June 27, 1913] 



SCIENCE 



981 



rectly, either in Center or one of the adjoin- 

 ing counties. 



The finding of " petrified wood " and 

 " bones " possibly " human," in sinking the 

 shaft, would seem to indicate talus deposits, 

 but the " fern impressions " point to little 

 shifted carboniferous strata. 



Is the wood really " petrified," that is silici- 

 fied, or does it consist of sandstone casts of 

 same? If the latter, and in strata in place, it 

 indicates carboniferous, and there could be no 

 human bones associated with it. The refer- 

 ence to " loose shale, sharply inclined " up ( ?) 

 the hill, might indicate overplacement due to 

 creep. The nature of the strata (sandstone 

 and shale) composing the Devonian and Car- 

 boniferous of that region would preclude the 

 possibility of extensive underground channels 

 or caverns, which require limestone and would 

 consequently render very problematic the ex- 

 planation proposed by Mr. Andrews for this 

 Potter ice mine phenomenon. 



If a seasonal reversal of underground air 

 currents is to be invoked as an explanation, 

 his diagrams introduced to illustrate this had 

 better be inverted. Every " freezing cave " 

 that has been represented in vertical section 

 shows the more remote recesses of the cave 

 lower than the mouth, and the more nearly 

 vertical the circulation of the air is in conse- 

 quence of this, the better the conditions for ice 

 accumulation. 



This may be illustrated by the conditions 

 which prevail in a " freezing cave " near Gap 

 Creek, Wayne Co., Ky., visited in August, 

 1898, by the author of this criticism. The 

 cave is in the nature of a vertical sink, the 

 opening of which is situated on the top of a 

 Knob, called " mountain " in that section. 

 The top is about 1,260 feet above sea level (220 

 above the drainage at the base). Descent into 

 this cave was made by means of rough lad- 

 ders. Various channels ramify from the main 

 body of the sink, mainly downward. The 

 " mountain " which consists of Mammoth 

 Cave Limestone capped by a thin coping of 

 Kaskaskia Sandstone, appears to be honey- 

 combed with subterranean passages. 



Into these passages the cold air tends to de- 

 scend in winter and from them to rise in 

 summer, due to changes in relative density 

 conditions of the internal and external at- 

 mosphere. It appeared also to the writer (and 

 this is the only contribution he wishes to offer 

 to the explanation of ice caves) that the de- 

 scent of the water through the underground 

 passages during the winter, when this region 

 has its abundant rains, would aid in the in- 

 take of cold air at the mouth of the sink some- 

 what after the manner of a Sprengle pimip, 

 and hence would contribute to the thorough 

 refrigeration of the mountain or knob. This 

 agency would operate with little force in sum- 

 mer, when the stage of cave waters is low. 



We did not find that the inhabitants in the 

 neighborhood of this Gap Creek ice cave at- 

 tributed any of the accumulation of ice in it 

 to the summer months, and while it is not im- 

 possible in accordance with the theory usually 

 advocated for the explanation of such ac- 

 cumulation, that it might continue after 

 freezing temperatures had disappeared on the 

 outside, there is no indication that this condi- 

 tion of affairs has persisted into the summer, 

 either in the case of this or of any other genu- 

 ine "glaciere." The result of all actual in- 

 vestigation thus far is in support of this nega- 

 tive. Mr. Andrews fails to offer any first-hand 

 observation in favor of his contention, for by 

 his own statement his visit was made to the 

 " ice mine " in the spring. His belief in sum- 

 mer accumulation still rests on hearsay evi- 

 dence. 



In view of the fact that such phenomena as 

 we have been here discussing are by no means 

 uncommon (Balch in his work, " Glacieres or 

 Freezing Caverns," cited and quoted from by 

 Mr. Andrews, lists some three hundred in- 

 stances) it would appear that the subject is 

 important enough to deserve treatment in our 

 text-books on physical geography, where it 

 would appropriately come up under the head 

 of " caverns." 



Arthur M. Mjller 



Kentucky State University, 

 Lexington 



