SPATIAL RELATIONS OF MAJOR TECTONO-IGNEOUS ELEMENTS AND THE ORIGIN OF MAGMAS 



593 



ficial granitic crust where various amounts of assimilation have occurred. 

 Of course, magmas intrusive into shallow sedimentary sections have 

 affected the overlying beds such as over laccoliths and bysmaliths, and 

 even over and around stocks in places, but these structures could not be 

 related to the origin of the magma. 



The Laramide structures of the alkalic province appear not to have 

 roots as previously suggested, and one of the most intriguing geophysical 

 studies is the seismic charting of the velocity layers in pursuit of this prob- 

 lem, and also the source of magmas there. 



Seismic Evidence of Crustal Structure. Tatel and Tuve (1955) report 

 the base of the basaltic layer ( Mohorovicic discontinuity ) under the 

 Colorado Plateau ( part of the alkalic province ) at the shallow depth of 30 

 kilometers. This was surprising because from isostatic considerations the 

 high plateau should have been supported by a crust some 50 to 70 kilo- 

 meters thick (50 to 70 kilometers to the M discontinuity). From this and 

 other data they conclude neither the Airy nor Pratt concepts of crustal 

 structure hold. Gravitv observations indicate a continent over which there 



J 



is isostatic compensation, and therefore, they conclude that the outer 

 mantle below the crust has density variations (see Chapter 31). Thus, 

 in turning to the outer mantle for causes of vertical movements, a column, 

 perhaps one to several hundred kilometers thick (or long) may be in- 

 volved, and if so, only small changes are necessary to elevate the plateau 

 5000 to 8000 feet. 



Basaltic Magma from the Mantle. Rasalt magma can originate (1) 

 by fractional melting of deep-seated earth material of different composi- 

 tion, (2) by complete melting of a deep-seated rock of the same composi- 

 tion, or (3 ) by complete melting of shallower rocks temporarily raised far 

 above the average temperature normally prevailing at that depth. If 

 surface basalts come from the mantle the process has been postulated to 

 be one of partial melting of a basic rock of the composition of stony 

 meteorites ( peridotitic ) , or of melting of eclogite, a heavy crystalline rock 

 of basaltic composition. Also the subcrust, presumably of basaltic com- 

 position, might melt in places to form a basaltic magma. The primary 

 olivine basalt of the Laramide Rockies is believed by the writer to have 

 come from the mantle and the following reasons are given. 



1. The seismic evidence of lavering, and the consequent interpretation 

 of gravity measurements indicate that rock density changes must occur 

 in the mantle. The Laramide structural province generally lacks roots 

 and is underlain by the thick 7.5 layer. Therefore, density changes in the 

 mantle are almost the sole explanation of isostatic adjustments. This sug- 

 gests that magmatic processes may be occurring there. 



2. The two primary basaltic magmas, tholeiitic and olivine, are best 

 explained as coming from the mantle. See discussion under a later head- 

 ing, "Tholeiitic Magma." 



3. The heat necessary for local partial melting of the upper part of the 

 mantle in places may adequately be provided by underlving conveetive 

 overturn, or possibly by generation along faults deep in the mantle. 



If partial melting of spots in the outer mantle shell is postulated as the 

 source of primary basaltic magma, a series of consequences must be en- 

 visaged, which, if the theorv is correct, must fit the pattern of structures 

 and igneous intrusions and extrusions through time and space as well as 

 geophysical observations and analyses. First, there are the considerations 

 of expansion. Since the Colorado Plateau is some 2 kilometers above sea 

 level and has no roots to buoy it up, we can think of expansion of the 

 mantle beneath to have raised the crust. In melting, a volume increase of 

 11.2 percent occurs, and if 6 percent of a 300-kilometer column should 

 melt, the crust would be elevated 2 kilometers. There would also be ex- 

 pansion in the solid state of the column over the convection cell as heat 

 is conducted upward. The problem is complicated and will require the 

 attention of experts, but evidently the amount of solid expansion is con- 

 siderable. The concept of rise of basaltic magma from the mantle is at- 

 tractive because it presents a plausible theory of the origin of the basaltic 

 subcrust. Instead of a primitive basic differentiate of a more silicic melt 

 from above, the subcrust would be the result of additions through time 

 from below. This view would hold for the basaltic substratum under the 

 continents, but for the ocean floors we would need to think of early out- 

 pourings, and after sufficient accumulation, increasing amounts of magma 

 from below to build up the basaltic layer. 



In connection with the arrest of basaltic magma from die mantle in the 

 subcrust we may think of uplifts like the Rlack Hills, Rig Horn Moun- 



