Analysis of kangaroo body part transport among the alyawara of central Australia

Journal of Archaeological

Science 1989,16,393-405

Analysis of Kangaroo Body Part Transport among the Alyawara of Central Australia JamesF. O’Connell” and BrendanMarshallb (Received

2 August

1988, revised manuscript

accepted

14 February

1989)

We analyse field processing and transport of red kangaroo (Macropus rufus) carcasses by Alyawara-speaking Aborigines in terms of an index of kangaroo body part utility.

We find that transport patterns are consistent with the goal of moving parts with the greatest amounts of attached edible tissuefrom butchering sitesto the hunters’ residential base. Only parts with relatively small amounts of associated edible tissue are consumed and/or discarded in the field. Selection of parts for transport is probably not affected by differences in butchering and transport costs. Contrary to suggestions based on data from elsewhere, Alyawara hunters do not preferentially consume relatively fat-rich or other high ranked parts at the butchering site. Almost all such parts are carried back to the residential base. Keywords: BONE TRANSPORT, ALYAWARA.

UTILITY

INDICES,

RED KANGAROO,

Introduction Recent ethnographic research clearly demonstrates that hunter-gatherers sometimes transport only a fraction of the available parts of medium and large bodied prey from kill-butchering sites to residential base camps (e.g. Gifford, 1977; Yellen, 1977;

Binford, 1978; Bunn, 1983; Bunn et al., 1988; O’Connell et al., 1988). It has long been suggested that this practice contributes to important differences in the composition of archaeological bone assemblages (e.g. White, 1952; Perkins & Daly, 1968). These in turn may provide clues to other aspects ofpast behaviour, such as foraging strategies, settlement patterns, and food sharing practices (e.g. Isaac, 1978; Binford, 1981; Bunn & Kroll, 1986; but see Grayson, 1988: 58-74). Interpreting differences in bone assemblage composition requires an understanding of the factors which determine the selection of parts for transport. The current literature identifies several, including the weight or nutritional value of edible tissue attached to various body parts, the costs of processing at the kill site, and the relative benefits of consumption at the kill versus transport and consumption at the residential base (e.g. Binford, 1978; Metcalfe & Jones, 1988; O’Connell et al., 1988; Speth, 1988). The importance of these and other potentially critical factors and the conditions under which they operate have yet to be firmly established. Until they are, inferences “Department of Anthropology, University of Utah, Salt Lake City, UT 84112, U.S.A. bDepartment of Archaeology, La Trobe University, Bundoora, Victoria 3083, Australia. 393 03054403/89/040393+13S03.00/0

0 1989 Academic Press Limited

394

J. F. O’CONNELL AND B. MARSHALL

about past human behavior based on differences in archaeological bone assemblage composition remain inherently weak. As a step toward improving this situation, we present quantitative data on bone transport by modern Alyawara hunters and on the economic anatomy of their principal prey, the red kangaroo (Macropus rz@s). We show that hunters usually transport most, but not all body parts from kill sites to base camps. Elements consistently discarded have only small amounts of edible tissue attached, or are subject to rapid spoilage and thus more likely to be lost in transport. This behavior is consistent with the goal of making body parts with the greatest amounts of attached edible tissue available to base camp residents. Although field processing and transport costs might also be a factor in element discard, we think it unlikely in this particular case, simply because the savings in transport costs resulting from discard are very small. Field consumption by hunters as a means of avoiding or pre-empting competition for nutritionally high ranked parts seems not to occur among the Alyawara, at ieast not in the cases reported here. In this sample, such parts are always carried back to base camps. The Alyawara

The Alyawara are an Arandic-speaking people whose traditional territory is centred on the Sandover River, about 250 km northeast of Alice Springs, central Australia. In preEuropean times, they lived in small, widely scattered groups, and supported themselves entirely by hunting and gathering. Direct contact with Europeans began in the late 19th century, but remained infrequent until the establishment of local sheep and cattle stations in the mid-1920s. Over the next 50 years, the Alyawara gradually abandoned full time hunting as they became increasingly enmeshed in the European economy. By 1970, most were living in large, semipermanent settlements near European homesteads or on government reserves. Some were regularly employed, but the majority were dependent on government welfare in the form of periodic cash payments and weekly supplies of rations. In spite of these changes, the Alyawara still relied on hunting and gathering for a significant part of their diet, at least through the late 1970s (for additional information, see Denham, 1975; O’Connell, 1977). Alyawara hunting and carcass processing

