IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 // 
 
 ^ .5^^. 
 
 ^J^ M^ 
 
 y^Ts' 
 
 .v'A,^^ 
 
 
 1.0 
 
 I.I 
 
 tta|2j8 |2.5 
 
 1.8 
 
 u 
 
 L25 iu ii.6 
 
 V] 
 
 Va 
 
 
 7 
 
 7 
 
 -«^ 
 
 Photographic 
 
 Sdences 
 
 Corporation 
 
 23 WEST MAIN STREET 
 
 WEBSTER, N.Y. 14580 
 
 (716) 872-4503 
 
 
 ^^"^^ 
 
 ^ 
 ^ 
 
 
CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHIVI/ICMH 
 Collection de 
 microfiches. 
 
 Canadian Institute for Historical Microreproductions / Institut Canadian de microreproductions historiques 
 
Technical and Bibliographic Notea/Notea tachniquaa at bibliographiquaa 
 
 Tha Inatituta haa attamptad to obtain tha baat 
 original copy availabia for filming. Faaturaa of thia 
 copy which may ba bibllographically uniqua, 
 which may altar any of tha imagaa in tha 
 raproduction, or which may aignificantly changa 
 tha uaual mathod of filming, ara chackad balow. 
 
 D 
 
 El 
 
 n 
 
 n 
 
 Coloured covara/ 
 Couvartura da coulaur 
 
 I I Covara damaged/ 
 
 Couverture endommagte 
 
 Covera reatorad and/or laminated/ 
 Couverture reataurAa «t/ou peilicuiie 
 
 Cover title miaaing/ 
 
 La titra de couverture manque 
 
 Coloured mapa/ 
 
 Cartea gAographiquea en couleur 
 
 Coloured ink (i.e. other than blue or black)/ 
 Encre de couleur (i.e. autre que bleue ou noire) 
 
 I I Coloured platea and/or iiiuatrationa/ 
 
 n 
 
 Planchea et/ou iiiuatratior • en couleur 
 
 Bound with otSer material/ 
 RailA avac d'autrea documentr 
 
 Tight binding may cauae shadows or diatortion 
 along interior margin/ 
 
 La reliure serrie peut cauaar de I'ombre ou de la 
 diatortion la long de la marge intArieure 
 
 Blank leavea added during restoration may 
 appajr within the text. Whenever poaaible, theae 
 have been omitted from filming/ 
 II se peut que certainaa pagea blanchea ajouttes 
 lors d'une restauration apparaiaaent dans la taxte, 
 maia. lorsqua cela Atait possible, ces pages n'ont 
 pas 6tA filmtes. 
 
 Additional commanta:/ 
 Commantairea supplAmentairas: 
 
 L'Institut a microfilm^ la mailleur exemplaire 
 qu'il lui a 6tA poaaible de ae procurer. Lea dttaiis 
 de cet exemplairo qui aont paut-Atre uniquea du 
 point de vue bibliographiqua, qui peuvent modifier 
 une image reproduite, ou qui peuvent exiger une 
 modification dana la mithoda normala de f ilmage 
 aont indiqute ci-dassous. 
 
 r~~| Coloured pages/ 
 
 D 
 
 Pages de couleur 
 
 Pages damaged/ 
 Pagea andommagiaa 
 
 Pagea restored and/oi 
 
 Pagea reatauriaa et/ou peliicultoa 
 
 Pagea discoloured, stained or foxei 
 Pages dAcolor6es, tachattea ou piquAas 
 
 Pages detached/ 
 Pages ditachAas 
 
 Showthroughy 
 Tranaparence 
 
 Quality of prir 
 
 Quality inigaia de I'lmpreasion 
 
 Includes supplementary materii 
 Comprend du mat6riel supplAmentaire 
 
 Only edition available/ 
 Seule Mition disponible 
 
 I I Pages damaged/ 
 
 |~n Pagea restored and/or lam'nated/ 
 
 I — I Pagea discoloured, atained or foxed/ 
 
 I I Pages detached/ 
 
 I I Showthrough/ 
 
 I I Quality of print variaa/ 
 
 I I Includes supplementary material/ 
 
 I — I Only edition available/ 
 
 Pagea wholly or partially obacured by errata 
 alipa, tiasuaa, etc., have been refilmed to 
 enaure the best possible image/ 
 Lea pagea totalament ou partiallement 
 obacurciaa par un feuillet d'errata, una palure, 
 etc., ont M filmtea A nouveau da fapon k 
 obtenir la meilleure image possible. 
 
