cause the canoe to turn turtle, if it fell on the other side, and thus pressed B—C into the water.
Another look at Fig. I (2) and (3) will help us to realise that the stability of the canoe will depend upon (i) the volume, and especially the depth of the dug-out; (ii) the distance B—C between the dug-out and the log; (iii) the size of the log C. The greater all these three magnitudes are, the greater the stability of the canoes. A shallow canoe, without much freeboard, will be easily forced into the water; moreover, if sailed in rough weather, waves will break over it, and fill it with water.
(i) The volume of the dug-out log naturally depends upon the length, and thickness of the log. Fairly stable canoes are made of simply scooped-out logs. There are limits, however, to the capacity of these, which are very soon reached. But by building out the side, by adding one or several planks to them, as shown in Figure I (4) the volume and the depth can be greatly increased without much increase in weight. So that such a canoe has a good deal of freeboard to prevent water from breaking in. The longitudinal boards in Kiriwinian canoes are closed in at each end by transversal prow-boards, which are also carved with more or less perfection (see Plates XXIV c, XLVII).