Alyawara hunting has been described in some detail elsewhere (O’Connell & Hawkes, 1984), and its principal features need only be reviewed in brief. The best available data on hunting were collected between 1973-78 near the settlement known as Bendaijerum, or MacDonald Downs. During that period, the Alyawara living at Bendaijerum took at least nine species of marsupials, birds, and reptiles. Of these, the most important was red kangaroo, a large macropod (mean weight 25 kg) which provided more than 90% of the meat the Alyawara obtained by hunting, and about 70% of their total meat intake, including that obtained as rations or through local cash stores. Kangaroo were taken by hunters armed with shotguns and small caliber rifles, operating on foot or in motor vehicles. Hunters on foot worked within a radius of about 10-12 km from the settlement, generally in groups of one or two. On any given day, they had about a 70% chance of making a kill, but seldom took more than one animal per trip. Hunters in motor vehicles (usually passenger cars or light trucks) operated over a wider radius (25-30 km) and in larger groups (mean party size of five). They had a higher probability of success than did pedestrian hunters (approaching lOO%), and usually took more prey, about four per trip on average. Once taken, kangaroos were either cooked and butchered in the field or brought back intact to the settlement for processing. Ease of carcass transport and availability of

ALYAWARA

BONE TRANSPORT

395

Figure 1. Hunter on foot with an adult female kangaroo.

firewood appear to be critical in choosing between these alternatives. A freshly killed red kangaroo of average size is easily carried by a single hunter on foot (Figure l), but a cooked and butchered animal is more difficult to manage. This suggests that hunters on foot should always carry prey to camp intact. On the other hand, dry firewood was scarce in the area immediately around Bendaijerum, which meant that a hunter on foot packing an uncooked animal to camp would have to bring firewood as well, or else make a second trip to collect wood. Under these conditions, hunters on foot typically carried their kill to a point where firewood was available close to camp (usually within 1 km or so of the settlement), cooked and butchered the animal there, and then continued home. Hunters in vehicles, who could easily carry both prey and fuel, often collected firewood late in the course of the hunt, and returned to the settlement to process their kill. On some occasions, however, they too cooked and butchered the animals in the field, sometimes only a short distance (3-5 min drive) from the settlement. The reason for this practice was not clear, though some informants suggested that it facilitated a quick distribution of meat shares in camp, and so reduced the opportunity for argument. It may also shave been as easy to unload and cook the animals in the bush as it was to load and later unload the firewood.

396

J. F. O’CONNELL

Figure

2. Cooked

AND

female

B. MARSHALL

kangaroo,

with disarticulated

tail and feet.

Regardless of processing location, all red kangaroos were treated in essentially the same way (Binford, 1983: 165-172; see also Gould, 1967 for a description of kangaroo processing among other central Australian groups; Solomon, 1985 for a comprehensive guide to the Australian ethnographic literature on this topic). The person who‘took responsibility for the carcass [usually, but not always an adult male; often the hunter himself, even when others were present (cf. Gould, 1967: 52)] dug a pit big enough to hold the animal (or animals), and built a large fire on the upwind side, immediately adjacent to the pit. He then dislocated both rear legs between the femur and tibia, made a cut in the lower abdomen, removed the intestines and internal organs, and set them aside. He tossed the carcass on the fire, turned it several times using the tail as a handle, then pulled it aside, and scraped it with a knife or stick to remove the singed hair. He cut off the tail and rear feet (tarsals, metatarsals, and phalanges), put the carcass, tail, and feet in the pit, covered them with coals and hot earth from the fire and waited about an- hour for them to cook. While waiting, he usually stripped the intestines of their contents, roasted them lightly in the ashes of the fire along with the liver, and ate them. About 10 min before pulling the carcass from the roasting pit, he dug out the tail and feet (Figure 2). Sometimes he stripped the skin from the tail, chewed or cut the meat from the vertebrae, cracked the latter and sucked at their contents, and discarded the bones. He also occasionally stripped one or both metatarsals of meat and tendons, ate the meat, cracked the large central metatarsal(s), sucked or picked out the marrow, and discarded the bones. At other times, he simply set the tail and feet aside. Later, he removed the carcass itself from the pit, set it on a brush palette, and divided it into 10 pieces, including the head and cervical vertebrae, thoracic vertebrae, right and left forequarters (each consisting of ribs and a complete forelimb), belly sheet, lumbar and sacral vertebrae, right and left innominate (split through the pubic symphysis and snapped off the sacrum), and right and left rear limb (each consisting of a femur, tibia, and fibula). If the animal was cooked and butchered in the field and the hunter was on foot, these pieces were skewer@ on a stick and carried back to camp. If a vehicle was available, they were simply loaded

