Volume 43 1934 > Volume 43, No. 170 > Australian spear-traits and their derivations, by D. S. Davidson, p 41-72
                                                                                             Previous | Next   

- 41

THE spear has long been recognized as the most important weapon of the Australians and the Tasmanians. Other missile and hand-weapons, such as throwing-sticks and clubs, are also universally employed but, in most cases, they are supplementary to the use of spears in both hunting and fighting.

In treating of the spear in any part of the world we are concerned with a very old culture-trait. Just how long spears have been used by man cannot be indicated with any certainty. We may be quite sure, however, that they antedate Homo Sapiens, if we may judge from what are commonly accepted as stone spear-heads in the Mousterian deposits of middle-Palælithic antiquity. It seems likely that plain wooden spears were known long before stone-heads were affixed to shafts, but no specimen has been preserved to show such a chronology. Since we cannot expect wooden objects to withstand the disintegrating influences of tens of thousands of years, except under most unusual conditions, it is possible that the time of the origin of the spear will never become accurately known. The most that we can say at present is that an indefinite period of time extending back into the lower-Palæolithic Age seems to be involved.

In respect to the derivation of the varieties of spears and the minor traits associated with them found in Australia and Tasmania, however, our problem is neither hopeless nor indefinite. There is every reason to believe that spears - 42 have been in the possession of the aborigines since the time they first arrived on the continent. This conclusion is suggested not only by the lack of evidence of a pre-Homo Sapiens population there and by the appearances of great antiquity of the spear in all parts of the world but also by the universal use of this weapon in Australia and Tasmania. As a matter of fact there is an abundance of information which, as I hope to show, demonstrates quite convincingly the derivations of the various spear-traits now found in Australia, as well as the chronology of their appearance on that continent.

In most parts of the world there has been a tendency for the spear to give way to the bow and arrow or other weapons. There are notable exceptions to this statement but, generally speaking, it holds true for Europe, Asia and the Americas and, in part, for Africa. Australia is the only continent where the spear has not felt the competition of these other weapons. How much longer such a condition would have continued is a moot question, for the bow is found in Torres Strait and may even have gained a foothold at the tip of the Cape York peninsula. 2 If the agencies responsible for diffusion under aboriginal conditions had not been interfered with by the arrival of European institutions it is possible that the bow might now be in the process of conquering the last major region of the world.

It seems evident, therefore, that in dealing with the history of the spear in Australia we are concerned with a weapon which has enjoyed an uninterrupted existence on that continent for many thousands of years, an existence undisturbed by other important missile weapons. We thus find in Australia an excellent setting for a study of one of the most important weapons in the history of the human race.

Australian spears have been described by a number of authors, including Brough Smyth, Roth, Basedow, Spencer and Gillen, Tindale, Love, and Hambly, and in some instances attempts at classification have been made. In all - 43 cases, however, these writers have been concerned primarily with a particular part of Australia and none of them has included all of the varieties found on the continent. On the basis of their descriptions and a study of the specimens of the various museums of Australia, however, it is now possible to consider the matter of classification from a continental point of view. There are still many relatively small regions from which we have no detailed information. Consequently, our data are by no means complete. Subsequent investigation, therefore, may bring to light some varieties of spears not known at present and may necessitate an extension of the boundaries of the different types we are to discuss. In view of the fact that our information as a whole contains only these small gaps, it is not to be expected that startling changes in distribution will be required.

The most meticulous classification of Australian spears is that made by Spencer in which he gives 19 types. 3 Although such a classification is an accurate one, in so far as it goes, it is nevertheless unsatisfactory in that it sets up as separate types spears which have the same fundamental characteristics but which differ in one or more details only. For instance, Spencer classes as one type a spear with barbs cut in the solid in a single series and as another type a spear with the same general features but with the same kind of barbs in two series. To my mind it is more reasonable to consider them as varieties rather than as types. If all the spears of the continent were differentiated on the basis of minor details we would have an infinite number of types. It seems quite clear that such a classification is ill suited to the purpose of segregating the several traits which in their various arrangements and combinations have given rise to the differences we find in Australian spears.

It also would be possible to classify spears on the basis of their means of propulsion, whether thrown by hand, or by use of a spearthrower. It seems obvious, however, that such a classification is meaningless from the point of view of the historical development of spear-types. Aside from the indentation in the butt to receive the point of the - 44 spearthrower, there may be no differences between hand-spears and those used with spearthrowers. The oldest type of spear, undoubtedly the plain unbarbed one-piece spear, can be equipped for use with a spearthrower as readily as those types of spear with the latest variety of barbs or stone heads and, vice versa, the latter, without the hole in the butt, might be said technically to belong to a pre-spearthrower age. Modifications and changes in spear-characteristics can be applied readily to various spears regardless of their intended means of propulsion and, for that reason, the boundaries of the use of spearthrowers in themselves do not constitute barriers against the diffusion of other spear-traits.

The presence or absence of the spearthrower in any region, therefore, is important only from the viewpoint of the history of spear-propulsion. It is not important from the point of view of the history of spear-types. Since there are no spear-types which could not be thrown by either method, it seems obvious that means of propulsion is not a criterion of classification of type.

The use of the spearthrower in Australia has usually been given so much attention by travellers and investigators that sight has been lost of the importance of the hand-spear or thrusting spear (see map, fig. 1). Although it is to be admitted that the Australians, generally speaking, are characterized more by the use of the spearthrower than by the hand-spear, it is nevertheless true that the latter is just as widely distributed as the former, indeed, more so, for it is found in several localities in which the spear-thrower is little used or even entirely lacking. Throughout a great part of New South Wales the spearthrower seems to be absent. In this region, however, the types of spears, on the basis of the little we know of them, seem to be similar to those in the adjacent regions where the spear-thrower is used. The lack of the spearthrower in this area, therefore, is a puzzle for which there seems to be no satisfactory explanation at the present time. Most of the tribes who lived in this area have been extinct for some time and it is probable that very little more information can be secured.

