336 



SCIENCE. 



[N. S. Vol. V. No. 113. 



yet to be supplied. The following method 

 has been devised to meet these difficulties : 



A condenser of considerable capacity is 

 first connected to the lighting mains and 

 charged at 220 volts. It is then discon- 

 nected and discharged through the primary 

 coil. The charging and discharging of the 

 condensers is effected by means of a com- 

 mutator. In this way the only current 

 passing through the coil is from the conden- 

 ser. The commutator is on the shaft of a 

 small fan motor. 



A six-inch Ritchie coil connected in this 

 way with a condenser of 25 microfarads, 

 its own condenser being disconnected, gives 

 a thick fuzzy spark about two inches long. 

 Removing the primary of the coil and re- 

 placing it by about seventy turns of rather 

 heavy wire, number 8 or 6 B. & S., we get 

 a multitude of fine zig-zag sparks about six 

 inches long, the discharge being identical 

 in appearance with that from an induction 

 worked in the ordinary manner under the 

 best conditions. The introduction of iron, 

 unless finely laminated, cuts down the dis- 

 charge to about one-tenth its value. In- 

 creasing the speed of the charge and dis- 

 charge of the condenser up to about 2,000 

 per minute, which is the limit of the very 

 crude commutator at present employed, im- 

 proves the discharge of the coil in quantity 

 and voltage. The sparking on the commu- 

 tator is very slight, and the amount of 

 power taken from the mains is small . 



The discharge obtained in this way, so 

 far as we can now judge, seems well suited 

 for driving X-ray tubes. Tubes so driven 

 give a brilliant fluorescent screen with 

 strong sharp shadows. An exposure of 

 twenty seconds gives a good photograph of 

 the hand. 



Chas. L. Noeton, 

 Ralph R. Lawrence. 



Rogers Laboeatoby of Physics, 



Mass. Institute of Technology, 



Boston, February 17, 1897. 



CURRENT NOTES ON PHYSIOQRAPHY. 

 THE COLORADO PLAINS. 



An essay by Gilbert on the ' Underground 

 Water of the Arkansas Valley in Eastern 

 Colorado' (17th Ann. Rep. U. S. Geol. 

 Surv.) affords more specific information as 

 to the topographic features of the plains 

 and their origin than is usually obtainable 

 from older reports. The general surface of 

 the plains does not accord with the surface 

 of their uppermost stratum, but bevels 

 across the strata at a faint angle. The 

 plains are therefore not in topographic 

 youth, but in topographic old age, the re- 

 sult of a cycle of denudation during a 

 lower stand of the land. On the peneplain 

 thus formed there are now strewn the 

 ' upland sands,' pebbly, cross-bedded, 50 to 

 200 feet thick ; the pebbles being derived 

 from the mountains on the west. This im- 

 plies a period of aggradation, after the long 

 preceding degradation. In explanation of 

 the change it is suggested that the region 

 may have been raised in the east or de- 

 pressed in the west so as to lessen the slope 

 of the rivers ; to this there may be added 

 a possible uplift of the mountains alone by 

 which the load of the rivers would have 

 been increased. To-day the sand- strewn 

 peneplain is trenched by broad valleys, 

 that of the Arkansas being fifteen or more 

 miles wide and 400 to 800 feet deep. Suc- 

 cessive pauses in the work of valley-cutting 

 produced broad straths at lower and lower 

 levels, whose remnants are now seen in 

 gravel-covered terraces, the seat of much 

 irrigated land. The eastward slope of the 

 terraces is greater than that of the present 

 grade of the river ; hence a progressive up- 

 lift is argued during the excavation of the 

 valley. The upland sands and the stream 

 beds at low water supply sand to the north- 

 west winds and extensive patches of dunes 

 are thus formed, a system of hills and 

 hollows without drainage by streams. The 

 relation of Cretaceous strata, upland sands 