ALYAWARA

BONE TRANSPORT

397

Table 1. Composition ofhuntingparty andrange ofprey taken on 19 Alyawara hunts, June 1974~September 1975. Women were present on two of 13 hunts in motor vehicles, children onfive. Immature kangaroos were taken on six hunts in vehicles and one hunt on foot. Euros were taken on two vehicle trips; rabbits, bustards, and rock wallaby on one each Vehicle hunts range

(N= 13) mean

Foot hunts range

(N=6) mean

Party composition Adult men Adult women Children

24 o-5 o-4

4.8 0.5 2.4

l-2

1.5 -

Prey taken Kangaroo Adult Immature Other

14 o-2 63

3.2 0.5 0.5

1-2 G-1

1.2 0.2

aboard and driven home. Animals cooked in the bush were occasionally carried back without further dismemberment, but this did not occur in the sample of hunting and butchering events reported here. Detailed quantitative data on bone discard in the field are available for 19 hunts, 13 in motor vehicles and six on foot (Table 1). The number of people involved in vehicle hunts ranged from three to 11, and averaged about six. Most were men, although women were present on two vehicle hunts, children on five. Hunts on foot involved one to two men, usually only one. The total number of animals taken on all hunts was 67, including 49 adult and eight immature red kangaroos, and 10 other animals. All but one of the immatures were very small (wt. < 2 kg) and still confined to the pouch. The 10 other animals included four euros (Macropus robustus, a medium sized macropod somewhat smaller than the red kangaroo), one rock.wallaby (Petrogale sp., a small macropod), two rabbits, and three plains bustards (Eupodotl’s australis). Hunters on foot took only red kangaroos, one per trip on five occasions, two adults and an immature on the sixth. Hunters in vehicles took more red kangaroos per trip on average (mean = 3*2), and also took all the other animals as well. Table 2 shows the parts of red kangaroos and euros consumed and/or abandoned at field butchering stations. All or part of the available intestines were eaten on at least 17 of 19 trips, including all hunts in vehicles and four of six hunts on foot. They may have been eaten on the other two hunts on foot as well, but the data are unclear on this point. Intestines and internal organs not consumed were generally taken back to camp, though on occasion they were abandoned at the butchering site. The reason for this was unspecified, but may have been related to the fact that the hunters had been very successful, having taken a total of seven animals, including five kangaroos and two euros. Tails were eaten on 13 of 19 trips. The total number consumed was 24, or about 45% of those available. Feet were abandoned on 11 trips. The total left was 56 of a possible 106 (53%). Of these, only 12 were actually stripped and cracked for marrow. The other 44 were simply discarded at the butchering station. Such discard took place on four hunts in vehicles, but on only one hunt on foot. In all five cases, the total number of animals taken was relatively high. For the one hunt on foot, it included two adult and one immature kangaroo, versus a mean for all hunts on foot of slightly more than one adult (Table 1). For the four hunts in vehicles, it ranged from four to seven adult kangaroos or euros each, versus a mean for all

398

J. F. O’CONNELL AND B. MARSHALL Table 2. Macropod

body parts

eaten or left atfield

butchering

sites

Vehicle hunts

Foot hunts

No. of trips Adult macropods taken

13 46

6 I

Elements eaten or left at butchering sites Skull Mandible Tail Forearm Femur Tibia Rear foot

