November 3, 1916] 



SCIENCE 



637 



fessor See proposed to explain many things, 

 among them being " the direct effect of sun- 

 light on a magnetic needle, as in Nipher's 

 experiment of 1913." This was a complete 

 surprise. Evidently this experiment had been 

 tried with success by I suppose Francis E. 

 ISTipher, of Washington University, St. Louis, 

 Mo. 



It seems to me that such an experiment 

 would be valuable to science in many ways. 

 The question arises as to the quantitative effect 

 produced — if appreciable, then might we not 

 expect or predict a change in all magnets 

 more or less with time — especially as they are 

 exposed to the sunlight? It is well known 

 that magnets lose some of their magnetism 

 during the process of ageing. Might this effect 

 be a contributing cause? 



The question as to the effect on small mag- 

 nets such as in use for the determination of 

 the earth's magnetic elements assumes some 

 importance when considered in this regard. 



What might be the effect of the sunlight on 

 the magnet if it were rotated about a hori- 

 zontal line through its center of mass and 

 perpendicular to its magnetic axis ? The theory 

 of magnetization by rotation has been treated 

 in two articles appearing recently in Science 

 by Barnett. 



Aside from the foregoing it would be inter- 

 esting to note the effect, if any, of radio- 

 active emanations upon a magnetic needle. 



There are two well-known cases of the trans- 

 formation of luminous into electrical energy, 

 the thermopile and the photo-electric cell. 

 However, in neither one is the transformation 

 direct, as would be the case of luminous energy 

 falling upon the magnetic needle. 



It would be interesting to see this matter in- 

 vestigated in the light of modern electrical 

 theory and to know of ISTipher's experiment 

 and of the results obtained. 



F. C. Loring 



Department of Terrestrial Magnetism, 

 Carnegie Institution 



gumbotil, a new term in pleistocene 



GEOLOGY 



The term gumbo has been used for many 

 years by some geologists in America for a 



dense, impervious clay, which, when saturated 

 with water, is sticky and tenacious. The 

 name has had no relation to the origin of the 

 material : in many cases it has been applied 

 to alluvial deposits on the flood plains of 

 streams : McGee, Leverett and others have 

 applied it to a gray to drab-colored clay over- 

 lying drift, the origin of the gumbo having 

 been attributed to various causes, some having 

 considered it to be, mainly, of ftuvio-glacial 

 origin, others to be aqueous, and still others 

 have thought it to be related to loess. 



In a recent paper in volume 27 of the Geo- 

 logical Society of America, pages 115 to 117, 

 the writer discussed a gumbo which lies on 

 Kansan drift and which he had studied in con- 

 siderable detail in southern Iowa. This 

 gumbo is limited in distribution to tabular 

 divides and other remnants of the Kansan 

 drift plain. The view was there expressed 

 that the field evidence suggested strongly that 

 the gumbo is the result, chiefly, of the chem- 

 ical weathering of Kansan drift. It was 

 stated, also, that detailed chemical analyses 

 of the gumbo and the underlying materials 

 were being made by Dr. J. N". Pearce, of the 

 chemistry department of the University of 

 Iowa, to ascertain whether the analyses would 

 strengthen or weaken the interpretations made 

 from the field evidence. These analyses have 

 now been completed and will soon be pub- 

 lished. They seem to show clearly that the 

 gumbo is the weathered product of the drift. 



During the present summer, the writer has 

 extended his studies into the western, north- 

 western and northern parts of Iowa, and at 

 scores of places sections have been examined 

 which show clearly the intimate relations be- 

 tween the gumbo and the underlying Kansan 

 drift. Moreover, it is of interest that in many 

 places a gumbo has been found on the Nebras- 

 kan drift, the relations of the gumbo to this 

 drift being similar to those of the super- 

 Kansan gumbo to the Kansan drift. Further- 

 more, after a somewhat careful study of the 

 gumbo which lies on the Ulinoian drift in 

 southeastern Iowa, and which has been dis- 

 cussed by Leverett in Monograph XXXVIII. 

 of the United States Geological Survey, 