 This item is filmed at the reduction ratio checked below/ 
 
 Ce document est film* au taux de rMuction indiqu* ci-daaaous. 
 
 10X 14X 18X 22X 
 
 12X 
 
 16X 
 
 20X 
 
 26X 
 
 30X 
 
 z 
 
 24X 
 
 28X 
 
 n 
 
 32X 
 
Th« copy fiimad h«r« has b««n raproducsd thanks 
 to ths ganarosity of: 
 
 Library Division 
 
 Provincial Archives of British Columbia 
 
 L'axamplaira filmi fut raproduit grAca A la 
 ginArosit* da: 
 
 Library Division 
 
 Provincial Archives of British Columbia 
 
 Tha imagas appaaring hara ara tha bast quality 
 possibia considaring tha condition and lagibiiity 
 of tha original copy and in kaaping with tha 
 filming contract spacifkations. 
 
 Original copias in printad papar covars ara filnrtad 
 baginning with tha front covar and anding on 
 tha last paga with a printad or illustratad impras- 
 sion. or tha back covar whan appropriata. All 
 othar original copias ara filmad baginning on tha 
 first paga with a printad or illustrated impres- 
 sion, and ending on the last page with a printed 
 or illustrated impression. 
 
 The last recorded frame on each microfiche 
 shall contain the symbol — ^ (meaning "CON- 
 TINUED"), or the symbol V (meaning "END"), 
 whichever applies. 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios. Those too large to be 
 entirely included in one exposure are filmed 
 beginning in the upper left hand corner, left to 
 right and top to bottom, as many frames as 
 required. The following diagrams illustrate the 
 method: 
 
 Les imagas suivantas ont At* reproduites avac la 
 plus grand soin, compta tenu de la condition at 
 da la nattati da raxemplaire filmA. et en 
 conformitA avac las conditions du contrat da 
 filmage. 
 
 Les exemplaires originaux dont la couvarture en 
 papier est ImprimAe sont filmAs en commen^ant 
 par la premier plat et en terminant soit par la 
 darnlAre page qui comporte une empreinte 
 d'impression ou d'illustration, soit par la second 
 plat, salon la cas. Tous las autres «>xemplaires 
 originaux sont filmAs en commenpant par la 
 premiAre page qui comporte une empreinte 
 d'impression ou d'illustration et en terminant par 
 la derniAre paga qui comporte une teMe 
 empreinte. 
 
 Un des symboles suivants apparaitra sur la 
 derniAre image de cheque microfiche, selon le 
 cas: le symbols — »> signifie "A SUIVRE ", le 
 symbole V signifie "FIN ". 
 
 Les cartes, planches, tableaux, etc., peuvent Atre 
 filmAs A des taux da rAduction diffArents. 
 Lorsque le document est trop grand pour Atre 
 reproduit en un seul clichA. il est filmA A partir 
 de Tangle supArieur gauche, de gauche A droite, 
 et de haut en bes, en prenant le nombre 
 d'images nAcessaire. Les diagrammes suivants 
 illustrant la mAthode. 
 
 1 
 
 2 
 
 3 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
n 
 
7 
 
 Jan. 1891.] INDIAN PHYSICAL CHARACTERISTICS. 
 
 25 
 
 PHTSICAL CHARACTBRISTICS OF THB INDIANS OF 
 THB NORTH PACIFIC COAST. 
 
 BY FRANZ BOAS. 
 
 During the past summer, when visiting the North Pacific coast, I 
 made a series of measurements of 263 Indians. The tribes included 
 in this series occupy the coasts of Oregon, Washington, and British 
 Columbia, and belong to many linguistic stocks. As the series is 
 not sufficiently large to allow a separate treatment of the individuals 
 of each tribe, I have grouped them together according to their former 
 habitat and, to some extent, according to their mode of life, the 
 tribes living on reservations being separated from those passing much 
 of their time in the canoe. I have distinguished the following 
 groups : 
 
 1. Southern Oregon, including principally tribes of Athapaskan 
 affinity, but also a few Klamath, Shasta, etc. The northern limit of 
 this group is situated at Euchre creek. 
 
