336 



SCIENCE 



[N. S. Vol. XLT. No. 1162 



fonner stream-planation and bToxial by moun- 

 tain wasli wliich by stream-action at a later 

 stage is again removed the fact remains that 

 the finest and most extensive rock-floors are 

 found in situations wbere no water-action 

 could possibly bave occurred. For all these 

 cases other suggestions of genesis is, of course, 

 necessary. 



The local exhuming of the rock-floors of 

 arid piedmonts by the removal of its wash 

 mantle does not really demand any elaborate 

 inductive reasoning in order to reach an ade- 

 quate explanation of the phenomenon. It is 

 one of the commonest features of tlie desert. 

 The effect is sometimes repeated over the same 

 district several times in a year. It has been 

 known to take place over night — by wind 

 action. In the semi-arid belt, or on the margin 

 of lofty mountains, as the Sierra Nevada in 

 California for example, the local removal of 

 the soil layer might be at first glance ascribed 

 to stream-action; but broader observation ex- 

 tending to typical desert regions, where only 

 low hills prevail, demonstrates at once that 

 the stream-planing hypothesis must be entirely 

 abandoned. The extension of moist-climate 

 principles of erosion to arid lands is done with 

 constantly growing difficulty. 



In support of the idea of the eolic derivation 

 of many rock-floored piedmont plains there are 

 ample published observations. The late W J 

 McGee's descriptions of the phenomenon as 

 displayed in Sonoran deserts are pertinent. A 

 single experience of my own when encamped 

 on the Jornada del Muerto at the northern end 

 of the Mexican tableland is by no means an 

 isolated instance. There at the foot of a 

 mountain apparently " biiried up to its 

 shoulders in its own debris " a strong gale 

 which suddenly arose completely swept away 

 in a half hour's time the supposedly deep soil 

 and laid bare the smoothest and hardest of 

 rock-floors worn out on the upturned edges 

 of most resistant strata. Since the situation 

 was at the mouth of a canyon and upon the 

 back of what appeared to be a broad alluvial 

 ■fan, a day's later visit might have ascribed the 

 phenomenon to stream work. 



Charles Keyes 



NOTES CONCERNING THE FOOD SUPPLY 

 OF SOME WATER BUGS 



In the literature dealing with aquatic 

 Hemiptera, we are informed that without ex- 

 ception they are predatory: those which dwell 

 upon the surface capturing such flies and 

 other terrestrial insects as may chance to fall 

 into the water, and those that pass their lives 

 beneath the surface preying upon aquatic in- 

 sects and similar organisms. 



In the light of recent observations along this 

 line, the above information seems inadequate. 

 Corixids for instance are largely herbivorous. 



The bulk of the food of our common water- 

 strider, Gerris marginaius, consists at certain 

 times of the year almost exclusively of the 

 Jassids and related forms that feed on Juncus 

 and other plants bordering on and growing 

 in the shallow waters. 



Our common species of Bheumatohates, 

 while it does not disdain to feed upon small 

 insects that fall into the water, obtains its 

 main supply from the little crustacean forms 

 such as Ostracods and Daphnians, which 

 swarm the quiet pools. These it captures as 

 they rest at the surface, scooping them out 

 and holding them aloft upon the upturned tip 

 of the beak, while the body of the little victim 

 is being depleted of its nutritive material. A 

 species of the genus Microvelia common in 

 Kansas has access to the same source for its 

 food supply and similar habits of consuming 

 it. Mesovelia mulsanti, our little green 

 Gerrid, has been observed exploring the sides 

 of stems of Juncus and Typha that lay just 

 beneath the surface for Ostracods, which they 

 occasionally obtained, while the well-known 

 marsh treader, Hydrometra martini, stalks 

 about over the floating vegetation in search of 

 whatever small beings chance to come to the 

 surface film. Its victims have been observed 

 to consist of mosquito wigglers, mosquito 

 pupse, emerging midges, nymphal corixids, and 

 Ostracods, as well as small terrestrial insects 

 floundering on the water. 



Among the bugs that live in the water, none 

 are more common than the back-swimmers, 

 or N"otonectids, and the water, boatmen, or 



