May 17, 1907] 



SCIENCE. 



765 



Another criticism deserved by all current 

 volcanologic research is that the investiga- 

 tions always follow the event, and are 

 wholly unsystematized. Even the Vesu- 

 vian observatory publishes nothing in sys- 

 tematic, tabulated form, and no instru- 

 mental records that may lead, eventually, 

 to prediction. Reviewing Pele, Soufriere, 

 Bandaisan, Tarawera and Krakatoa, aU 

 had scientific commissions, and in no case 

 did the commission observe the first erup- 

 tion or its forerunning phenomena. The 

 writer would point out the exceptional op- 

 portunity now possessed by the United 

 States for seismo-volcanologic research. 



Professor Jaggar's paper was discussed 

 by Professors Reid and "Willis. 



Experiments Illustrating Erosion and Sedi- 

 mentation: T. A. Jaggar, Je. 

 The fii-st experiments illustrating rill 

 erosion were made in 1899, and described 

 by Dr. E. Howe and the author (21st An- 

 nual Report U. S. G. S.). A coarse spray 

 was used, and the miniature streams of 

 water were an inch or more in breadth, 

 eroding models made of sand, marble dust 

 and coal dust. In 1901 a finer spray was 

 used, produced by a direct jet of water 

 deflected from an inclined and embossed 

 surface. An experiment in ' grand canyon ' 

 topography was executed with this appa- 

 ratus by R. W. Stone. Lateral planation 

 and delta sedimentation were well shown. 

 The latest apparatus consists of a series of 

 atomizers operated by compressed air. 

 Digitate drainage of some delicacy of pat- 

 tern is etched by the riUs on a model made 

 of modeling clay or mill slimes. These 

 rivulets apparently reproduce the mechan- 

 ism of 'bad land' drainage, or that of a 

 newly uplifted coastal plain. The models 

 made hitherto throw new light on the angle 

 of intersection of streams, on piracy, on 

 parallelism of tributaries and on the gen- 

 eral principles which govern the migration 



of divides. H. G. Ferguson has assisted in 



the last experiments. 



Characteristics of Various Types of Co^v- 



glomerates: G". R. Mansfield. (Intro- 

 duced by W. M. Davis.) 



The paper gave the results of a critical 

 review undertaken in connection with the 

 study of the Roxbury (Boston) conglom- 

 erate; examining the descriptions of con- 

 glomerates of many ages and countries as 

 to the conditions of their formation. 



Mr. Mansfield's paper was discussed by 

 Professors Grabau, Barrell, Miller, C. W. 

 Brown and Davis. 

 Dome Structure in Conglomerate: Ralph 



Arnold. Abstract not received. 

 Biver Sediment as a Factor in Applied 



Geology: W J McGee. 



Various recent events and movements 

 render it clear that the sediments trans- 

 ported by rivers require consideration, not 

 merely as a factor in geologic process, but 

 as an industrial factor of much moment. 

 In systematic geology it commonly suffices 

 to deal simply with the stream as a vehicle 

 transporting load in solution, in suspension, 

 and rolled or pushed along the bottom; 

 while in most industrial problems the 

 stream itself is regarded as the primary 

 agency, and the load is commonly neg- 

 lected, save merely as a disturbing factor. 

 Yet in that constantly increasing control of 

 streams (as sources of power, as supplies 

 of water for irrigation and other purposes, 

 as canals for transportation, etc.) which 

 must eventually result in the complete arti- 

 ficialization of river systems, the load either 

 immediately or remotely governs the char- 

 acter and behavior and ultimate utility of 

 the stream. The bearing of these consid- 

 erations on such engineering works as those 

 of the lower Mississippi region and on such 

 projects as that of the proposed Lakes-to- 

 Gulf Waterway, is particularly worthy of 

 discussion. 



