SECT. 1] THE UNCONSOLIDATED SEDIMENTS 75 



dense than the water alone, hence the following relationships can be written as 

 typical of the water-sediment interface 



Ci ^ C2 piCi ^ P2C2 pi < P2. (1) 



2. Evidence for Gradients and Low-Velocity Sediments 



The evidence for the velocity structure shown by the velocity versus depth 

 curve in Fig. 1 is derived from both refracted and reflected waves. In a time- 

 distance graph, there is, theoretically, a reflection curve for each interface and 

 a refraction line tangent to it, the inverse slope of which gives the velocity in 

 the material below. If there is a velocity discontinuity at each interface and if 

 velocity does not vary with depth in any layer, the time-distance graph will 

 consist of straight line segments (for refracted arrivals) and h3rperbolic curves 

 (for reflections). If the velocity increases in each successively deeper layer, Riv 

 will cross Riv-i. In the case shown by Fig. 1, the average velocity, C2, in 

 Layer 2 is higher than that in Layer 1, but Rn does not cross Ri. This indicates 

 that the longer range arrivals on the Rn curve were not reflected from the 

 second interface, as were those at shorter range, but were bent upward by a 

 velocity gradient before penetrating to the depth of the interface. This behavior 

 of reflected arrivals is typical of all ocean-basin areas where detailed seismic 

 studies have been reported (Hill, 1952; Officer, 1955; Katz and Ewing, 1956; 

 Nafe and Drake, 1957; Ewing and Ewing, 1959). It is convincing evidence 

 that appreciable velocity gradients exist in the deep ocean sediments. Evidence 

 for velocity gradients is found also in refraction data and has been summarized 

 by Nafe and Drake (1957). Refracted arrivals from various depths in the 

 sediments taken from a number of profiles throughout the Atlantic Ocean 

 show a systematic increase of velocity with depth. The average value of 

 gradient is in agreement with that computed from reflection data. 



As mentioned before, the fact that refracted arrivals are not received from 

 the water-sediment interface is interpreted to mean that either there is no 

 velocity discontinuity across the interface or the velocity below is lower than 

 that above. We have evidence for both cases. In certain areas the first sub- 

 bottom reflection curve, Rn, is observed to approach a line parallel to and 

 above Ri on the time-distance graph. This can occur only if the velocity in 

 the upper sediments is lower than water velocity. In other areas the Rn curve 

 approaches Ri, indicating no velocity discontinuity across the water-sediment 

 interface. 



Further evidence for low-velocity sediments has been shown by Officer 

 (1955) and Katz and Ewing (1956) from the study of guided waves traveling 

 in the bottom just beneath the water-sediment interface. These waves have 

 been recorded in many areas of the Atlantic. They appear at a constant fre- 

 quency at ranges beyond that of the water-borne refracted wave which grazes 

 the sea floor (see Fig. 5, p. 79). To account for the constant frequency 

 and travel time of these arrivals, it is required that they be coupled waves, 

 excited by the grazing ray in the water, and that they travel in a wave guide 