 2. Central Oregon, including the tribes between Euchre creek 
 and Umpqua river. 
 
 3. Northern Oregon, including the Yakonanand Salish tribes be- 
 tween Umpqua and Columbia rivers. 
 
 4. Columbians, including the tribes in the immediate neighbor- 
 hood of Columbia river and in the lower Willamette valley. 
 
 5. Tribes of Washington, including the whole coast of Washing- 
 ton west of the Cascade Range. 
 
 6. Harrison Lake and Lillooet. 
 Vancouver Island. 
 The Bilqula, on the central part of the coast of British Colum- 
 
 7 
 
 8 
 
 bia. 
 
 ■9 
 
 Tribes of British Columbia north of the last-named group. 
 
 Only a short series of measurements of each individual was taken, 
 such as could be made by the removal of only a small portion of the 
 clothing. Following is a list of the measurements : 
 
26 
 
 THE AMERICAN ANTHROPOLOGIST. 
 
 [Vol. IV. 
 
 1. Stature. 
 
 2. Finger-reach. 
 
 3. Height of ear. 
 
 4. Height of 7th vertebra. 
 
 5. Height of acromion, 
 
 6. Height of point of second 
 
 finger. 
 
 7. Width between acromia. 
 
 9- 
 10. 
 II. 
 12. 
 
 13- 
 
 14- 
 IS- 
 
 Length of head. 
 
 Width of head. 
 
 Jugal width of face. 
 
 Distance from root of nose to 
 
 chin. 
 Distance from root of nose to 
 
 mouth. 
 Height of nose. 
 Width of nose. 
 
 8. Height when sitting. 
 
 In discussing the series, I have excluded all males of less than ao 
 and more than 50 years of age, and all females of less than 19 and 
 more than 50 years of age, the former because they have not reached 
 the adult stage ; the latter on account of the incipient changes inci- 
 dent to old age. It may seem that the lower limits are fixed too low, 
 but, as the growth after 19, viz., 20 years, is very slight, the gain in 
 accuracy resulting from the average of a greater number of indi- 
 viduals far outweighs the slight error that is introduced by the im- 
 maturity of a certain number of them. 
 
 On examining the series, it appears that there are very marked 
 differences between the various local groups. In order to make these 
 differences apparent, a table of those measurements ,which vary most 
 among the different groups is here presented. In addition to the 
 average, the table conUins the limits between which one-half of the 
 series is found — /. <r., the probable variation. The last three figures 
 in each column are the finger-reach, height when sitting, and length 
 of arm, expressed in ^ermills of the stature. 
 
Jan. 1891.] INDIAN PHYSICAL CHABACTGBISTICS. 
 
 27 
 
 
 li 
 
 
 a 
 
 13 
 
 |3 
 
 Is 
 
 a 
 
 1^ 
 
 II 
 
 So 
 
 -H-H-H-tl -H 4) •« -H ■« -H 
 
 SPSS gS3|2« 
 
 q -H-H-H-H -H •«■+)-« -H -H 
 
 SPSS sssssi 
 
 ^ •W-«-H-«. -H -H-H-H-H 
 
 H-H-H- 
 
 _ -H-H-W-H -H-H-H-H-H-H 
 
 
 99«». 33SSS« 
 
 „ -H-H-H-H 
 
 -H-H-H-H-« 
 
 liiH |§iiS3 
 
 •H-H-H-H rJ-H-H-H-H-H 
 
 as22 1222*- 
 
 -H-H-H-H --H-H-H-H-H 
 
 SS3si 
 
 iisis' 
 
 -H-H-H-H -H-H-H-H-H-H 
 
 -H-H-H-H -tl-H-H-H-H-H 
 
 I 
 
 ! ! 
 
 «il 
 
 o 
 
 I 
 
 ii 
 I I 
 
 : i 
 
 a ! 
 
 3 I 
 
 •o?-- 
 
 «a 
 
 ■« M-e 
 
 i s 
 
 ill •S'S'S*-"-* 
 *._•« -a-,. ^^^ 
 
 
 ,^ 23 
 
 Iesj aisles 
 
 a 
 
 29SS 
 
 -H-H-H-H 
 
 M« 
 
 e< 
 
 -H -H-H-H-H 
 
 -H-H-H-H -H-H-H-H-H-H 
 
 -H-H-H-H 
 
 
 32«J22 
 
 -H -H-H-H-H 
 
 *am 
 
 8X0900 'l^^ei _ 
 -^ -H+l-H-H -H-H-H-H-H-H 
 
 8SS s ri^S ass 
 
 -H-H -H -H-H -H-H-H 
 
 ilii 
 
 : o ! i 
 it :m 
 
 ii5 
 
 ! « ! a S 
 
 i<eC»o 
 
 
 ISO 
 
 
 I HIS IIIII2 
 
 I 
 
28 
 
 THE AMEniCAN ANTHKOPOLOGIST. 
 