Although it is hardly valid to base an opinion upon a quantitative study of museum collections unless it is

- 45
1. Known Distribution of Hand Spears., HAND-SPEARS., Western Australia. W.A.M. (Western Australian Museum)—Ashburton, Roebourne, Derby, Kimberley, Lake Way, West Kimberley, Broome, Pender Bay, Drysdale, Beagle Bay, Prince Regent River, Isdell Range, Collier Bay, Laverton, Geraldton, Murchison, Gascoyne, Wiluna, York. S.A.M. (South Australian Museum)—Lyons River, Eucla, Gascoyne. U.P. (Museum of the University of Pennsylvania)—Byro Station, Paynesville, Nannup, Lawlers, Southern Cross. Hambly-Murchison, Gascoyne, Ashburton. Clement, Pls. 2-3, figures what probably are hand-spears from Sherlock River, Hammersly Range, Desert Region and Nullagine., South Australia. S.A.M.—Murray River, Fowler's Bay, Kopperamanna, Goolwa, Lake Eyre, West Coast of S A., Algehuckina, Yardea, S.A.-Queensland border, Minanicka, Upper Murray River, Yorke Peninsula, Mount Gambier, Birdsville, Coopers Creek, Adelaide, Mount Burrell, East Hamilton, Gawler Range, Dieri, Warrego, Port Lincoln to Fowler's Bay. Hale and Tindale, pp 47-48, Wailpi, Flinders Range. Cawthorne, p. 4, Adelaide Tribe. Schurmann. p. 213, Port Lincoln (fish-spear). Taplin, pp. 39-40, Narrinyerri, Kukatha, Port Lincoln to Fowler's Bay. Meyer, p. 193, Encounter Bay (fish-spear). Horne and Aiston, p. 79, Wonkonguru. Bolam, p. 94, Ooldea (unsharpened stick used by modern native boys for amusement)., North Australia. S A.M.—Port Essington, Palmerston, Croker Island, Melville Island, Bathurst Island, Katherine River, McArthur River A.M. (The Australian Museum, Sydney)—Borroloola U.P.—Katherine River. Foelsche, p. 12, Coastal country., Central Australia. S.A.M.—Musgrave Range, Macdonnell Range. Alice Springs, Barrow Creek, Powell's Creek, West of Alice Springs., Queensland. S.A.M.—Diamantina, Boulia, Cairns, Herbert River, Port Stewart, Clermont. A.M.—Boulia, Gregory District, Clonagh, Cloncurry, Kalkadoon, Burketown. Roth, 1897, p. 146, Boulia, Leichhardt-Selwyn, Cloncurry, Flinders District, Upper Diamantina; p. 147, Mitakoodi, Pitta Pitta; Bulletin 13, p. 195, Tully River, Rockhampton, Upper Fitzroy River; p. 196, Westwood (Rockhampton), Maryborough, Brisbane., Victoria. S.A.M. A.M.—Lake Victoria. N.M.V. (National Museum of Victoria)., New South Wales. A.M.—Western New South Wales, Edward River, Murrumbridgee River, Lower Darling River, Lachlan River, Dubbo, Mount Poole, Murray-Darling District, North-central New South Wales, Angledool, Milparinka. Flanagan, p. 79. Peron, plate 22, no. 3., Tasmania. Tasmanian Museum; N.M.V.—Peron, plate 13, no. 1.
- 46

definitely known that the selection of specimens in the field has been made in a relatively true proportion of their occurrence, it is nevertheless interesting to note that apparently a very large proportion of the spears from the Gascoyne-Ashburton region of north-western Australia belong to the hand-type. It seems more than just a coincidence that such a proportion is found in a peripheral area when other peripheral regions lack the spearthrower completely.

The two most important negative areas for the spear-thrower are Tasmania, and Melville and Bathurst Islands, in both of which regions hand-spears are exclusively used. In Melville Island the spearthrower is said to be known but it is regarded as a toy for boys. This condition should not be interpreted as indicating that the spearthrower was formerly an important adult characteristic. In view of the peripheral location of Melville Island it seems more probable that the unimportance of the spearthrower is the result of the introduction of a trait which either did not impress the Islanders or was found by them difficult to apply to the spears in use. Since there seem to be no natural reasons why the diffusion of the spearthrower should be hindered in diffusing into this locality we should look to cultural forces for an explanation of this anomaly.

Generally speaking it is difficult to hurl heavy spears with a spearthrower, and it is not surprising, therefore, to find them usually thrown by hand. There is no definite rule in this matter, for light spears are also sometimes used in the hand; for example, the fishing-spear of Central Australia. As we have said, however, in most cases very heavy spears are not thrown with the aid of a spearthrower and, as a result, we should not expect the spearthrower to be important in areas where heavy spears prevail. It also follows that spearthrowers will not be readily introduced into a region where heavy spears are found alone unless the latter are modified in weight or unless new lighter-weight spears diffuse along with the spearthrowers. Such an explanation seems to be suited to the conditions on Melville Island where the natives prefer relatively heavy spears. The situations in New South Wales and northwestern Western Australia, where heavy spears, among others, are also found, may be analagous in a general way.

- 47

Historically speaking, the hand-spear must have preceded the use of the spearthrower, and this chronology in the Pacific is demonstrated, apparently, by the lack of the spearthrower in Tasmania and Melville Island and, seemingly, by the relatively little use of this device in north-western Western Australia. The situation in New South Wales, as already mentioned, is rather difficult to explain. This is a region of which we know little and, as a result, it is impossible to take into consideration many factors which, if known, might aid in the explanation of the case.

A third basis for classifying spears might be their use or function. In Australia spears are employed for fighting individual combats, for war, for hunting, for fishing, and for ceremonies, and very often in certain tribes different spears are associated with one activity but not with another. It seems obvious upon a moment's consideration, however, that such a classification would be of little value except for each tribe, or possibly group of tribes, taken individually, for some tribes use the same spear for two or more activities whereas other tribes have different spears to serve the same function. 4 Since there are many varieties of spears, some with great distributions, some with small distributions, a number of tribes or regions are found to possess a relatively great number of them, whereas others have only a few. As a result of such a condition, specialization of function can be carried further in one area than in another. A tribe with a great variety of spears can easily allocate them to the activities for which they feel each is best suited, but the tribe with only one variety of spear must use it for all occasions.

- 48

The specialized function of a spear, furthermore, is neither obvious nor discernible from its structural qualities, but depends entirely upon individual or local opinion, as determined by tradition and historical influences. A spear preferred by one tribe for hunting is considered by another as best suited to fighting. The functions of spears as a group vary but little from one end of the continent to the other, but within this distribution we find a number of varieties of spears, and within each variety a wide range of weights, dimensions, and details, as the answers to these apparently similar needs.


By this time it should be evident that if we are to arrive at any understanding of the history of development of the different varieties of spears in Australia it will be necessary to confine our attention to the fundamental physical traits of spears as they are found singly in their respective distributions rather than in their combined forms with various other traits.

It seems clear that except for the indentation in the butt for the point of a spearthrower there need be no difference in the structure of spears for thrusting, for throwing by hand, or for throwing with the aid of a spear-thrower. In so far as means of use are concerned, therefore, we may disregard for the most part the presence or absence of an indentation at the butt.

It also seems evident that function has no direct relationship with the modern distribution of the different varieties of spears. This should not be implied to mean that different spear-types may not have been directly - 49 associated with special functions at the time of their origins, but that is quite a different problem.

