ages and thickens, magma can rise to greater heights in volcanic 

 conduits; for example, Azores volcanoes are not nearly as high as 

 Hawaiian ones. The greater the highes, the slower the spreading rate; 

 this explains the increase in the width of the pinelike region of 

 partial melting below the ridge. At high spreading rates, the wide 

 pipe can easily discharge the "hot spots" mass excess without 

 accumulation of a large mantle "mountain" such as that under 

 Iceland and the Azores. 



Plume-derived volcanic material may also be important in 

 producing certain geochemical anomalies found in sea-floor basalt. 

 Chlorine content, and strontium 87/86 ratios and 

 lanthanum/samarium ratios show unique distributions in the north 

 Atlantic that are difficult to reconcile with the single source of basalt 

 that is implied by the sea-floor spreading model. Work supported by 

 ONR and NSF has suggested that two sources may be supplying 

 volcanic material to the Mid-Atlantic Ridge. The first source is the 

 low-velocity zone of the upper mantle, which is depleted in chlorine 

 and large-ion rare earths, such as lanthanum, and has low strontium 

 isotopic ratios. The second source is a rising plume, which 

 transports material upward from the deep mantle. Iceland was 

 formed where this plume supplies material at a rate exceeding that 

 required to fill the gap left by the spreading plates. 



Continued refinement of the geomagnetic time scale is taking place 

 under NSF, ONR, and USGS support. Most notable ac- 

 complishments in this area are the clarification of the magnetic pole 

 reversal chronology between 10 and 25 million years ago and the 

 establishment of ages for the reversal sequence between 110 and 135 

 million years ago. Significant research is presently underway in an 

 attempt not only to continue refining the time scale but also to 

 understand the processes by which reversals of the magnetic field 

 are recorded by the ocean crust. 



A histogram of the number of polarity intervals during the last 45 

 million years, plotted against the length of the intervals, shows an 

 exponential decline in the number of intervals as their length 

 increases. This was suggested as evidence that reversals are 

 independent of each other. More complete statistical arguments have 

 recently suggested, however, that the reversal pattern may be 

 systematic. Research is attempting to resolve this conflict because 

 the reversal mechanism is a key to understanding the forces in the 

 earth's core that produce the geomagnetic field. 



Critical to understanding the statistics of the reversal sequence is 

 an evaluation of reversals lasting only a few thousand years. 

 Analysis of reversals during the last 5 million years indicated 

 numerous short polarity intervals. These short intervals, however, 

 have been much harder to find in older crust. Although this may 



38 



