273 JOURNAL OF FORESTRY 



dency seems to be toward the angle iron and wire entanglement type, 

 for which ordinary treetops, anchored to the banks by cable, would 

 seem to be a cheap and satisfactory substitute. 



Another fundamental point of technique is that there must be some 

 sort of working plan for the creek as a whole. This plan need not be 

 complicated. It should consist merely in charting an ample fiood chan- 

 nel, as straight as practicable, and adjusted to such fixed points as 

 rock outcrops, bridges, headgates, existing willow banks, banks pro- 

 tected by heavy timber, masses of driftwood, or other existing barriers. 

 Having mapped out the flood channel, the work of each landowner 

 must be correlated thereto. Uncorrelated work too often merely pushes 

 the trouble down the creek. As an example of costs, the plan for 

 demonstration work on about half a mile of channel on the Walnut 

 Creek Ranger Station, Prescott National Forest, Arizona, calls for 

 two wagon loads of 18-inch willow cuttings, fifty pounds TNT, one- 

 fourth mile of old barbed wire, one team two days for placing drift- 

 wood logs and treetops. and about a week of the ranger's time. It 

 will be seen that the cash cost is not heavy. Of course, it remains to 

 be seen whether the work will be successful. If it proves successful, 

 the neighboring ranchmen, who have recently sustained heavy losses of 

 orchards and alfalfa, are already committed to extending the work to 

 their lands, under the ranger's supervision. 



UPLAND EROSION 



Erosion elsewhere than creek valleys is largely a problem of gullies, 

 although wind erosion is serious in certain dry-farming communities. 

 Upland erosion is more extensive in point of area affected but seems 

 less severe in threatened economic loss, than the creek valley problem. 



It has heretofore been almost dogma that well-sodded or well-tim- 

 bered land would not gully. As to timber, this is generally true in the 

 Southwest. As to sodded open land, it is often not true. Given the 

 foci of stock trails or trampled spots, and gullies are liable to run 

 through any deep soil in the Southwest, no matter how luxuriant the 

 sod. It is true that accidental removal of foci often causes gullies to 

 resod. But this is no solution. Erosion foci are bound to exist so 

 long as the range is used. Therefore artificial works are necessary to 

 prevent the spread of gullies. 



Brush and rock dams are undoubtedly the solution of the gully prob- 

 lem. But we know nothing about where and how to build such dams or 