0 0 19 6 1 6 46

1 1 5 1 0 0 10

Trips on which viscera were eaten or left

13

4

Immatures taken Immatures consumed

I 5

1 0

hunts in vehicles of three to four. Immature kangaroos were completely or partly consumed on five of six occasions when they were available. All those eaten were very small. Other large macropod body parts were left at field butchering sites only rarely. In our sample, such pieces included six distal tibia broken during hunts, seven forearm units (radius, ulna, carpals, metacarpals, and phalanges) which disintegrated during cooking, one skull and mandible, and one partly fused proximal epiphysis of the femur, which became separated during butchering. Of these various elements, only the skull and mandible and one forearm unit were actually stripped of attached edible tissue. The other forearm units were simply dropped and abandoned. The tibia and femur fragments had no edible portions. Differences in the numbers of parts left at butchering stations on vehicle and foot hunts, respectively, are minor. The most obvious ones are that feet and tails are almost twice as likely to be eaten and/or abandoned in the field by hunters on foot as by those in vehicles. Intestines may be less likely to be consumed by hunters on foot, though as we indicated above the data are not clear on this point. Economic Anatomy of the Red Kangaroo Data on the distribution of edible tissue on the red kangaroo are available from several sources (e.g. Tribe & Peel, 1963; Hopwood, 1976, 1981; Hopwood et al., 1976), but are generally not presented in the form required for this analysis. To remedy this situation, we acquired five animals (two males, three females) from a professional shooter based near Pooncarie, New South Wales, butchered the animals in a dissecting room at La Trobe University in Melbourne, and weighed and measured the meat and marrow associated with various skeletal elements. All these animals were killed with a small bore high velocity rifle. All were acquired by us within 12 h of the time they were shot, were completely intact except for head wounds, and were kept under refrigeration from the time they were killed until they were processed. Processing was complete within 7 days after the animals were killed. All five carcasses were skinned, stripped of internal organs and intestines, and divided into 13 sets of elements closely approximating those normally produced by the Alyawara.

ALYAWARA

BONE

TRANSPORT

399

These sets were then weighed, as were the skin, internal organs, and intestines (stripped of their contents). Larger units (forequarters, rear legs, and lumbar and sacral vertebrae) from four animals (two male, two female) were then divided into smaller sets, specifically ribs, scapula, humerus, radius and ulna, forepaw (including carpals, metacarpals, and phalanges), femur, tibia and fibula, lumbar vertebrae and sacrum, and distal sacral vertebrae. Flesh was stripped from all bones and weighed separately, as were the bones themselves. Time required to strip the flesh was recorded for one male and one female carcass. Articular ends were sawn off the long bones (i.e. humeri, radii, ulnae, femora, tibiae, and central metatarsals), the marrow removed, and the volume of the marrow cavity measured. Cavities in other marrow bearing bones, including the fibulae and mandible, were not measured because they were very small. Results of this exercise are presented in Table 3. Analysis

The recent literature contains at least three explanations for patterns observed in prey body part transport among hunters. One is that within the limits of transport capacity, hunters seek to maximize the amount of edible tissue, measured by weight or caloric content, moved from kill/butchering sites to base camps (Binford, 1978; Metcalfe &Jones, 1988). According to this argument, items consumed or abandoned in the field should be low ranked relative to other body parts in terms of the absolute weight or caloric content of attached edible tissue. A second, closely related hypothesis is that hunters seek to maximize the amount of edible tissue moved relative to the costs of field processing and transport (O’Connell et al., 1988). Since transport has a cost, one might expect hunters to strip edible tissue from bone and abandon the latter in the field. All else being equal, the probability of abandonment should be correlated with bone weight: the heavier the bone, the more likely it is to be stripped, cracked for marrow if it contains any, and left at the butchering site. On the other hand, since field processing often takes place under time constraints, one might expect, all else being equal, that the probability of abandonment would vary with processing time requirements: bones more easily stripped of meat and marrow would be more often left in the field. The third hypothesis is quite different. It suggests that hunters or other members of the transport party take advantage of the relatively low level of competition at the butchering site by eating the best parts (i.e. those which produce the highest nutritional return relative to processing costs) there, and moving only the remaining lower ranked parts back to base (O’Connell et al, 1988; Speth, 1988). Data pertinent to the first hypothesis are presented in Table 4, which shows the average weight of meat and marrow attached to each set of skeletal elements normally produced by Alyawara butchering. Parts from males show a broad range of values, with five fairly clear cut groups: (1) lumbar and sacral vertebrae, (2) forequarters and rear legs, (3) internal organs, intestines, and belly sheet, (4) innominates, caudal vertebrae, thoracic vertebrae, and skulls and cervical vertebrae and (5) rear feet. Parts from females show the same number of groups across a narrower range: (1) lumbar and sacral vertebrae, (2) rear legs, (3) internal organs and intestines, (4) forequarters, belly sheet, thoracic vertebrae, innominate, skull and cervical vertebrae, and caudal vertebrae and (5) rear feet. Apart from the fact that females have much less edible tissue per skeletal element set, the principal difference between the rankings is that forequarters are much bulkier and more heavily fleshed in males than females relative to most other elements, and thus higher ranked. Parts consistently selected for consumption or abandonment at field butchering stations are feet, tails, and intestines. Feet are the lowest ranked items for both males and females; tails are the second lowest ranked for females, and the fourth lowest for males.

