Arrow from Tsitsutl Peak Glacier, Western B.C.

400 year old arrow or dart from Tsitsutl glacier, B.C. Source: Keddie and Nelson: 2005.

In 1924, a land surveyor found an arrow at an elevation of 2,100 metres near Tsitsutl Peak in west-central British Columbia (map).  The arrow made its way to the Royal BC Museum where it lay for over 80 years, until a timely inquiry and increased awareness of ice-patch archaeology stimulated a small research program.  This research, initiated by RBCM curator Grant Keddie and reported in the Canadian Journal of Archaeology (Keddie and Nelson 2005), establishes that the arrow is about 400 years old.

Keddie and Nelson note that the arrow is made of western Yew wood (unusual) and is anomalously long for an arrow – 86 cm,but is missing an unknown portion of the nock end.   They estimate the original length at closer to 1.2 metres.  Combined with a larger-than-average side-notch point mounted on it, they suggest it may actually be an atlatl dart (humans need to throw things).  Atlatl technology from Yukon ice patches disappears around 1200 years ago, replaced by bows and arrows, but the authors cite an account of coastal Atlatl use in historic times (1788) in the Alaskan Panhandle near Sitka:

Atlatl's in 1788 as noted by Colnett in Galois 2004. Source: Keddie and Nelson 2005: 116

It is not commonly recognized that the atlatl is as much a maritime hunting technology as it is a terrestrial one!

Inuit hunter with atlatl lashed to kayak, ca. 1929. Photo: E.S. Curtis, so specific authenticity is questionable but general setting may be accurate. Source: Wikipedia.

It is not out of the question that this arrow is a dart and was in the possession of a coastal person hunting inland in the 17th century.  Or, perhaps such technology was in use much more recently in this area than in the Yukon. The unusual combination of attributes cited by Keddie and Nelson reinforces how little we know about the relationship between stone and organic  material cultural, especially when they are combined into composite tools of which only the stone portion usually survives.  So much extrapolation, so few actual specimens.  And now, as I noted before, and as was forcefully put by blog commenter “BCArchaeologist”, these glaciers and ice patches are melting quickly, and there is only a small degree of archaeological action to salvage their valuable payload – and essentially no such action in British Columbia, even by BC Parks.  As Keddie and Nelson demonstrate, we’ve known about frozen arrows in B.C. glaciers since 1924, and still no real action.

Detail of Tsitsutl arrow. Source: Keddie and Nelson 2005.


G Keddie, E Nelson. An Arrow from the Tsitsutl Glacier, British Columbia. Canadian Journal of Archaeology/Journal canadien d’archéologie. 2005;29(1):113-23

The atlatl quote is:

Galois, Robert (ed.) 2004.   A Voyage to the North West Side of America: The Journals of James Colnett, 1786–89. . Vancouver: UBC Press.

24 responses to “Arrow from Tsitsutl Peak Glacier, Western B.C.

  1. I’m really starting to envy west coast preservation! That’s an incredible artifact and I agree – without these all too rare examples, making the leap from stone artifact to a complete hafted tool is one of the most difficult journeys in archaeology.


  2. This paper by Otis T. Mason “Throwing Sticks in the National Museum” has many illustrations of ethnographic examples, some collected on the north Pacific coast, many from across the arctic. The coastal ones seem to be characterised by complicated grips for the hand or fingers – perhaps because of the conditions prevailing in small boats in cold waters; wet clumsy fingers, etc.:


  3. Interesting also that Mason experimented with their use – I wonder if he used the museum collection for his experiments.


  4. One of the reasons that throwing boards persisted in northern coastal areas alongside the bow and arrow was that they were so well suited for use from small single person boats, like the kayak. You can launch a dart or harpoon with much greater force than with your arm alone, but unlike a bow you only need one hand to do it. You can still hold the paddle with your other hand. Throwing boards were part of the kayak hunters gear in the arctic well into the historic period.


    • Good point, that makes a lot of sense. The kayaker has his by the left hand, presumably using the left hand to fit it onto the right, then the left to position the dart. The Colnett report suggests support with the left hand which probably means stabilizing the whole atlatl-dart combination before throwing with the right hand.

      I’d guess you could tell a left handed from a right handed atlatl? Or not?


      • Most arctic throwing boards are so ergonomically made that you can tell whether they are right or left handed. The one in the Curtis’ photo looks like it has an isolated pad for the fingers carved out to fit a right hand. It would have looked a lot like the one in this image in use:

        Another water connection is the name “atlatl” itself. Atlatl comes from Nahuatl, the language of the Aztecs. Interestingly, the Nahuatl word for water is “atl”, so the literal translation of “atlatl” is something like “water thrower” because it was used by the aztecs mainly for fishing.