Structurally a spear consists of a head and shaft which in the most simple form are in one piece. Presumably this is the oldest type of spear from which all others have directly or indirectly developed. In the composite spears the head and shaft are separate, and joined by various means of splicing, gumming, or tying. In the more complex spears there may be a head, an after-head or fore-shaft, the shaft, and an after-shaft or tail-piece. In addition to these fundamental traits there are such minor peculiarities as the shape of the head; the number of prongs; the various forms of barbs cut in the solid, as they appear singly, or in one or more series; reversed barbs; “closed” barbs; detachable barbs, singly, or in series, of bone or of round or flat wood; stone heads of different types; small stone chips affixed in gum to the head; and so forth.

Since any of the above traits can be applied to any “type” of spear, whether it is well made or poorly made, long or short, light or heavy, of one piece or more than one, it seems evident that all of them, except those which by definition exclude others, could be found together in any one spear. For this reason any and all of these traits could have diffused to all regions of the continent. That all of them are not ubiquitous, however, is a condition which we should expect for many reasons. In the first place all of these traits cannot be of equal age, and for that reason alone we should expect some to have greater distributions than others. Secondly, local methods of spear-manufacture undoubtedly have been influenced by tradition, with the result that some tribes may be conservative and less subject to accepting new ideas, whereas others may readily adopt the innovations of their neighbours. Thirdly, willingness or unwillingness to change must not be regarded as a staple condition; for a tribe which may be averse to adopting one spear-trait, may at the same time readily accept another spear-trait, and a group which may resist an influence at one time in its history may welcome the same trait, or a different one, at a later time. As with all other traits of culture we should expect as a normal condition to find considerable variation in the distributions of the various spear-traits. - 50 By careful analysis of these, however, we may hope to perceive some clues to the history of spears in Australia. Furthermore, with the possibility of infinite variation in spear-trait combinations it will be interesting to see how ingenious man has been in adapting material objects to suit his fancied needs.


The most simple type of spear found in Australia and Tasmania is the plain spear made from one piece of wood and having only a simple point, round or oval in cross-section. The plain spear is very well known for Tasmania, where it is the only type present, but its appearance in Australia has never been given much attention, probably because of the more complex types which prevail in most regions. However, as the map shows (fig. 2), plain spears

FIG. 2.
Hatching—Appearances of plain one-piece spears., Broken line—Limit of two-piece wooden spears., Dotted line—Limit of two-piece reed spears., Solid line—Limit of three-piece wooden spears.
- 51
ACICULAR, PLAIN, ONE-PIECE SPEARS, NOT FOR USE WITH SPEARTHROWERS.,Western Australia. S.A.M.; W.A.M.—Eucla., South Australia. S.A.M.—Coopers Creek, Kopperamanna, Murray River, West Coast of S.A., Mount Gambier, Goolwa, Algebuckina, Yardea. Lake Eyre. Taplin, pp. 39-40, Narrinyerri. Schurmann, p. 213, Port Lincoln., North Australia. S.A.M.—Croker Island, Port Essington, Katherine River, Melville Island, Bathurst Island., Central Australia. S.A.M.—Macdonnell Range, 120 miles west of Alice Springs, Barrow Creek., Queensland. S.A.M.—Diamantina, Boulia, Cairns, York Peninsula, Birdsville, Port Stewart, Warrego, Herbert River, Clermont. Roth, 1897. p. 146, Boulia, Leichhardt-Selwyn, Cloncurry, Flinders District. Upper Diamantina; Bulletin 13, p. 195, Rockhampton, Upper Fitzroy, Tully River, p. 196, Brisbane., New South Wales. Peron, plate 22., Victoria. Brough Smyth, 1, p. 307., Tasmania. Tasmanian Museum. Peron. plate 13. There seems to be no reason for believing that any other spear type was known in aboriginal days although claims made otherwise for historic times may be justified., ACICULAR, PLAIN, ONE-PIECE SPEARS, FOR USE WITH SPEARTHROWERS., Western Australia. W.A.M.—Esperance, Geraldton. S.A.M.—King George Sound, Geraldton. Swan River. N.M.V.—Chianga, Whajuk, Ballardong, Kadagur, Wonundar-mining tribes. U.P—Nannup. Stokes, 1, p. 176, King George Sound (Bathurst Island). Carnegie, P. 339, Desert Tribes. Love, p. 25. Wororra (light for fish; heavy for turtles and dugong—probably not used with spearthrowers)., South Australia. S.A.M.—Yorke Peninsula, Eyre Peninsula. Fowler's Bay. Gawler Range, Port Lincoln. Howitt. p. 761, Yerkla-mining. Taplin. p. 95, Kukatha tribe, Port Lincoln, Fowler's Bay (implied). Schurmann, p. 3, Port Lincoln. Bolam, p. 83. Ooldea (for rabbits), North Australia. S.A.M.—Groote Island. Wardaman (Fieldnotes)., Central Australia. N.M.V.—Arunta., Victoria. Brough Smyth, 1, p. 307., New South Wales. A.M.—Lake Victoria, Murray-Darling, Mount Poole, Milparinka, Queensland. S.A.M.—Diamantina., ONE-PIECE ACICULAR SPEARS WITH GUM BANDS OR RINGS., Western Australia. U.P.—Nannup, Albany, Esperance. W.A.M.—Kookynie, Bunbury, Northampton, Ashburton, Gascoyne, Williams., Queensland. A.M.—Normanton., PLAIN SPEAR WITH OVAL OR FLATTENED POINT (FOR SPEARTHROWERS). Western Australia. W A.M.—Wiluna. Geraldton, Ashburton, Kimberley, Esperance, Kookynie, Pingelly, Bunbury, Greenough, Israelite Bay, Lake Way, Murchison, Mount Magnet, Gascoyne, Kanowna, Boogardie, Laverton (with diamond shape cross-section, Lake Way, Kanowna, Geraldton, Pingelly. Kookynie). U.P.—Peak Hill, Gwalia. Meekathara, Mount Magnet, Paynesville, Sandstone. Carnegie, p. 339, Desert tribes., Central Australia. N.M.V.—Arunta, Barrow Creek. S A.M.—Alice Springs. Arunta, Luritia (hand-spears), Barrow Creek., South Australia. S.A.M.—Fowler's Bay, Eyre Peninsula, Port Lincoln., New South Wales. A.M.—Murray-Darling District., PLAIN ONE-PIECE SPEARS WITH SPATULATE HEADS (FOR SPEARTHROWERS)., Western Australia. S.A.M.—Eucla, Gascoyne, Ashburton. Carnegie, p. 339, Desert Tribes., South Australia. S.A.M.—Gawler Range, Yardea, West Coast of S.A., Central Australia. S.A.M.—Barrow Creek, Macdonnell Range., PLAIN ONE-PIECE SPEARS WITH SPATULATE HEADS, NOT FOR USE WITH SPEARTHROWERS., Western Australia. W.A M.—Laverton., South Australia. S.A.M.—Algebuckina, Lake Eyre, Gawler Range, Dieri, Wallara, Minanicka, Upper Murray, Kopperamanna, West Coast of S.A. Hale and Tindale, pp. 47-48, Flinders Range., Central Australia. S.A M.—Arunta, Luritja, Barrow Creek. West of Alice Springs, Macdonnell Range. Spencer and Gillen, 1904, p. 671., North Australia. S.A.M—Powell's Creek. A.M.—Borroloola.
- 52
New South Wales. S.A.M.—Upper Murray River., Queensland. A.M.—Cloncurry, Rankin River, Boulia, Gregory District, Clonagh. S.A.M.—Diamantina. Roth, 1897, p. 147, Boulia, Mitakoodi, Pitta Pitta; Bulletin 13, p. 195, Tully River., Tasmania. A M., TWO-PIECE SPEAR WITH PLAIN SPATULATE HEAD AND WOODEN OR BAMBOO SHAFT, FOR USE WITH SPEARTHROWERS., Western Australia. S.A.M.—Gascoyne-Ashburton District., Central Australia. S.A.M.—Macdonnell Range, Alice Springs, Mount Burrell. Helms, plate XVI, no. 4, Everard Range., North Australia. S.A.M.—Daly River.

