwier, taimer. 
etc (W.), goar, tapim (R); the spines can 
be seen at the base of the tail in tig. 305 F. 
Another sting-ray (fig. 309) is probably a 
Fteroplatea, it was identified as ptikai (W.) 
and kumaaar (E.). 

Fio. 811. Shorel-noeed sluite engnTed on b tab*eoci 
ptp« (Exeter Moaenm, eth. 116*), (. 

Pia. 8ia. SbOTel-noeed skate drawn b; OIed, J. 

The largest of the