      • Thanks Tim.

        IN that Curtis picture, what is the other device beside the atlatl, do you think? It is too fat to be a dart, it looks like a harpoon foreshaft plus shaft mounted tip towards stern, to me. But then why the atlatl? a Curtis bogus reconstruction, perhaps.

        Speaking of West Coast preservation — there must be permanent ice patches in Labrador, for example. Has anyone checked any out, do you know?

        Here’s a bonus picture of a guy with a large fish caught via atlatl:


      • It’s probably a harpoon. Throwing boards used with kayaks launched a variety of light and heavy darts, including harpoons.


  5. Despite the length of this projectile, I still think it is likely that it is an arrow. Without the nock end we’ll never know. (Arrows have a straight slot for the string, whereas darts have a round socket for the atlatl hook). I note that the apparent shaft width at the haft, and the probable neck width of the point, is probably a little under a cm accounting for the binding materials. One cm is the ‘crossover’ point where the great majority of arrows in ethnographic collections are smaller, while darts are larger; though there is considerable overlap. Thus a chance remains that this is an arrow.

    Corliss published a series of articles and comments about this, the latest and perhaps definitive was amusingly titled “Arrowpoint or Dart Point: An Uninteresting Answer to a Tiresome Question”.

    Incidentally, the large size of the point is consistent with arrow points from the very late precontact period across southern BC interior. The AVERAGE width of side-notched points from the late matlodge we nearly totally excavated at the Kamloops Railyard was greater than one or two std deviations from the mean of the reported Kamloops side-notch points at the time (can’t remember exactly, it was nearly 30 years ago I wrote the report).

    Another thing to keep in mind about the dating is to check if this is, as I believe it is, one of the ‘flat’ periods on the calibration curve that can translate a +- 30 year error into a +-200 year error (Two Eyes will remember without looking it up I’m sure)!


    • Hi Morley,

      here is the calibration table from Keddie and Nelson:

      It is a flat period: a date of 335 +/- 30 at two standard deviations, falls between AD 1482 and AD 1639

      I find it interesting that the sample name as sent to the lab is “Dart-1”, suggesting that they had already decided it was a dart before they got the date, which is not precisely how the article explains thing!


      • The arrow/dart was briefly displayed at the RBCM in 2001. I took a picture of its label which was titled at that time: “Rare atlatl dart” and stated it “is the first atlatl dart found in British Columbia”. It was identified as yew and noted “The style of point suggests the dart may be between 3000 and 6000 years old”.


  6. Ouch, APM!! I dunno where they would have got that date from; plains sequence? Maybe they were thinking the Oxbow-style points from the Shuswap phase. The early side-notched on the coast, to the south, would be Cold Springs side-notched, much earlier.

    I was pleased to see one of the atlatls discussed by Otis was left-handed! Reject dextrocentricism!


  7. Another thought about the length; I saw one Internet ref about arrows to 1.2 m long; were some arrows especially long for certain functions? Fishing? Not likely on a glacier, but….


  8. Sorry for three posts in a row, but couldn’t resist posting this:


  9. “humans need to throw things”
    See also W.H. Calvin, The Throwing Madonna (McGraw Hill 1983) Chapter 4. Did Throwing Stones Lead to Bigger Brains?
    The hominid brain has enlarged threefold in the last few million years and, along the way, acquired specializations for language, handedness, and even music. What were the selection pressures which shaped this rapid evolution? An examination of one-handed throwing of stones at prey and how this cultural practice could have led to the enlargement of the brain-and provided the foundations for language cortex.


    • Hi george,

      One thing about throwing is, it must be under pretty clear selective pressure whether defensive or offensive or both so I can see it as a plausible prime mover. If this is the same Calvin who proposed thrown handaxes as deadly projectiles I’m not sure I buy that – but it doesn’t take away from the point that throwing things (spears, rocks) has probably been pretty important human behaviour under a lot of mission-critical pressure for a long time.


  10. Anne was just telling me about a TV show this evening that showed Brazilian monkeys not only selecting and storing nuts over a 5 or 6 day period, but processing them with stone mortars and pestles; large rounded rocks brought a long way from rivers then dropped onto the nuts to open them. Repetition has caused pecked depressions, essentially mortar features, in the bedrock (which till the primate studies were thought to be archaeological). They also bring rocks to store on ledges and clifftops to throw or roll at predators below to drive them away. Amazing, I presume this is now peer reviewed etc. They have pretty small brains still. Are these complex behaviours selecting for intelligence I wonder?


