140 GRAND CANON DISTRICT. 



autumn. They vary in size and depth very considerably. Some are as 

 narrow as 20 feet ; some are ,300 to 400 feet across. The depths vary 

 from a yard or two to a hundred feet. The form is crater-like— always 

 approximately circular. They do not appear to occur under any special 

 set of conditions. They are found as often upon the platforms as in the 

 valleys and are not uncommon upon the slopes of the ravines. In a few 

 instances traces may be seen of rain gullies or washes leading into them 

 but not often, and none have ever been noted leading out of them! 

 Whatever running water may enter them sinks within their basins; 

 but it is certain that many of them rarely receive any running watei In 

 the cases of those which do the wonder is that they donot soon till up wit h 

 sand and silt, for the water generated by heavy rain storms or by melt. 

 ing snows, when sufficient in volume to run in a stream, is always thick 

 with mud. The scarcity of running water on the Kaibab has been men- 

 tioned. Yet the precipitation is comparatively great and the evapora- 

 tion small. It is apparent that all the water which falls upon its vast 

 expanse, with the exception of a slight percentage evaporated, must sink 

 into the earth, where it is doubtless gathered in subterranean drainage 

 channels which open in the profound depths of the great amphitheaters 

 of the Grand Canon. In those depths are large creeks of perennial 

 water issuing from the openings of those underground passages. This 

 implies a system of subterranean rivulets, but it is not more wonderful 

 than the endless caverns in the limestones of Kentucky and Indiana, 

 and it is probably not upon so large a scale nor so greatly ramified. 

 It also argues a high degree of permeability both in the upper strata and 

 in the overmanning soil. The water sifts through them as easily as 

 through sand, and rarely gathers into streams even in the most copious 

 showers or most rapid melting of the snow. Whether these « lagoons" 

 and "sink-holes," as we termed them, are the openings of pipes lead 

 ing down into the subterranean rivers and kept open by a gradual 

 solution of the limestone, it is difficult to say. There are some diffi- 

 Jultes in the way of this theory. 



Moving rapidly southward, at length we reach the Svlvan Gate at the 

 lower end. Passing through we immediately find ourselves at the head 

 of a second park very similar to De Motte's, but smaller, having a length 

 of nearly three miles. It is named Little De Motte Park, and the Sylvan 

 Gate occupies a divide between the two. It contains a largo lagoon 

 holding stagnant water. There is a chain of these parks reaching from 

 the northern end of De Motte's southward, a distance of 25 miles, sepa- 

 rated only by necks of forest. 



Our first objective point is a spring situated in one of the large ra- 

 vines which head in the heights overlooking these two parks. Without 

 some foreknowledge of the way to reach it, or without a guide, it would 

 be impassible to find it, and the same is true of any other spring on the 

 summit, but with this foreknowledge we seek the southwestern border 

 of Little De Motte and enter the timber. During half an hour there is 



