NOV. 19, 1921 WASHINGTON: GRANITES OF WASHINGTON 459 



2.0°. The rapid fall of temperature with comparatively slow rise 

 of field is what w^e would expect from the equation. Now if we assume 

 that the equation above is true, and find tq by substituting the known 

 values, we shall get 5.3° for the normal transition temperature, against 

 the estimate of Onnes of 6°. This I believe must be considered to be 

 within possible experimental error. 



Another detail of the behavior of the magnetic field is also con- 

 sistent with this view. Onnes found that at a fixed temperature 

 the threshold value of the field was slightly higher for a field parallel 

 to the wire than for one transverse to it. The probable reason is 

 that when the field is transverse there are regions inside the wire 

 itself where the circumferential field due to the current in the wire 

 is in the same direction as the external applied field, whereas this 

 is not the case for the longitudinal field. Hence in the transverse 

 case the maximum field is greater than the measured field, and the 

 applied field does not have to be raised so high in order that the 

 critical value may be reached. 



The temperature of transition of the four metals hitherto measured 

 is depressed in a magnetic field. This means that the phase stable 

 at the lower temperature has the greater susceptibility. It does 

 not appear to me whether there is any reason why this should be 

 universally true, or whether it is possible that there may be substances 

 whose temperature of discontinuity is raised in a magnetic field. 



This view of the discontinuity as due to a polymorphic change 

 ought to be capable of independent experimental verification. It 

 should be possible to detect discontinuous changes in other physical 

 properties — so far as I know the search has not been made for these — 

 and it may be possible to realize supraconductivity in metals which 

 do not normally show it, by recrystallizing a wire at higher tempera- 

 tures, as can now be done for tungsten, before subjecting it to low 

 temperatures. 



PETROLOGY.— r/z^ granites of Washington, D. C} Henry S. 

 Washington, Geophysical Laboratory, Carnegie Institution of 

 Washington. 



The granites of the District of Columbia and its vicinity belong 

 to the great belt of Archean intrusives (mostly granite with smaller 



' Received October 21, 1921. Presented before the Geological Society of Washington 

 October 26, 1921. 



