44 F. B. TAYLOR — CORRELATION OF BEACHES AND MORAINES. 



erable distance on a winding course through this part is the famous 

 " stone wall," a remarkable ridge of boulders, low and narrow, generally 

 only nine or ten feet wide. Locally it is supposed by some to be a work 

 of prehistoric man, and the marl which lies just below the surface among 

 the stones is supposed to be the crumbled mortar of the original wall. 

 This wall is said to extend ten miles or more, but was seen in only two 

 places by the writer. In some respects it resembles ice-beaches formed 

 on the shores of small lakes, but this explanation seems unsatisfactory 

 here and no better one has been suggested so far. The general course of 

 the wall in this part is parallel with the ice-front. In its upper part the 

 spillway valley grows narrower and divides into divergent head branches, 

 with small swampy patches here and there. Its floor is slightly uneven, 

 somewhat stony and sandy, but does not show effects of a flowing river 

 in a marked way. One branch comes in south of Freiburger and passes 

 northwest of this place to join the others. Two or three branches are 

 crossed between Freiburger and Tyre, and one of them, the most north- 

 erly, appears to open out of the upper part of the south branch of the 

 Tyre channel. The head branches of the Cumber spillway seem to show 

 that it was a smaller and more temporary affair even than the early Tyre 

 channel, and it seems clear that it does not, as at first supposed on seeing 

 only its lower part, mark the outlet for a distinct lake stage with a cor- 

 responding beach. Instead of this, it was probably one of several tem- 

 porary spillways that carried off the water, or a part of it, while the ice- 

 front was in active retreat from its previous position at the Detroit 

 moraine. It is doubtful whether the Cumber spillway ever carried off 

 more than a fraction of the whole overflow of the lake, and this probably 

 for only a comparatively short time. 



There is another river valley about ten miles south from Tyre drain- 

 ing westward from the crest of the thumb. It is a branch of Cass river 

 that flows westward about two miles south of Argyle and a mile north 

 of Shabbona. Near Argyle it has two branches, one coming from the east 

 and another from the southeast. The first has a small creek in it and its 

 valley is from 10 to 15 feet deep and from 400 to 500 feet wide. The sec- 

 ond is as large, but is swampy and carries only a small spring rill. Both 

 stream beds put together, however, appeared to fall far short of sufficient 

 capacity to carry the whole overflow. The valley is larger and deeper 

 toward Shabbona and its floor is well covered with boulders. It seems 

 too large for the present stream, although it is not greatly out of propor- 

 tion. This valley may have been a spillway for a small part of the over- 

 flow for a short time, but it lacks the characters of the abandoned bed of 

 a great outlet river. The present drainage divide on this part of the 

 thumb is five or six miles east of Argyle. The two small streams are 