are widely distributed, and it would not be surprising to find them in occasional use everywhere.

It is in only a small part of the continent, the southern and western coastal districts, that plain spears may be said to be at all frequent. In these regions they are extremely crude (fig. 3). Very little attention seems to have been

FIG. 3.
Crude Unbarbed Spears, South Australia., 1. Yorke Peninsula; 2. Eyre Peninsula; 3. Dieri tribe; 4 and 6. Port Lincoln; 5. Fowler's Bay. (South Australian Museum.)

given to their manufacture and, as a rule, they are neither straight nor refined by polishing. In some areas such a condition is undoubtedly the result of the necessary use of mallee, but in others there seems to be no excuse for the selection of poorly-suited material. In most cases these spears are thrown with the aid of a spear-thrower. The Tasmanian spears, as a class, are much better made and are usually longer and straighter. They are all for use by hand, for the spearthrower, as already stated, was unknown in Tasmania.

In the central portions of Australia excellent plain spears are not infrequently found. In this region they seem to be of specialized ceremonial use and are usually spoken of as king-spears (fig. 4). They are made of very heavy

- 53
FIG. 4.
King Spears, Central Australia.

wood, and consequently are generally not associated with a spearthrower. Many have wide spatulate heads; others have acicular or slightly flattened heads.

The ceremonial function of these spears is very likely the cause of the careful workmanship they have received, but it is only in the degree of refinement that they differ from the ordinary plain spear. Since both are structurally the same there seems to be no reason for placing them in separate classifications or for regarding the king-spears historically as other than carefully made plain spears.

Well-made simple spears are also found in central and north-western Western Australia, where they seem to be used for ordinary purposes (fig. 5). For the other regions

FIG. 5.
Unbarbed Wooden Spears, Western Australia., 1. Wiluna; 2. York; 3-5. Ashburton. (Western Australian Museum.)

of Australia plain spears are not characteristic. Yet, as the map shows, there seem to be few well known regions from which some specimens of this type have not been collected.

On the basis of both distributional and technological considerations it seems safe to conclude that this simple one-piece spear is the oldest in Australia. It has the most widespread distribution, and is the only spear which appears - 54 to be present in all or most all of the peripheral regions. This type of spear, presumably, must have been among the possessions of the invading Tasmanians and possibly also the first Australians. In such an event it could not have had an Australian origin.


The composite spear, it cannot be doubted, is historically later than the simple spear, but how it developed is a matter of conjecture. Practically all of the Australian tribes who use the one-piece spear, either plain or with barbs cut in the solid, as a characteristic weapon do not make any use of the composite spear with separate wooden head and wooden shaft. This is true for Tasmania, Victoria, South-western Australia, Melville and Bathurst Islands, and certain districts of the eastern coast of Queensland. It is important to note that all are peripheral or isolated regions. It is only in Victoria that a composite spear with a wooden head and a reed shaft is also present.

In Central Australia, part of North Australia, and central and north-western Western Australia, composite spears with wooden heads and wooden shafts are in general use, and the question arises as to their historical relationship to the one-piece spear. Off hand one might expect that the repair of a broken spear, by joining the two pieces together, would occur to anyone, and that this might be the logical origin of a composite spear. Regardless of the validity of such a suggestion it is interesting to note that composite spears are generally lacking in most of the peripheral regions of the continent, as already stated. Our conclusion, therefore, must be that the peoples in these regions either found it easier to make a new spear than to repair a broken one, were so bound traditionally to spears of one piece that cultural forces discouraged repair, or that the thought of joining two broken parts together never occurred to them.

The fact that museum collections do not show composite spears from these peripheral regions, however, should not be taken as proof that repaired spears were not present in any particular region. It is quite understandable that repaired spears might have been rejected by collectors in the field in favour of perfect specimens. On the other - 55 hand, there are numerous examples of composite spears in the collections which one would have great difficulty in classifying as originally two-piece spears or as repaired one-piece spears. In spite of such difficulties in arriving at criteria for interpretation, it nevertheless remains true that one-piece spears seem to be the only ones found in most peripheral regions, as the map seems to demonstrate (fig. 2), and that composite wooden spears are grouped in a more central location. Since this compilation is based upon the study of many hundreds of spears in all the Australian museums the possibility that selection in the field may be responsible for such a distribution would seem to be discounted.

The whole problem of the origin of the composite spear in Australia, however, is complicated by the presence of two types of two-piece spears which we have mentioned in passing. The first, which consists of a wooden head and a wooden shaft, is the one we have discussed. The second, with a wooden head and a reed or bamboo shaft, must now be considered in the possible bearing it may have had on the development of the composite wooden spear.

Reed-spears in Australia, as we shall see later, are undoubtedly more recent than the simple wooden spears, and it may be that they have been the example which suggested the change of the latter to the composite spear with separate head and shaft. Reed-spears have an almost continental distribution, but are noticeably lacking throughout a large part of Western Australia and other peripheral districts and in the isolated cul-de-sacs of eastern Queensland. The distribution of the composite wooden spears falls partly within that of reed-spears but extends beyond it in the north-western part of Western Australia. In the Gascoyne-Ashburton area reed-spears seem to be unknown and here composite wooden spears are found in conjunction with one-piece wooden spears. In view of this outlying appearance we cannot be certain of the chronological relationship of the two types of composite spears which otherwise might suggest itself. Composite wooden spears may have developed locally before reed-spears diffused into the area of their distribution or they may be the results of the influences of the latter and subsequently spread alone to the north-west.