 [Vol. IV. 
 
 Considering the series as a whole, we may describe the population 
 of the North Pacific coast as of average size, ranging, when con- 
 sidered in groups, from i,6ii to 1,699 mm., with short heads and 
 rather wide faces and large chins. Their skin is of light color, and, 
 if protected from sunlight, can hardly be distinguished from that of 
 European brunettes. The hair is of a very dark brown, straight or 
 slightly wavy. The eye almost always has a well-developed plica 
 interna ; the eyebrows are wide ; the ears coarse, with small lobes. 
 
 A study of the distribution of the single observations in each group 
 of the series shows that the groups may be considered as homogene- 
 ous, so that we are justified in forming the averages given in the above 
 table. 
 
 Although the series are rather small, a number of conclusions may 
 safely be drawn from them. Wherever the series of males and females 
 shows analogous variations these can hardly be due to accident, but 
 it is fair to assume that similar figures would result from a larger 
 series. 
 
 Considering the observation on stature, we are struck by the tall- 
 ness of the Columbians and the small size of the tribes of Harrison 
 lake. A comparison of the series of males and females proves that 
 these averages are trustworthy. When, we express the stature of the 
 women in per cents, ofthat of the men, we obtain the following re- 
 sults : 
 
 Bflqulo. 
 
 94.6 
 
 These figures agree very well with each other, and with those ex- 
 pressing the same relation among other races. 
 
 It appears that the stature increases as we approach Columbia 
 river from the north and from the s .-'th. It is of interest to note 
 that the greater size of the natives of this region has been re- 
 marked upon by a number of writeis.* We find in this region 
 Sahaptin tribes, who d*^ cended from th". interior to the coast ; the 
 Molalla, part of whom live far inland, and the Chinook, whose ter- 
 ritory extends far up Columbia river, fo that an intrusion of a taller 
 race coming from the east and settling here becomes very likely. 
 The proportions of the body and of the head of the Columbians pre- 
 sent no very marked divergence from those of the neighboring tribes. 
 
 Southern 
 OreguD. 
 
 Central 
 Oregon. 
 
 Northern 
 Oregon. 
 
 Wuhlng- 
 ton. 
 
 Harrison 
 Lake. 
 
 94-3 
 
 94-8 
 
 93-6 
 
 94.2 
 
 94- S 
 
 * bancroft, Native Races, Vol. I, p. 254. 
 
Jan. 1891.] INDIAN PHYSICAL CHARACTERISTICS. 
 
 29 
 
 They are in the habit of deforming their heads ; consequently no 
 conclusions can be drawn from head measurements. The faces of 
 the Columbians are a little wider than those of their neighbors on 
 both sides. 
 
 The tribes of Harrison lake, who are remarkable for their short 
 stature, differ fundamentally from all others. The great majority of 
 individuals belonging to these tribes may easily be recognized by 
 their peculiarities. Their heads, although not deformed, are exceed- 
 ingly wide; their faces very chamseprosopic. These peculiarities 
 may be observed in males as well as in females. 
 
 In Vancouver Island is found a group of tribes which resemble 
 closely those of the northern part of the coast of British Columbia. 
 They are characterized by a size ranging between that of the tribes 
 of northern and those of southern Oregon, and by a much longer 
 head than is found anywhere farther south. The index of the three 
 individuals of Vancouver Island is 79.9; that of the seven individ- 
 uals belonging to northern tribes, 79.5. This interesting fact is con- 
 firmed by measurements of skulls from these regions. I published in 
 the Verb, der Berliner Ges. f. Ethnol., 1890, p. 30, measurements of 
 a series of 10 undeformed crania from the southeastern part of Van- 
 couver Island. These give an average index of 77.7. To this may 
 be added a skull described by Flower* belonging to the west coast 
 of Vancouver Island, which has an index of 77.4. If we allow, ac- 
 cording to Broca, 2 per cent, difference between the cranium and 
 the head -of living individuals, we find that the results of these series 
 agree very closely. We have also a series of measurements belong- 
 ing to the northern part of the coast of British Columbia. Four 
 Tsimshian skulls have been described by me.f Three others have 
 been described by Barnard Davis. Still another has been descrilied 
 by the same author as a " round head " from Vancouver Island (p. 
 229). Finally, I measured a Haida cranium in the Provincial Mu- 
 seum of Victoria, B. C. Theaverage of these 9 skulls gives 78.4. It 
 will be seen that all these values agree very closely, and confirm the 
 results obtained on the living individuals. 
 
