NO. 1 EMERY AND HOLSEMANN: SUBMARINE CANYONS 37 



of the isopleths in a few of the canyons. At Redondo Canyon the tem- 

 perature and oxygen content is higher and the salinity and nutrients are 

 lower near the head than farther seaward. This difference is just what 

 is to be expected of local upwelling. A similar conclusion is indicated 

 by the less complete data at Dume Canyon, but on the other hand possi- 

 ble downwelling may have occurred at Mugu and La Jolla canyons. 

 Clearly, upwelling was not marked at the times of the surveys, but 

 then the wind and sea conditions were fairly calm at these times. At 

 times of strong winds, movements of water along the canyons may be 

 more intense. 



It seems evident that the water is not of such unusual character as 

 to present an abnormal environment for benthic animals; thus any ab- 

 normalities in size of individuals or groupings of the fauna must be due 

 to some aspect of the environment other than the water within the 

 canyon. 



A major abnormality in the benthic fauna is indicated by the fact 

 that 22 samples from six canyons (Table 2) consist almost exclusively 

 of Capitella, a polychaete worm which ordinarily lives in estuarine water 

 (Hartman, 1962). These same samples are free of marine worms and 

 of other marine animals except carnivores such as squid, which may 

 not really inhabit the sites. Since Capitella lays its eggs in the tubes in 

 which it lives, wide dispersion through sea water is unlikely. It is sug- 

 gested that the samples represent sites at which fresh water escapes into 

 the ocean from aquifers which have been intersected by cutting of the 

 canyons. Escape of fresh water is known to occur from many nearshore 

 areas of the sea floor of the world. Accounts of its escape from sub- 

 marine canyons go back at least to Benest (1899). Johnson (1938- 

 1939) even postulated an origin for submarine canyons on the basis of 

 submarine erosion by escaping ground water, but his concept is now gen- 

 erally considered less plausible than others. 



It is quite reasonable that a submarine canyon should be a local 

 focus for escape of ground water because it is the farthest landward 

 point of outcropping horizontal strata, and thus a point of steep pres- 

 sure gradient of confined waters. The coarse sediment which floors the 

 canyon should form no impediment. The rate of escape of the water is 

 likely to be so low that a dilution of the overlying sea water cannot be 

 detected. Thus, the benthic fauna may be the best indicator of escaping 

 fresh water. At shallow depths escape is less likely, at least for Hueneme 

 and Redondo canyons, owing to probable sea-water intrusion into aqui- 

 fers produced by artificially lowered water tables of the adjacent land. 



