One of the most important elements in the history of the canoe is its early adoption by the French. Champlain was the first to recommend its use by white men. He stated that the bark canoe would be very necessary in trade and exploration, pointing out that in order to penetrate the back country above the rapids at Montreal, during the short summer season, and to come back in time to return to France for the winter (unless the winter was to be spent in Canada) the canoe would have to be used. With it the small and large streams could be navigated safely and the numerous overland carries could be quickly made. Also, of course, Indians could be employed as crews without the need of training them to row. This general argument in favor of the bark canoe remained sound after the desirability of going home to France for the winter had ceased to influence French ideas. The quick expansion of the French fur trade in the early seventeenth century opened up the western country into the Great Lakes area and to the northward. It was soon discovered that by using canoes on the ancient canoe route along the Ottawa River goods could reach the western posts on the Lakes and be transported north early enough to reach the northernmost posts before the first freeze-up occurred. The use of sailing vessels on the Lakes did not enable this to be accomplished, so that until the railroads were built in western Canada, the canoe remained the mode of transport for the fur trade in this area. Even after the railways were built, canoe traffic remained important, until well into the first half of the twentieth century as part of the local system of transportation in the northwestern country of Canada. [Illustration: Figure 3 CANOES FROM LAHONTAN'S _Nouveaux Voyages ... dans l'Amerique Septentrionale_, showing crude representations typical of early writers.] The unsatisfactory illustrations accompanying early published accounts have been mentioned. The earliest recognizable canoe to be shown in an illustration is the reasonably accurate drawing of a Micmac canoe that appears in Bacqueville de la Poterie's book, published in 1722. LaFiteau, another Frenchman, in 1724 published a book that not only contains recognizable drawings but points out reasons for the variation in the appearance of bark canoes: The Abenacquis, for example, are less high in the sides, less large, and more flat at the two ends; in a way they are almost level for their whole extent; because those who travel on their small rivers are sure to be troubled and struck by the branches of trees that border and extend over the water. On the other hand, the Outaouacs [Ottawas] and the nations of the upper country having to do their navigation on the St. Lawrence River where there are many falls and rapids, or especially on the Lakes where there is always a very considerable swell, must have high ends.