 It is very remarkable that we find located between these two ho- 
 mogeneous groups a somewhat taller and much more brachycephalic 
 
 * W. H. Flower, Catalogue of the Specimens illustrating the Osteology, etc., in 
 the Museum of the Royal College of Surgeons, p. 148. 
 f Report to the B. A. A. S. on the tribes of British Columbia, 1889, p. 16. 
 
30 
 
 THE AMERICAN ANTHROPOLOGIST. 
 
 [Vol. IV. 
 
 people — the Bilqula. Fortunately the series of measurements on this 
 tribe is sufficiently large to give the result a high degree of proba- 
 bility. North of British Columbia we find the Tlingit, also con- 
 siderably taller than the tribes of northern British Columbia, and 
 much more short-headed. Unfortunately no satisfactory series of 
 measurements of this people is accessible to mc. As there is a con- 
 stant intercourse between the various tribes of this region, and as the 
 conditions of life of all of them are uniform, we must assume that 
 these differences are racial characters. Our knowledge of the native 
 tribes of the interior of British Columbia is very limited, but we 
 may adduce some material from the most southern part of this region. 
 Seven skulls from Lytton, B. ,C., which are in the Museum of the 
 Geological and Natural History Survey of Canada, in Ottawa ; one 
 from the same place^ which is in my possession, and two Shushwap 
 skulls described by Barnard Davis, give an average cephalic index 
 of 78.9. A single Shushwap whom I measured in New Westminster 
 has a cephalic index of 82.9, corresponding to about 80.9 on the 
 skull. 
 
 On account of the remarkable differences in the cephalic indices 
 of these various groups of tribes, it would be of great interest to know 
 the natural form of head of the tribes of northern Vancouver Island, 
 of the delta of Fraser river, of Puget Sound, and of northern Oregon. 
 Unfortunately all of them deform their heads. On Fraser river, on 
 Puget Sound, and north of Columbia river this custom is, however, 
 dying out, and I was able to measure a series ot children. Twelve 
 boys from Puget Sound give an average of 83.9, with a probable 
 variation of 1.8; thirteen from Fraser river give an average of 87.3, 
 with a probable variation of 2.0. Although the results obtained by 
 measuring boys are not identical with those obtained by measuring 
 adults, the difference is probably not very great. At least the results 
 of the measurement of the cephalic indices of 13 boys from southern 
 Oregon are almost identical with the results obtained from the meas- 
 urement of 1 7 adults. According to these figures, thelndians of Puget 
 Sound would have to be grouped in close connection with those of 
 central Oregon ; those of Fraser river with the tribes of Harrison 
 lake. The Bilqula resemble the tribes of Puget Sound and of north- 
 ern and central Oregon. 
 
 It appears, therefore, that there is a comparatively uniform, meso- 
 cephalic population on Vancouver Island and on the coast of Brit- 
 ish Columbia, with the exception of Fraser river and Bentinck Arm. 
 
Jan. 1891.1 INDIAN PHYSICAL CHARACTERISTICS. 
 