J. F. O’CONNELL AND B. MARSHALL

400

Table 3. Weights and volumes of red kangaroo body parts. M= weight of meat, brain andsinus contents. B= weight ojbone, including marrow. Weights jorjemale 3 are for meat, marrow, and bone combined. Note that for female 2, cervical vertebrae were inadvertently combined with thoracic in disarticulation. NR=not recorded. Under basic units, each forequarter includes ribs, half the sternum, scapula, humerus, radius, ulna, carpals, metacarpals, and phalanges; each rear leg includes femur, tibia, and jibula; each rear foot includes tarsals, metatarsals, andphalanges

Males

Females

1

Basic

units (wt.

1

2

2

3

M

B

M

B

M

B

M

B

886 1635 6166 1545 1600 2100 4487 4335

615 1120 1347 496 1071

930 826 7289 1550 1325 2000 5076 4349

820 821 1282 382 1054

625 580 3620 639 325 700 841 1810

542 367 585

323 488 293

330 705 3745 568 404 710 734 2021 -

188 475 594 156 405 355 483 263

in g)

Skull &cervical vertebrae Thoracic vertebrae Lumbar & sacral vertebrae Innominate Caudal vertebrae Belly sheet Forequarter Rear leg Rear foot Internal organs Intestines

-

1465 1252 647

-

1500 1072 506

-

150 382

-

2800 2565

-

3259 2449

-

1250 1445

-

1600 1448

-

Lumbar vertebrae & sacrum Distal sacral vertebrae

4256 1910

511 830

4509 2780

780 502

1990 1630

330 225

2011 1734

264 330

Ribs & sternum Scapula Humerus Radius & ulna Carpals, metacarpals, & phalanges

1999 960 1051 411

533 149 305 282

2472 936 1182 485

579 147 309 295

562 115 118 46

171 21 39 42

481

194

99 124 30

26 41 47

-

196

-

171

-

-

47

Femur Tibia & fibula

3090 1245

540 112

3169

485

1180

587

1256 554

210 278

1650 371

200 283

-

315 332

-

-

118 175

-

113

Unit breakdowns

(wt.

cavity

Humerus Radius Ulna Femur Tibia Metatarsals

volumes

NR NR NR

in g)

Calcaneum & metatarsals Phalanges Marrow

1000 1000 4400 750 900 800 1300 2650

221 285

50

150

(ml) 15 12

15 12

8

11

26 56 4

36 48 4

5

1 1 19 24 4

5 3 3 19 30 5

Intestines, however, are much higher ranked for both sexes (fifth for males, third for females). Field consumption and/or abandonment of feet and tails is consistent with this first hypothesis, but that of intestines is not, especially since carcasses were almost always fully butchered at field cooking sites, making parts lower ranked than intestines potentially available for consumption. One explanation for the exception may be that intestines are more likely to be spoiled or fly infested than other parts of the carcass, and mu$ therefore be eaten fairly quickly. We favor this explanation, but note that it does not

ALYAWARA

BONE TRANSPORT

401

Table 4. Weight (g) of meat and marrow attached to sets of kangaroo skeletal elements normally produced by Alyawara butchering. Figures are the average of valuesfor individuals of each sex presented in Table 3. Weight of marrow is estimated at 1.2 g ml-‘. Figures for female skull and cervical vertebrae and thoracic vertebrae are those for individual 1 only (see note Table 3) Males Lumbar & sacral vertebrae Forequarter (ea) Rear leg (ea) Internal organs Intestines Belly sheet Innominate (ea) Caudal vertebrae Thoracic vertebrae Skull &cervical vertebrae Rear foot (ea)

