850 



SCIENCE. 



[N. S. Vol. XXIII. No, 596. 



diameter and tlie height of fifty-nine mounds, 

 located on an area of about a hundred acres 

 of ground near Keller's Bay, northwest of 

 Olivia post-office. The land has an elevation 

 of some thirty feet above the veater in the 

 bay and consists of clay sediments alternating 

 with some sand. The following table shows 

 the variations of diameters and heights of the 

 mounds measured: 



MEASUREMENTS OP FIFTY-NINB MOUNDS NEAE 

 OLIVIA, TEXAS. 



CONDITION OF FrFTT-NINE MOUNDS NEAB OLIVIA, 

 TEXAS. 



About one third of these mounds thus had a 

 diameter of from twenty-one to thirty feet. 

 About forty-sis per cent, had a smaller diam- 

 eter than this and only twenty-one per cent, 

 had a diameter exceeding thirty feet. The 

 extremes were three feet and ninety feet. Their 

 height ranged from two inches to eighteen 

 inches, and it averaged only a little more than 

 four inches. I noticed that their apparent 

 height was- quite deceptive. Measurements 

 invariably fell below my first estimates on 

 this dimension. 



Observations on two other features were 

 also made: on the presence of sunken pits on 

 the mounds, and on live anthills. The former 

 increased in frequency with the size of the 

 mounds and the latter decreased. One fifth 

 of the mounds exhibited sunken pits. In 

 most cases there was only one pit on each 

 mound, sometimes there were two, and in one 

 instance there were three. The shape of 

 these pits is irregular, and they have a tend- 

 ency to occur near the center of the mounds. 

 The anthills, on the contrary, often have a 

 peripheral situation. Of the forty-five mounds 

 noted as having anthills three mounds had 

 two and the others had only one each, thus : 



The clay on which these mounds occur is 

 evidently a lagoon sediment. It is red and 

 greenish-blue and too uniform in texture to 

 be a stream alluvium. 



I was informed by several parties that un- 

 der these mounds the sand continues down for 

 some distance below the surface of the sur- 

 rounding ground, while the rest of the sub- 

 soil is clay. It has also been found that when 

 land with such mounds has been inundated 

 for a rice crop, the water sinlss so rapidly 

 through the sand under them, that this land 

 can not be used for such purpose, if the 

 mounds are too numerous. 



It seems to me that from what is at present 

 known of these mounds the following views 

 may merit consideration, if we wish to apply 

 the method of multiple hypothesis in their 

 study : 



1. Differential settling of coarse and fine 

 sediments, as suggested by Hill. The cause 

 of the localization of such action then re- 

 quires a separate explanation. 



2. The anthill hypothesis. Some of the 

 mounds seem too large to be accounted for 

 in this manner, but my own observations 

 rather support it. 



3. The wind-drift hypothesis. The persist- 

 ent circular form is an objection to this hy- 

 pothesis, as is also the downward continuation 

 of the sand. 



4. Vertical brisk seepage of water under 

 hydrostatic pressure thi-ough thin clay strata 

 underlain by water-bearing sands, might re- 



