The need to leave messages created perhaps the most widespread occasion for indigenous mapping. In the northern part of the continent, where the bark of the white tree birches afforded a widely available medium, maps were inscribed on the inner surface of the bark with wooden, stone or, later, metal styli and left behind in prominent or predetermined places. Until the middle of the nineteenth century such maps were particularly common in the upper Mississippi valley and Greak Lakes basin, where they were often found inserted into the end of a pole which had been stripped of its bark to make it conspicuous. They usually indicated the number and composition of the party, the mode of travel, the completed and proposed routes and the immediate destination. The cartographic component, usually limited to the representation of rivers, lakes, portages (at which canoes had to be carried between waterways) and camp sites along the route, was perfectly intelligible to those with a knowledge of the area but variable scale, stereotyped representation of features and disregard for direction resulted in patterns quite unlike those on a modern topographic map. Because of their short-lived function and fragile nature few examples have survived but one which was found in 1841 on the watershed between the Ottawa River and Lake Huron was forwarded to the United Services Institution by Captain Bainbrigge R.E. in the hope that it might 'shew young officers how small an effort is needed to acquire that most useful art, Military Sketching ... ' (Fig. 6).[15]


Other examples are related to hunting activities. One from Maine depicts in relation to a river system the movements of two Passamaquoddy hunters. They set out together up the main river but separated at a fork. One followed the right-hand branch as far as a lake, where he erected his tepee. Failing to make a kill he became hungry but was away hunting when his partner arrived from the left-hand fork. Finding a birchbark message with an upturned pot symbolizing hunger the partner added topographical information about the area around his tepee on the left fork, indicated his hunting success by means of a mooses' head at the site where he had made the kill and inserted the route between the two tepees as an invitation to his partner to join him (Figs. 7a and b).[16]


Extant examples of early maps drawn for indigenous use are relatively rare, either because they were not made to last, were made on fragile or otherwise impermanent materials or were considered unimportant by the whites who saw them. Occasionally they come onto the market. Recent examples include the British Library's purchase from the Royal United Services Institution of the birchbark map found by Captain Bainbrigge in 1841. As recently as October 1978 eight birchbark maps of northwestern Ontario were auctioned in Montreal. They were collected by Robert Bell about 1890 but Douglas W. Marshall of the William L. Clements Library has reservations about the provenance of four pieces purchased by the library. More exciting, but far more problematical, is a painted skin which is supposed to date from 1607 and to incorporate within its complex design a crude chart of the James River, Virginia. Purchased about 1940 by Frank H. Stewart of Philadelphia it is now in the Special Collections Division of the Learning Resource Center at Glassboro State College, New Jersey. It would appear to have been overlooked since Stewart wrote a newspaper article about it in 1945 and merits re-examination in the light of recent historical research and with the aid of new analytical techniques. [17] Items of this type in public collections are almost always classified as artefacts rather than maps and their cartographic content has not always been apparent to custodians or included in their descriptions. Whereas transcripts of Indian maps are usually readily found in archival and map collections, often associated with the papers of those who collected them, artefacts are usually included in the ethnographic collections of museums and private collectors. Hence, they are more difficult to locate. Dewdney's analysis of Southern Ojibway migration charts was based on eight certain and seven possible examples located in nine public (US 4, Canada 3, UK 2) and three private collections. An interesting Paw nee sky chart on buckskin in the Field Museum, Chicago, could provide the beginning for an equally interesting analysis of astronomical knowledge but would involve an act of faith that other examples could be found (Fig. 8).