- 56

From the point of view of the historical development of composite spears, regardless of whether they consist of two pieces of wood or of wood and reed or bamboo, the evidence from their distributions would seem to furnish satisfactory proof that the simple one-piece type was the original form. This conclusion is further substantiated by technological considerations for it is only reasonable to suppose in this case that the simple type is the basis for the more complex.

The place of origin of the two types of composite spears is still another problem. As we shall see in a further consideration of reed-spears there can be no doubt but that they are of New Guinean derivation in so far as Australia is concerned. At the present time we have neither information to imply that the composite wooden spear diffused from New Guinea nor evidence to show that it developed in Australia, except its localized distribution there. Judging from this relatively small contiguous distribution, however, it seems reasonable to believe that it is of Australian origin, either the result of independent forces prior to the arrival of reed-spears, or subsequent to their diffusion if a derivation from the latter is assumed.

A different type of composite spear is the multi-pronged variety generally associated with fishing. These spears are found in Australia with two, three, or four prongs (fig. 6).

FIG. 6.
Bi-dent Spear., Groote Island (after Tindale).

Very few specimens are found in museum collections and, for that reason, we cannot discuss their distribution with any certainty. Judging from their known appearances this type seems to be lacking in Western Australia, and concentrated in the Cape York peninsula region (fig. 7). This suggests that the idea of multi-pronged spears may have come from New Guinea, although I have been unable to find reference to them in that region. They are used in other parts of the Pacific. The apparently restricted distribution in Australia, however, would seem to indicate a chronological relationship between multi-pronged spears

- 57
FIG. 7.
Distribution of Number of Prongs in Multi-pronged Spears and of Spears with Stingray-spine Heads (S)., MULTI-PRONGED SPEARS., North Australia. S.A.M.—Groote Island (2), Daly Waters (4 wire), Port Essington and Croker Island (2 long prongs cut in the solid). Spencer, 1914, plate 13, figs. 2, 3, 5, Kakadu, Geimblo, Umoriu, Iwaidji (3 and 4), Melville Island (2 cut in solid, plain and barbed). Basedow, 1907, p. 32, Larrikiya and Port Essington (2 barbed in the solid)., Queensland. S.A.M.—Batavia River (3), Cape Bedford, Cape York Peninsula and Batavia River (4). A.M.—Bentinck Island (2) with North Australian type of barbs cut in the solid. Roth, Bulletin 7, p. 194, Lower Tully River (3 and 4), said to be recent arrivals; Bulletin 13, p. 192, Pennefather River (4); p 197, Brisbane (4); p. 193, Cape Bedford, Princess Charlotte Bay, Middle Palmer and Bloomfield Rivers (4); p. 190 and 1897, p. 148, Wellesley Islands and Karunti tribe at Normanton to Upper Leichhardt (Coolullah) (3)., New South Wales. Peron, plate 22, no. 4 (4). Saturday Magazine, no. 252, p. 217 (4), all jagged on inner side., South Australia. S.A.M.—Point McLeay (2), Narrinyerri (2). Woods, p. 42 (3). Taplin, plate facing p. 32 (2)., Central Australia S.A.M.—(2 and 3). Spencer and Gillen, 1904, fig. 231a. Warramunga (3) barbed in the solid; 1899, p. 577, said to be lacking further south., Victoria. Brough Smyth, 1, p. 306, Swan Hill (2 and 3)., Torres Strait. Cambridge Expedition, 4, p. 180 (2 and 4), barbed in the solid in double rows., STINGRAY-SPINE PRONGS., North Australia. S.A.M.—Groote Island (as many as 17)., Queensland. S.A.M.—Cape Bedford (various). Roth, Bulletin 13, p. 192, Princess Charlotte Bay, Bloomfield and Middle Palmer Rivers; p 191, Pennefather River (various); p. 193, Gulf Coast; p. 194, Lower Tully River.
- 58

and ordinary spears which, technically considered, might be said to have only one prong.

A different type of multi-pronged spear and one which has a much smaller distribution is the spear with a head of stingray spines found in North Australia and Queensland (fig. 7). Very little information is available about this spear or its distribution. It does not seem to be found except in coastal regions, as one might expect for obvious reasons. In view of its restricted distribution it seems likely, at least in so far as Australia is concerned, that its origin may be relatively recent. Its distribution also suggests that a non-Australian origin is possible although information is not available to establish its presence in adjoining New Guinea.

A third type of composite spear is the three-piece spear found only in Central Australia. It consists of an ordinary composite wooden spear to which a tail-piece has been attached. The tail-piece is short and suggestive of the short shaft found in the reed-spears of southern and eastern Australia. It is quite possible that the latter has been the influence for the application of the tail-piece to the composite wooden spear. The distribution falls within those of both the wooden composite spear and the reed-spear (see fig. 2).


The reed-spear, as we have already said, is quite common in Queensland and the Cape York peninsula, Victoria and adjacent South Australia, Central Australia, North Australia and contiguous Western Australia. It consists of a sharply-pointed hard wood head inserted into a reed or bamboo shaft and fastened by thongs or gum. There seem to be two extremes between which are many varieties; one, in which the length of the head is relatively short, possibly only one quarter the length of the spear; the other, in which the proportions are reversed, the short shaft constituting the minor length of the weapon (fig. 8). In the latter the reed part is hardly more than a tail-piece and may be the influence from which the true tail-piece of Central Australia developed.

The spears with the short head appear to predominate in North Australia, whereas the short-shaft variety seems

- 59
FIG. 8.
Reed-spears., 1. Daly River, N.A.; 2. Caledon Bay, N.A.; 3. Lake Alexandrina, S.A. (South Australian Museum).
- 60

to be more numerous in Queensland and the southern part of the continent. It is impossible, however, to draw a definite line of demarcation, for considerable variation is found in all regions.

There is also a lack of unison in total length, for we find some of these spears less than four feet long, whereas others may be twelve feet or more in length. Such a wide range is seldom found in the spears of any one locality, but is noticeable immediately when specimens from a number of tribes or regions are compared. Some variation is a normal condition. In addition, allowance must be made for spears manufactured for different purposes, as well as for the differences in spear-making tradition in various regions as determined by local historical factors.

In all cases, except one, the reed-spear seems to be made to be thrown with the aid of a spearthrower. In North Australia, at least in the western part, a special type of spearthrower with a large wad of gum for the point which fits the butt of the spear, is associated exclusively with this type of spear. In the other regions the prevailing type or types of spearthrower in each area seems to be used. The only tribes known not to employ a spearthrower with the reed-spear appear to be the Arunta and their neighbours of Central Australia, who use these spears in the hand for fishing.

This instance must be regarded as a very specialized use, however, for these spears by their very nature, in their light weight and fragility, are not suited for either thrusting or for throwing by hand in war or in hunting. In fact it seems quite certain that the spearthrower must have been in use before reed-spears were invented, for these spears are the only type which would be impractical for ordinary purposes without propulsion by means of a spearthrower. It even seems likely that the existence of the spearthrower may have suggested an experimentation with reed instead of wood in order to find a lighter-weight shaft and to avoid the necessity of carving an indentation in the butt. At any rate reed-spears have a much more restricted distribution in Australia than spearthrowers and the more substantial wooden spears which, as we have seen reason to believe, appear to antedate the use of spear-throwers.