 81 
 
 The latter are at the same time the regions where intermarriage with 
 the Athapaskan tribes of the interior has been frequent. We know that 
 the Bilqula have intermarried very extensively with the Carriers and 
 Chilcotin ; the Lillooet with the latter. Therefore it would appear that 
 brachycephalism becomes more frequent in those regions where inter- 
 mixture with Athapaskans makes itself felt. In order to strengthenor 
 to refute this theory, it would be necessary to obtain measurements of 
 theSishiatl, of Jervis Inlet, and of the upper Tsimshian, both of whom 
 are in close contact with Athapaskans. The Tlingit, of Alaska, who 
 are immediate neigiiborsofAtnapaskan tribes, are also niorebrachy- 
 cephalic than the tribes of the coast of British Columbia, and their 
 stature seems to be higher. There can be little doubt that among 
 the Athapaskans on the whole brachycephalism is very frequent. 
 Quatrefages and Hamy* mention seven skulls from various A'^ i- 
 paskan tribes of Canada, and find them to be brachycephalic. \ .le 
 proportions of their faces, also, closely resemble those of the Bilqula 
 and the other tribes discussed in the present pai>er. I had occasion 
 to question a number of former officers of the Hudson Bay Company 
 regarding the general appearance of the Athapaskans of the interior 
 of British Columbia and of the Mackenzie Basin. According to 
 their descriptions, the Athapaskans resemble the tribes of the north- 
 west coast much more closely than the Algonquin. For these reasons, 
 as well as on account of the form of their language and a number of 
 customs and beliefs, I am inclined to class the Athapaskans as one 
 of the peoples of the north Pacific coast. A good series of measure- 
 ments from the Mackenzie Basin would decide this question. The 
 almost complete absence of Dolichocephali — at least, according to 
 the present state of our knowledge — distinguishes th^m most clearly 
 from the eastern group — the Algonquin and Iroquois — as well as from 
 the central and eastern Eskimo. 
 
 The divergence of the tribes of Harrison lake from all their neigh- 
 bors is very curious, more especially the high degree of brachy- 
 cephalism and of chamaeprosopy. The width of both head and face is 
 much greater than is found in any other tribe. Their small stature 
 is also quite unique. If I should include three men of about 55 
 years in the series, the average would be less than 1,600 mm. I may 
 mention here that three tibiae and femora which I collected at Lyt- 
 ton with the mesocephalic skulls mentioned above are very small. 
 
 '* Crania ethnica, p. 470. 
 
82 
 
 THE AMERICAN ANTHROPOLOaiST. 
 
 [Vol. IV. 
 
 Assuming the tibia to be about 33 per cent., the femur 37 per cent, 
 of the stature, the three individuals would have measured about 1,430, 
 1,500, and 1,560 mm., thus corresponding in size to the females of 
 the Harrison lake series. This, however, is, of course, a very un- 
 satisfactory comparison. 
 
 We will finallj consider the proportions of the bodies of the vari- 
 ous groups. It appears that the finger-reach of the southern tribes, 
 especially of those of southern and central Orep:on, is much smaller 
 than that of the northern tribes. I am inclined to attribute this to 
 a difference in occupation, the two first-named groups living on res- 
 ervations, while the others are fishermen. Hand in hand with this 
 lengthening of the finger -reach seems to go an increase in the length 
 of the arm. These variations may be seen in males as well as in 
 females. The latter also pass much of their time in the canoe. The 
 table shows that the trunk of these Indians is much longer than that 
 of Europeans, and also longer than that of the Iroquois, which, ac- 
 cording to Gould, is 53.4 per cent. It seems that the trunk of the 
 southern groups is longer than that of the northern ones. 
 
 It is of the greatest interest to observe that upon study i;, the 
 physical characters of the tribes of the Pacific coast in detail such a 
 variety of forms is found. Each tribe appears composed of many 
 types, but in each we find a marked prevalence of a certain type. 
 Unfortunately we have hardly any detailed studies on the physical 
 char'cters of living Indians, and yet these studies are just as impor- 
 tant s those on languages, customs, and beliefs. The disappearance 
 of tribes, their intermixture with each other, and with whites, the 
 changes in their mode of life, are so rapid that little time remains 
 for studies of this character. 
 
 Banana Cultivation among the Natives of New Guinea. — 
 Sir William Macgregor, governor of British New Guinea, in his 
 recent exploration of the Fly river, found that the natives of the 
 large island of Kawai, in the delta of that river, gave much atten- 
 tion to the cultivation of the banana. In a vocabulary of their 
 language which he compiled there is a list of not fewer than thirty- 
 six varieties of banana, and from personal experience he testifies 
 that the distinctions were not fanciful, but indicate real, substantial 
 differences. This is among a people whose sole utensil, used as 
 bucket, basin, bowl, and plate, is a large slipper shell. (Proc. Roy. 
 Geog. Soc; June, 1890, p. 354.) 
 
 
 H..te:.VrVttiS*S > 
 
■1