314 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



park. Owing to an effect of shadow one can distinguish in this view 

 all the details of the relief of the ground. They are significant. At 

 the same time that the descending tide washes the elongated sand 

 bank at the middle of the river from above downward water runs lat- 

 erally to join it from the channel at the right and from a rivulet fur- 

 rowed in the mud at the margin of the shore at the left. Ripple 

 marks are formed at this place. 



Let us now return to the great beach of Goulven, represented at 

 thelbeginning of this article (fig. 1). If one introduces this view be- 

 tween the following (fig. 14) and that which we have given in figure 

 5, the accompanying panoramic view (fig. 15) will be obtained. We 

 see at the right the shore and the most distant point of the bay dom- 

 inated by a picturesque roclcy mass called " The Cathedral '' ; at the 

 left, stretched across the bay, a series of high granite ledges which 

 extend above the water. The waters have receded and leveled the 

 slope of the sands accumulated around these knobs. But directed in 

 consequence toward the middle of the bay they have been obliged, in 

 order to get out, to seek two lateral outlets. They have taken the 

 direction of " F " and " F'." Do we not have here again, on an im- 

 mense scale, our pool of Pouldu, with its two lateral outlets and its 

 ridges parallel with one another and with the shore ? 



The similarity of the two illustrations will be better comprehended 

 by turning again to figure 1 and examining very closely (with a mag- 

 nifying glass, one might say) the smallest details of the landscape. 

 The lines of ripple marks are drawn out over a long distance, but in 

 reality they show at a number of points a lack of continuity. They 

 appear to be composed of bars which follow one another irregularly, 

 or do not stand in line at all. To summarize, this immense field of 

 undulations may be divided into an infinity of pools similar to that 

 of Pouldu and of banks much less important even than that of Trieux. 

 They form a mosaic. This admirable design is effected by means of 

 fragments and pieces, pools and banks of sand in juxtaposition, and 

 separated by a series of parallel rivulets, some still full of water and 

 others already completely drj^ just as shown in the first illustration, 

 where one may see at the right ridges becoming progressively de- 

 formed and losing themselves at the place where the two generating 

 currents have taken a common direction. 



The same spectacle presents itself within the bay as at its mouth. 

 Everywhere, from whatever point we take up the question, we always 

 arrive at the same conclusion. All these marks (with certain excep- 

 tions due to the movements of the ground, which, if we stop to study 

 them, witness in favor of our theory) in their entirety are directed 

 parallel to the shore toward a rapid creek formed on the right side 

 of the bay by the retreat of the water which the sea every day carries 