- 61

The main value of reed-spears, it seems evident, lies in their sharp point; for, in being light of weight, reliance cannot be placed upon momentum, as in a heavy projectile. For this reason small reed-spears can be deflected easily. On the other hand, these difficulties are, in part, compensated for by the increased range which the light weight makes possible, and by the large number of spears which can be carried without hardship.

The reed-spear seems to be a good example of a trait which has been combined with a number of other spear-traits in various parts of the continent. It is to be noticed that it tends to be plain in a number of areas along the periphery of its distribution, such as Victoria and adjacent South Australia, Central Australia, northern North Australia, and in the lower Murchison region of Western Australia. The appearance in the last-named locality is reported only by Hambly, and is based upon specimens in the Field Museum. In all of these regions plain reed-spears prevail. In some instances, however, combined types are also present (see map, fig. 9).

In South Australia some reed-spears have been equipped with stone chips, embedded in gum along the head, to become “death-spears.” This practice seems to be a local one for most “death-spears” in other areas are made from single pieces of wood.

An important type of spear with which reed-spears appear to be very closely allied, if they are not the foundation for it, is the Kimberley stone-headed spear (fig. 29). As a general rule the after-shaft is made of bamboo and, with the exception of the stone head, there seems to be no important difference from the long variety of reed-spear. In origin, therefore, the Kimberley spear would seem to represent the combination of the plain reed-spear and a chipped stone head. Both plain and Kimberley spears are similar in their fundamental characteristics; both seem to be best suited to throwing with the aid of a spearthrower, rather than by hand or by thrusting; and both rely upon a sharp point rather than weight for their effectiveness.

Reed-spears are obviously the basic trait for the spears found throughout a large part of eastern Australia which are equipped with detachable barbs (fig. 25).

- 62
FIG. 9., Distribution of Reed-spears.
1. Plain., 2. Plain and with Detachable Barbs., 3. Plain and in “Death” Spears., 4. Plain and with Barbs cut in the Solid., 5. Plain and with Kimberley Stone Heads., REED-SPEARS., North Australia. S.A.M.—Daly River, Port Essington, Alligator River, Katherine River, Darwin, Daly Waters. Spencer, 1914, p. 357, Kakadu. Basedow, 1907, pp. 32 et seq., North-western Coastal Tribes, Larrikiya, Wogait, Berrigin, Ginnen. U.P.—Wardaman, Ngainman., Central Australia. Spencer, 1914, p. 357, Arunta (hand spear for fish); Spencer and Gillen, 1904, p. 676, said to be common., Queensland. S A.M.—Herbert River, Hughenden, Flinders River, Coen River. Roth, Bulletin 13, p. 191, Pennefather River (for play). Embley River (heavier and longer); 1897, p. 148, North-west Central Queensland, Cloncurry (varying proportions of heads and shafts), Mullangera, Boulia (traded from Burke and Wills region). A.M.—Batavia River, Croyden, Gregory River, Normanton., South Australia. S.A.M.—Point McLeay, Murray River, Southeast, Mount Gambier. Woods, p. 40, Narrinyerri. Taplin, pp. 39-40 and plate facing p. 32, Narrinyerri. Cawthorne, p. 5, Adelaide Tribe., Victoria N.M.V.; S.A.M.—Etheridge. Brough Smyth, 1, pp. 305-306, common everywhere in Victoria., Western Australia. S.A.M.—Kimberley. Hambly, p. 13, Lower Central Murchison.

Another variety of spear to which the reed-spear seems to have contributed is the two-piece spear of North Australia consisting of a long bamboo shaft and a short hardwood head with barbs cut in the solid (fig. 17). Except for the proportions of head and shaft the barbed feature of the head is the only important point of difference between - 63 it and the plain reed-spear. This North Australian weapon apparently represents the combination of the reed-spear, with its light-weight shaft; the composite wooden spear, with its separate head and shaft, which is also found in this region; and the idea of barbs cut in the solid, from the one-piece spear. The latter is still occasionally found in North Australia but now prevails in peripheral Melville and Bathurst Islands.

In view of the contiguity between the distribution of reed-spears in Australia and the distribution of reed-spears and reed-arrows in Torres Strait and New Guinea it would seem that a historical relationship for all can be reasonably implied. This should not be construed to mean that the Australian spear is a degenerate New Guinean arrow, although such may be the case. In view of the presence of reed-spears in both Torres Strait and New Guinea, however, it seems more likely that the New Guinean reed arrow developed from a reed-spear subsequent to the time the latter diffused into Australia.

Although the detachable barb is found associated with the reed-spear in New Guinea it seems probable that this is a relatively recent development. This conclusion is suggested by the peripheral appearances of reed-spears in Australia without the detachable barb. Detachable barbs, as we shall see, occupy a contiguous distribution throughout a large part of Australia, Torres Strait, and New Guinea.


On the basis of available information, spear-heads with barbs cut in the solid seem to be found in three main distributions, a south-eastern area, a western area, and a northern area (see map, fig. 10). It is quite possible that all three are more extensive than indicated, but until more information is forthcoming it is necessary to consider their modern distributions as shown.

In the south-eastern area a contiguous distribution is suggested from the Coopers Creek district of north-eastern South Australia to eastern Victoria and southern New South Wales. An apparently isolated appearance in the east is to be noticed at Broad Sound, Queensland. Since no information seems to be available for the intervening region of New South Wales and southern Queensland we cannot

