90 



SCIENCE. 



[N. S. Vol. XXI. No. 525. 



condition produced by the action of the oil. 



An attempt to measure directly the dif- 

 ference of potential between the two ends 

 of an isolated unequally heated bar of 

 metal would, apparently, encounter ob- 

 stacles quite as great as those which have 

 thus far proven unsurmountable in the 

 case of attempts to measure directly the 

 contact difference of potential between 

 metals. The outlook is, therefore, not 

 bright for any immediate and final answer, 

 on experimental grounds, to this question 

 of the direction and magnitude of the local 

 electromotive forces with which we have 

 been dealing. 



I wish to add one afterthought. If elec- 

 tricity flows like a perfect gas through a 

 homogeneous solid conductor of uniform 

 cross-section, its velocity at any given cross- 

 section of the conductor must be, approxi- 

 mately at least, proportional to the abso- 

 lute temperature of this cross-section. Now 

 the ordinary law of resistance in the case 

 of a fluid moving through small passages 

 is this : Resistance is proportional to the 

 velocity. Accordingly, we are led to the 

 conclusion that the resistance encountered 

 by our electric stream should be propor- 

 tional to its velocity, that is, other things 

 being equal, proportional to the absolute 

 temperature at the part of the conductor 

 considered. Now we know that in pure 

 metals this is the general law of resistance, 

 and the fact that this law finds an explana- 

 tion in a conception of the electric current 

 formed without any reference to electrical 

 resistance adds considerable weight to the 

 argument in favor of that conception. 



Edwin H. ITali.. 



IIakvari) T'niversity. 



777 A' ALAMOGORDO DESERT.* 



TiiK Alairiogordo desert of southern 



New Mexico lies immediately west of the 



* Address by the vice-pi esidont and chairman of 

 Section (i for" 1904. Philadelphia, Pa. 



106th meridian, west, and approximately 

 between thirty-two and thirty-four, north. 

 It is bounded on the north by the Oscuro 

 range of mountains, on the east by the 

 Sacramentos, on the south by the Jarillas 

 and the Organ mountains, on the west by 

 the San Andreas. As here defined, there- 

 fore, the desert is of comparatively limited 

 area, one hundred or one hundred and 

 twenty-five miles from north to south, and 

 perhaps thirty-five to fifty from east to 

 west ; a very convenient little deserr, easily 

 manageable, one might suppose, for any 

 naturalist, who, with inborn love of ad- 

 venture, starts out in search of the wilder- 

 ness to find scenes and pastures new. 



A year ago in this presence, it may be 

 i-ecalled, the present speaker, by aid of 

 photographic illustrations, attempted to 

 sketch the relations obtaining, as would 

 appear, between the geology of the desert 

 and its flora ; in the present paper it is 

 intended briefly to resume the earlier argvi- 

 ment with such added reflections as may 

 be suggested by present conditions and by 

 recent renewed acquaintance with the prob- 

 lem. 



The desert of Alamogordo or Tularosa 

 is a great plain, not unmarked, however, 

 by singular topographic inequalities later 

 on to be described. Only the most casual 

 geologic examination is sufficient to show 

 that the plain floor corresponds strati- 

 graphically with the beds in some places 

 exposed at or near the tops of the sur- 

 rounding mountains, in any case far up 

 their flanks. On the east especially lime- 

 stones of carboniferotas age rise sheer some 

 1,000 feet or more straight up from the 

 desert floor, and are again capped by other 

 strata only at length, perhaps 1.000 feet 

 higher, surmounted by materials corre- 

 spondent with those in the level of the 

 plain. 



On the west the same thing is true : but 

 more emphasized still is the difference in 