6728 4826 4462 3030 2507 2050 1548 1463 1230 908 5

Females 3683 799 2169 1425 1446 705 604 364 580 625 5

account for the transport of at least some intestines in some instances. Occasional abandonment of feet without processing and consumption should not be surprising, since these elements have very little edible tissue attached. As indicated above, such abandonment occurred only in those instances when the total number of animals taken was relatively large. The second hypothesis is that under conditions of limited transport capacity, hunters may discard inedible elements (mainly bone) in the field in order to move the maximum amount of edible tissue possible back to their residential base. Since field processing often takes place under time constraints, the choice of bones to be discarded may be affected by differences in processing costs as well as by the weight or bulk of the bones themselves. Given these conditions, we might expect hunters to field process those body parts which yield maximum savings in transport costs (measured here as weight of bone) per unit of time spent stripping flesh and cracking bones for marrow. Data pertinent to evaluating this expectation are presented in Table 5. On this criterion, the bones most likely to be discarded in the field for both sexes should be feet and tails. Defleshing other elements yields lower savings in transport costs per unit time expended. Alyawara behavior is consistent with this hypothesis, again with the reservation. that intestines may be consumed in the field for other reasons. In spite of these results, we reject this second hypothesis because the savings in transport costs achieved by processing feet and tails in the field are relatively small, only about 1.5 kg of bones on male kangaroo carcasses, about half that on female carcasses. These savings cannot be significant for hunters operating in motor vehicles, who nevertheless consume and/or discard in the field about half the tails and feet they take. Most of the feet they abandoned in this sample (40 of 46) were not even processed. Still, it is interesting to recall that feet and tails are discarded even more often by hunters on foot (Table 2), whose transport capacity is at least somewhat limited relative to those in motor vehicles. The third hypothesis is that hunters and other members of the transport party take advantage of the relatively low level of competition at the butchering site by eating the best parts (those which produce the greatest nutritional benefit per unit time invested in processing) there, and moving only the remaining lower ranked parts back to base. Data pertinent to this hypothesis are presented in Table 6. For both males and females, tails and feet, the elements most often consumed and/or abandoned, are of low or intermediate

402

J. F. O’CONNELL AND B. MARSHALL Table 5. Bone weights and processing times for red kangaroo body parts. Bone wt. data are from Table 3. Processing time a is the time required to strip raw meat from body partsfor males andfemales, respectively. Processing time b is the estimated time required to break bones and remove marrow from skeletal elements in each set. This number is taken to be a constant 3 min per bone (cf Jones & Metcalfe, 1988). Data on processing times are not available for skulls, which were damaged in carcass acquisition

Processing time (min) Bone wt. (g)

a

b

Wt. saved (g min-‘)

0

3 0 0 6 0 9 0 0 0 0 NA

192 106 97 13 55 25 23 0 0 0 NA

3 0 6 0 0 0 9 0 0 0 NA

93 39 37 28 20 19 6 0 0 0 NA

Males (ave) Rear foot (ea) Caudal vertebrae Thoracic vertebrae Rear leg (ea) Innominate (ea) Forequarter (ea) Lumbar & sacral vertebrae Belly sheet Internal organs Intestines Skull &cervical vertebrae

1063 971 1162 439 1483 1315 0 0 0 718

10 10 10 8 50 56 0 0 0 NA

Females (ave) Rear foot (ea) Caudal vertebrae Rear leg (ea) Thoracic vertebrae Lumbar & sacral vertebrae Innominate (ea) Forequarter (ea) Belly sheet Internal organs Intestines Skull & cervical vertebrae

278 394 486 367 590 153 339 0 0 0 542

10 I 13 30 8 45 0 0 0 NA

511

0

rank. The highest ranked elements, rear legs, are never eaten in the field and are always brought back to camp. Alyawara field consumption patterns are clearly inconsistent with this third hypothesis. Another version of this hypothesis has recently been offered by Speth (1988) who suggests that hunters may preferentially consume relatively fat-rich parts of the carcass at the kill or butchering site, again in order to avoid competition for these parts at the residential base. During the time these data were collected, the Alyawara consistently displayed a strong desire for animal fat, as evidenced by their frequent comments on the nutritional condition of native fauna (especially red kangaroo), their preferential selection of fat-rich parts from the carcasses of domestic livestock occasionally made available to them by local European pastoralists, and even by the care with which they removed and consumed even the smallest bits of congealed fat found in tinned meats purchased at European stores. Red kangaroo are comparatively lean (less than 1% dissectible fat by weight), but when in good condition (as they probably were throughout the study period as a function of exceptionally high local rainfall), they contain small amounts of fat around the kidneys, on the lateral aspect of the thigh, and on the ventral surface of the tail (especially near the