- 64
FIG. 10., Distribution of Barbs cut in the Solid.
SPEARS WITH BARBS CUT IN THE SOLID., Multi-barbed—One Series., Western Australia. W.A.M.—Ashburton, 80 miles east of Mount Mortimer, Derby, Lake Way, Wyndham, Halls Creek, Roebourne. S A.M.—Lyons River. Gascoyne. N.M V.—Minderu, Majanna, Nichol Bay. U.P.—Mundi Windi, West Creek (Wiluna), Peak Hill, Black Range, Gwalia, Lawlers, Southern Cross, Wells Range. Clement, plates 2-3, Roebourne., Queensland. A.M.—Broad Sound (2-8 barbs), Bentinck Island (24 barbs). Roth, Bulletin 13, p. 196, Westwood (Rockhampton)., Victoria-New South Wales-Eastern South Australia. N.M.V.—Victoria A M.—Murrumbidgee, Dubbo, Edward River (Kyalite River), Lachlam-Darling. S.A.M—Adelaide Tribe, Murray River beyond Moorundi, Everard Range. Angas, plate 13, nos. 13 and 33. Brough Smyth, 1, p. 305, Victoria., North Australia. S.A.M.—Katherine River, Borroloola, McArthur River, Pine Creek, Roper River, Daly Waters, Port Essington (all with wooden shaft), Port Essington, Darwin, Escape Cliff. McArthur River, Alligator River, Daly River (with bamboo shaft), Melville and Bathurst Islands (one-piece), Goyder River, McArthur River, Borroloola, Daly River, Port Essington, Groote Island (with closed barbs). Spencer and Gillen, 1904, pp. 672-673, Warramunga and tribes to the north., Single Barbs., Western Australia. W.A.M.—Murchison, Roebourne (two-piece spear), Ashburton S.A.M.—Gascoyne (two-piece spear). Clement, plate 2, no. 1. Roebourne district (two-piece spear). U.P.—Black Range., North Australia. S.A.M.—Port Essington, Croker Island., New South Wales. Peron, plate 22., Queensland. Roth, Bulletin 13, p. 196, Maryborough., Multi-barbed—Two Series., With numerous large fine barbs—, Western Australia. W.A M.—“Northwest.” U.P.—Black Range, Lawlers, Southern Cross., North Australia. Melville and Bathurst Islands (typical).
- 65
With few large coarse barbs—, Western Australia. W.A.M.—“Northwest.” S.A.M.—Lyons River. U.P.—Halls Creek. Clement, plates 2-3, Roebourne area., North Australia. Basedow, 1907, pp 30-34, North-western Coastal Tribes. S.A.M.—Palmerston (Darwin)., Queensland. Roth, Bulletin 13, p. 196, Westwood (Rockhampton)., South Australia-Victoria-New South Wales. S.A.M.—Murray River. Coorong, Adelaide Tribe, Cooper's Creek. N.M.V.—Victoria. Angas, plate 5, no. 3, Murray River Brough Smyth, 1, p. 304, fig. 70, Victoria. Cawthorne, p. 4, Adelaide tribe., With numerous fine sharp barbs in two planes at a 120 degrees angle—, Western Australia. W.A.M.—Ashburton. N.M.V.—Ashburton, Nichol Bay. Clement, plate 13, Maratunia tribe., With coarse, uneven and reversed barbs—, Western Australia. W.A.M.—Kookynie. U P.—Black Range, Mount Hahn. Clement, plates 2-3, Roebourne area., With numerous coarse barbs (some in reverse)—, North Australia. S.A.M.—Borroloola, Alligator River, Port Essington, Daly River (squared barbs)., With a few long flaring barbs—, North Australia. S.A.M.—Darwin, Borroloola. Spencer and Gillen, 1904, p. 673, fig. 229, Daly River, Three to Five Series of Barbs., Three Series—, Western Australia. W.A.M.—Ashburton, Gascoyne, Murchison. S.A.M.—Gascoyne, Murchison. U.P.—Weld Range., North Australia. S.A.M.—Darwin W.A.M.—“Northern Territory.”, Queensland. A.M.—Burketown., Four Series—, Western Australia. W.A.M.—Roebcurne, Ashburton. U.P.—Byro Station, Wiluna. Clement, plates 2-3, Roebourne area., North Australia. S.A.M—Darwin, Daly River., Five Series—, Western Australia. U.P.—Byro Station.

say definitely whether this Queensland appearance is a modern separated occurrence or part of a large contiguous eastern distribution. Regardless of which alternative is correct for the modern aspect, there can be little doubt but that these two appearances are historically related.

In the western, north-western, and northern parts of the continent, a historical relationship is also suggested. There seem to be two main distributions at the present time, the central and north-western region of Western Australia, on the one hand, and North Australia, the Wellesley Islands and adjacent Queensland, and northern Central Australia, on the other. These two distributions are separated, however, by what appear to be several sporadic appearances in the Kimberley district, and these not only suggest a more widespread distribution in the north, but also a present or past contiguity with the distribution in the west. It also seems quite likely that the appearance of this type of spear-head in the Everard Range of north-western South Australia may be connected geographically with the western and possibly also with the northern distribution, or was contiguous to them in the - 66 past. The Everard Range, furthermore, is not far from the Coopers Creek region and a historical connection between these two appearances would seem to be a reasonable inference.

There are good grounds, therefore, for believing that all the observed appearances of spears with barbs cut in the solid occupied a contiguous distribution in the past and, on the basis of this reasoning, a unitary origin of the trait appears to be quite definitely indicated.

In view of the great variety in the forms of barbs which may be cut in the solid, we should not be surprised to find some variation in the three major distributions, each of which is characterized by important geographical, climatic, and floral differences. In spite of these differences, however, considerable consistency in generalized types is to be noticed not only throughout each area individually, but also throughout the three areas severally.

In the three regions we not only find the use of single barbs, but also of single rows of barbs and of double rows of barbs (figs. 11 to 20). A close relationship between the western and northern distributions is also indicated in the use of three and four rows of barbs (see figs. 21 to 23). In Western Australia this line of development

FIG. 11.
Victorian Spears (after Brough Smyth).
FIG. 12.
Spears with Barbs cut in the Solid. Victoria, upper; Adelaide tribe, lower (South Australian Museum).
- 67
FIG. 13., FIG. 14., FIG. 15., FIG. 16.
Spears with Barbs cut in the Solid., 1-3. North-western Western Australia (Western Australian Museum); 4. Everard Range, border of Central Australia-South Australia (South Australian Museum)., Melville Island Spears (South Australian Museum)., Spears with Single Row of Barbs, Western Australia. Weld Range, Peak Hill, Lawlers (U. of Pa. Museum)., Two-piece Spears with Barbs cut in the Solid. Geraldton, upper, and Roebourne, lower (W. Australian Museum).
- 68
FIG. 17.
North Australian Spears., 1. Groote Island; 2-3. McArthur River (heads made at Groote Island); 4. Roper River; 5-6. Pine Creek; 7. North Australia; 8. Wardaman tribe (1-7. South Australian Museum; 8. University of Pennsylvania Museum).
FIG. 18.
Spears with Double Row of Barbs, Western Australia. 1-3. North-west District (Western Australian Museum); 4-5. West Kimberley; 6. Ashburton; 7. North-western Western Australia.
- 69
FIG. 19., FIG. 20., FIG. 21., FIG. 22.
Victorian Spears. With two series of barbs (after Brough Smyth)., North Australian Spears with Specialized Barbs (South Australian Museum)., Spears with Three Series of Barbs, Western Australia. 1. Weld Range (U. of Pa. Museum); 2. North-western Western Australia (Western Australian Museum); 3. Kimberley and northwestern Western Australia., North Australian Spears with Four Rows of Barbs, Darwin.
- 70
FIG. 23.
Spears with Four Series of Barbs, Western Australia.1. Wiluna; 2. Byro Station (U. of Pa. Museum); 3. North-western Western Australia with four double series (Western Australian Museum).

apparently has been carried further than elsewhere, for in some specimens five rows of barbs are to be found. The spears in the south-eastern area seem to have undergone less specialization than in the west or the north, not only in lack of great variety but also in the lack of development of unusual forms. For the most part the barbs in this area are coarse, large, and few in number (figs. 11, 12 and 19).