  11. Youtube has everything!

    Similar behaviours have been seen in higher primates in Africas as well, where the variety of chimp technology falls out into distinct “chimp cultures” each with a different suite of material culture and technology.. The interesting thing about this is, you could indeed study primate tool use archaeologically and see if there was a correlation with brain size. I could imagine a baboon cave or something with all kinds of technology in it once you learned to look.

    Here is a recent paper:

    Selection of Effective Stone Tools by Wild Bearded Capuchin Monkeys

    Current Biology, Volume 19, Issue 3, 213-217, 15 January 2009

    Elisabetta Visalberghi1, , Elsa Addessi1, Valentina Truppa1, 2, Noemi Spagnoletti1, 3, Eduardo Ottoni4, Patricia Izar4 and Dorothy Fragaszy5

    * Appreciation of objects’ affordances and planning is a hallmark of human technology. Archeological evidence suggests that Pliocene hominins selected raw material for tool making [1,2]. Stone pounding has been considered a precursor to tool making [3,4], and tool use by living primates provides insight into the origins of material selection by human ancestors. No study has experimentally investigated selectivity of stone tools in wild animals, although chimpanzees appear to select stones according to properties of different nut species [5,6]. We recently discovered that wild capuchins with terrestrial habits [7] use hammers to crack open nuts on anvils [8,9,10]. As for chimpanzees, examination of anvil sites suggests stone selectivity [11], but indirect evidence cannot prove it. Here, we demonstrate that capuchins, which last shared a common ancestor with humans 35 million years ago, faced with stones differing in functional features (friability and weight) choose, transport, and use the effective stone to crack nuts. Moreover, when weight cannot be judged by visual attributes, capuchins act to gain information to guide their selection. Thus, planning actions and intentional selection of tools is within the ken of monkeys and similar to the tool activities of hominins and apes.


    • This link below shows the stone anvils, pitted from long use. Also, near the end, the monkeys throwing rocks off the cliff in the direction of jaguars, at least that is what we are lead to believe. Irritating music score.

      Those monkesy sure are strong – I’d hate banging away with a rock of that size relative to my body weight, or even relative to my head size (no wise cracks please, so to speak :)).

      In any case, very interesting with lots of implications for early hominid archaeology.


  12. Wow! Those are convincing videos. This blog is rather like going to the pub after an ASBC meeting or rockwash – but cheaper and easier to hear whats being said!


  13. As predicted, 4,000 year old Chimpanzee archaeology, stone tools included:


  14. When I first examined the glacier arrow or dart, I thought is was more like a dart on the basis of its length and the fact that it was made of yew wood. I started making my first bows and arrows when I was 6 years old and still have the first fibre glass bow I bought when I was 7. This glacier arrow was larger than the ones I used as an adult. I though it was a late style point, but when Earl Nelson showed images to Phil Hobler who was more familiar with the region of the find than me, Hobler suggested – based on its size – that it could be 6000 or 7000 years old . Given that I thought it was more likely to be an atlatl which would likely be earlier, I suggested the possibility of that date in the small temporary weekend exhibit. After dating it I am more inclined to assume it is an arrow because of the late date and lack of evidence that atlatls were used in the region in later times. A few other comments: The Aztec descendants used the atlatls out in boats to hunt mostly birds, not fish. This was also how they were used in historic times in Siberia. Geese were hunted on lakes when they were molting and could not fly far. Most atlatls are not left or right handed. The Aleut atlatls are one of the exceptions. I leaned the hard way when I arrived on a the movie set of “The Golden Seal” to train the main actor to throw the weapon. The movie took place in Beecher Bay but was – in theme – taking place on the Aleutian Islands. I arrived early on a Saturday morning with two Aleut style right-handed throwing boards and several darts, only to find that the actor was left handed. I have thrown my Aleut style darts and larger ones with a larger throwing board from my Kayak. I have found it is very difficult to use the larger throwing boards this way. Because you are sitting down you get a lot less distance. The accuracy, even with the lighter darts, is only good for short distances of a about 10 meters. When Aleut hunted sea otters they surrounded groups of them in their Kayaks and went in to get them at close range.


  15. Pingback: Atlatls to Bows: A Suspiciously Large Arrow « Gambler's House

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s