ALYAWARA

BONE TRANSPORT

403

Table 6. Weight of meat and marrow available from various body parts relative to processing time. Data on meat and marrow weights from Table 4, processing times from Table 5. Data on skull and cervical vertebrae not available

Meat and marrow(g)

Processing time (mm) a b

Return rate (g min-‘)

Males (ave) Rear leg (ea) Innominate (ea) Caudal vertebrae Thoracic vertebrae Lumbar & sacral vertebrae Forequarter (ea) Rear foot (ea) Skull & cervical vertebrae

4462 1548 1463 1230 6728 4826 5 908

10 8 10 10 56 50 0 NA

6 0 0 0 0 9 3 NA

219 194 146 123 120 82 2 NA

Females (ave) Rear leg (ea) Lumbar & sacral vertebrae Innominate (ea) Thoracic vertebrae Caudal vertebrae Forequarter (ea) Rear foot (ea) Skull &cervical vertebrae

1962 3683 604 580 364 799 5 625

7 30 8 13 10 45 0 NA

6 0 0 0 0 9 3 NA

151 123 76 45 36 15 2 NA

proximal end), as well as in their bone marrow (Hopwood, 1976,198 1, pers. comm.). With the exception of the tail and the metatarsals (which have very small marrow cavities, Table 3), none of these parts were consumed at butchering stations by Alyawara hunters. As noted above, metatarsals were far more often than not left in the field uneaten; and in at least one instance, hunters who ate part of a tail at a butchering site consumed the distal rather than the comparatively fat-rich proximal half. Alyawara field consumption practices are thus also inconsistent with Speth’s version of the third hypothesis. Conclusion

On the basis of this analysis, we conclude that the data on Alyawara bone transport are consistent with the proposition that hunters seek to move parts with the greatest amounts of edible tissue from kill/butchering sites to base camps. Parts consumed or abandoned in the field have only small amounts of edible tissue attached, or are likely to deteriorate quickly. Although the data are also consistent with the idea that field processing and transport costs determine element transport patterns, we think it unlikely that this is true in the Alyawara case, simply because the Alyawara can generally carry all parts of their prey without difficulty (especially on hunts in motor vehicles), and because the savings gained by abandoning tails and feet are very small. Field processing and transport costs are more likely to be important in cases where prey size exceeds transport capacity by a greater margin than it does in the Alyawara case (e.g. Binford, 1978; Brooks & Yellen, 1987: 80; Bunn & Kroll, 1988: 145; O’Connell et al., 1988). Finally, the Alyawara data are clearly inconsistent with a third proposition, that hunters or other members of the transport party eat the best parts in the field and take only the remaining lower ranked items back to camp. We might expect such behavior only in situations where individuals whom the hunters are attempting to provision are actually members of the transport party (e.g. O’Connell et al., 1988).