In the northern and western distributions, however, not only are these crude barbs found, but also specialized types, some fine and sharp, some long and flaring, and some small and numerous. In addition, it is important to note that reversed barbs and barbed composite spears are also important in these two areas. On the basis of these similarities, a historical relationship between the two regions is further supported. Other evidence to strengthen this contention is found in the intervening Kimberley region, in the presence of barbs of similar types. The so-called “closed” barbs (fig. 17) seem to be indigenous to North Australia (Groote Island).

In all three major distributions some spears with barbs cut in the solid are made for use by hand, and others for use with spearthrowers. Weight probably is an important factor in determining the means of propulsion. Nevertheless, it is important to note that all the districts in which hand-spears of this type are important, Victoria, Ashburton-Gascoyne area of Western Australia, Melville and Bathurst Islands, the Wellesley Islands in the Gulf of Carpentaria, and Broad Sound, Queensland, are peripherally located in respect to those regions where their

- 71
FIG. 24.
Distribution of Detachable Barbs. Oblique hatching—Bone barbs., Vertical hatching—Round wooden barbs., Horizontal hatching—Flat wooden barbs., SPEARS WITH BONE-BARBS., Queensland. Roth, Bulletin 13, pp. 191 et seq. Examples are common in all collections from North Queensland and the Cape York Peninsula., New South Wales. N.M.V.—“New South Wales.”, Victoria. Brough Smyth, 1. p. 306, Swan Hill., SPEARS WITH SINGLE ROUND WOODEN DETACHABLE BARBS, Western Australia. W.A.M.—Kanowna, Mount Magnet, Wiluna, Geraldton, Lake Way, Israelite Bay, Greenough, Kookynie, Esperance, Laverton. S.A.M.—Lyons River, Nannine, Lake Lefroy, Geraldton, Eucla. Carnegie, p. 339, Desert Tribes south of Kimberley. U.P.—Bougardie, Mount Magnet, Meekathara, Peak Hill, Abbots Well, Sandstone., South Australia. S.A.M.—Fowler's Bay, Yardea, Cooper's Creek, Gawler Range, Great Bight. Schurmann, p. 3, and Woods, pp. 213-214. Port Lincoln. Howitt, p. 761, Yerkla-mining. Taplin, p. 95, Kukatha tribe. Angas, plate 13, no. 18, Port Lincoln., Central Australia. S.A.M.—Macdonnell Range, Alice Springs, Musgrave Range, Barrow Range. Spencer and Gillen, 1899, p. 578, Arunta; 1904, Urabunna, Ilparra, Luritcha. Helms, pp. 268-269, Warrina, plate 16, Fraser Range, p. 271, Everard Range. Bolam, p. 83, Ooldea White, p. 727, Everard Range., New South Wales. A.M.—Lower Darling River., SPEARS WITH SINGLE FLAT WOODEN DETACHABLE BARBS., Western Australia. W.A.M.—Greenough, Pingelly, Murchison, Geraldton, Bunbury (some with two barbs on opposite sides), Ashburton and Kookynie (some as preceding, some in tandem), York, Northampton. S.A.M.—Nannine, Gascoyne, Eucla, King George Sound. Lake Lefroy. U.P.—Albany, Esperance, Bougardie. Helms, p. 272, Murchison. King, 2, p 138, King George Sound.
- 72

use with spearthrowers is general. It would seem likely, therefore, that this type of spear was in common use before the spearthrower diffused into Australia. Apparently this type of barb was originally associated with heavy hand-spears which in many regions, seemingly those where the spearthrower has been in use for a long time, were modified in weight following the introduction of the spearthrower, or discarded in time as subsequent influences introduced lighter spears with other characteristics.

In view of the importance of the use of barbs cut in the solid in New Guinea and Melanesia it seems likely that the appearances in Australia are not indigenous but are the results of a diffusion to the continent from New Guinea by way of Torres Strait. Although there are a number of variations in Australia which, it cannot be doubted, are localized developments, there is such general similarity for the continent as a whole with the appearances of this type of barb in both spears and arrows of New Guinea and Melanesia that a common point of derivation in New Guinea seems to be the only plausible explanation. When the total distribution of this type of barb in these areas is taken into consideration, the lack of their appearance in the Cape York peninsula does not appear an important barrier to such a conclusion. This region, as well as other districts in Australia which intervene between the positive appearances of barbs cut in the solid, is now occupied by a different sort of barb, the detachable barb, which, we have every reason to believe, is a more recent invader.


Detachable barbs, as we have remarked in passing, are of widespread use in Australia. As is shown in fig. 24 their distribution extends in a broad band from Cape York on the north-east to the Perth region on the south-west.

(To be continued.)

1   This paper represents one of the studies carried on in the museums of Australia under a Fellowship grant by the Social Science Research Council of New York, and in the Museum of the University of Pennsylvania.
2   The Cambridge Expedition, vol. IV, p. 173, reported the bow and arrow at Cape York but attributed its presence to the visits of the island natives. Bows and arrows have been introduced as toys by Europeans in various parts of Australia but they seem to have never impressed the natives as suitable weapons.
3   Spencer, 1914, pp. 356-357.
4   Among the Wheelman tribe, who formerly lived in south-western Australia, the plain spear was preferred for hunting, as was also the case for the natives of Groote Island, off the coast of North Australia. On the Lower Tully river of Queensland this simple spear was used for fishing and for fighting. In the Rockhampton-Townsville region spears with stingray spine heads were employed in fighting but on Groote Island they were associated with fishing. Spears with detachable barbs were the fighting-spears of the Wheelman but were used by the natives of the Everard Range, a few hundred miles to the north-east, for both fighting and hunting. In the Pennefather river region of the Cape York peninsula spears with detachable barbs were the hunting-spears but in the Rockhampton-Townsville area they were preferred for fighting. In some parts of Victoria they were used for fishing. At Swan Hill, however, they were employed for fighting. On the Lower Tully river reed-spears served the dual purpose of fighting and hunting, whereas in Victoria they were used for fishing and for war. In Central Australia these spears were associated only with fishing and were always used in the hand. In North Australia, however, they are always associated with spearthrowers and the short variety used for individual combat, the long variety for hunting and war. This list could be extended indefinitely. See Hassell; Basedow, 1907, pp. 31 et seq.; White, p. 727; Tindale, 1925, p. 93; Roth, Bull. 7, pp. 190-192; Brough Smyth, vol. 1, pp 304-307