404

J. F. O’CONNELL AND B. MARSHALL

Acknowledgements We thank A. Howard (Director, New South Wales Parks and Wildlife, Broken Hill) for authorization to collect the necessary carcasses, J. Nicholls (Pooncarie) for providing them, A. Lee (Department of Agriculture, La Trobe University) for access to laboratory facilities, P. Hopwood for critical data on kangaroo anatomy, and G. Caughley, D. Grayson, J. Hall, K. Hawkes, R. L. Lyman, D. Metcalfe, A. Newsome, S. Solomon, J. Speth, and J. P. White for timely advice and discussion. We are especially grateful to F. J. Allen, R. Frank, and other members of the Department of Archaeology at La Trobe for their generous assistance at all stages of the project, and to B. Bancroft for essential financial support. The manuscript was prepared by D. Gillett. References Binford, L. R. (1978). Nunamiut Ethnoarchaeology. New York: Academic Press. Binford, L. R. (1981). Bones: Ancient Men and Modern Myths. New York: Academic Press. Binford, L. R. (1983). In Pursuit ofthe Past. London: Thames and Hudson. Brooks, A. & Yellen, J. (1987). The preservation of activity areasin the archaeological record: ethnoarchaeological and archaeological work in northwest Ngamiland, Botswana. In (S. Kent, Ed.) Method and Theory for Activity Area Research: An Ethnoarchaeological Approach. New York: Columbia University Press,pp. 63-106. Bunn, H. T. (1983). Comparative analysis of modern bone assemblagesfrom a San huntergatherer camp in the Kalahari Desert, Botswana, and from a spotted hyaena den near Nairobi, Kenya. In (J. Clutton Brock & C. Grigson, Eds) Animals and Archaeology, Vol. I. Hunters and their Prey. British Archaeological Reports International Series163, pp. 143-148. Bunn, H. T., Bartram, L., & Kroll, E. (1988). Variability in bone assemblageformation from Hadza hunting, scavenging, and carcassprocessing. Journal of Anthropological Archaeology 7, 412457. Bunn, H. T., & Kroll, E. (1986). Systematic butchery by Plio/Pleistocene hominids at Olduvai Gorge, Tanzania. Current Anthropology 27,43 l-452. Bunn, H. T., & Kroll, E. (1988). Reply to L. R. Binford: fact and fiction about the Zinjanthropus floor: data, arguments, and interpretations. Current Anthropology 29, 135-147. Denham, W. W. (1975). Population properties of physical groups among the Alyawara tribe of central Australia. Archaeology and Physical Anthropology in Oceania 10,114-l 51. Gifford, D. P. (1977). Observations of modern human settlements as an aid to archaeological interpretation. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Berkeley. Gould, R. (1967). Notes on the hunting, butchering, and sharing of game among the Ngatatjara and their neighbors in the West Australian Desert. Kroeber Anthropological Society Papers 36, 41-66. Grayson, D. G. (1988). Danger Cave, Last Supper Cave, and Hanging Rock Shelter: the faunas. Anthropological Papers of the American Museum of Natural History 66, l-l 30. Hopwood, P. R. (1976). The quantitative anatomy ofthe kangaroo. Unpublished Ph.D. dissertation, University of Sydney. Hopwood, P. R. (1981). Carcassmuscle weight distribution and yield: a comparison between grey kangaroos, Macropus giganteus, and red kangaroos, M. rufus. Australian Wildlife Research 8, 263-268. Hopwood, P. R., Hilmi, M., & Butterfield, R. M. (1976). A comparative study of the carcass composition of kangaroos and sheep. Australian Journal of Zoology 24, 16. Isaac, G. Ll. (1978). The food sharing behavior of proto-human hominids. Scientific American 238,9C-108. Jones, K. T. & Metcalfe, D. (1988). Bare bones archaeology: marrow indices and efficiency. Journal of Archaeological Science 15,415-423. Metcalfe, D. &Jones, K. T. (1988). A reconsideration of animal body part utility indices. American

Antiquity

53,486504.

O’Connell, J. F. (1977). Ethnoarchaeology of the Alyawara: a report: Newsletter of the Australian Institute of Aboriginal

Studies 7,47-49.

ALYAWARA

BONE

TRANSPORT

405

O’Connell, J. F. & Hawkes, K. (1984). Food choice and foraging sitesamong the Alyawara. Journal of Anthropological

Research 40,504-535.

O’Connell, J. F., Hawkes, K., & Blurton Jones, N. (1988). Hadza hunting, butchering, and bone transport and their archaeological implications. Journal of Anthropological Research 44(2), 113-161. Perkins, D. & Daly, P. (1968). A hunters’ village in neolithic Turkey. ScientiJic American 219(1 l), 97-106. Solomon, S. (1985). People and other aggravations: taphonomic research in Australia. Unpublished B.A. Honours thesis, Department of Prehistory and Archaeology, University of New England, Armidale, New South Wales. Speth, J. (1988) Seasonality, resource stress,and food sharing in “so-called” egalitarian societies. American Anthropologist (In press). Tribe, D. E. &Peel, L. (1963). Body composition of the kangaroo (Macropus sp.). Australian Journal of Zoology

11,273-289.

White, T. E. (1952). Observations on the butchering technique of some aboriginal peoples: I. American

Antiquity

17,337-338.

Yellen, J. (1977). Archaeological

Approaches

to the Present. New York: Academic Press.