8 ALLAN HANCOCK PACIFIC EXPEDITIONS VOL. 27 



(4) Many species are represented by peak or high numbers in 

 one or a few samples, from widely dispersed places ; such peak numbers 

 may be correlated with optimum conditions for the species, with 

 gregariousness, or with recent spat-falls of individual species, but not 

 usually with food concentrations. 



(5) The replacement of species from one canyon to the next is 

 such that from 30% to 60% are different. These differences may be 

 partly correlated with latitude, with change in sediments, with dis- 

 tance from shore, or with other, still unknown factors, concerned 

 with the biology of specific entities. 



(6) Replacement of species within a canyon with increasing depth 

 is also abrupt, so that more than 50% of the species may differ from 

 one depth class to the next. These step-down effects are illustrated 

 below. 



Analyses of these canyon samples have resulted in thousands of 

 individual facts relating to the abundance or sparsity of life, to the 

 distribution and numbers of kinds of animals of individual species, 

 and to their occurrences in major phylogenetic groups. It is believed 

 they represent conditions as they exist in nature, but the figures of 

 abundance can be regarded only as minimal since only those unit 

 parts which reach the final stage in processing can be counted. 

 Some of the most conspicuous features are the associations of species in a 

 sort of community structure, suggesting an obligatory relation that 

 transcends systematic categories. Most of the species encountered in the 

 sediments are presumably deposit feeders and perhaps not competitive 

 for food supplies. Predators, such as large nemerteans, ceriantharians, 

 seastars, show spatial relations that space them at more than a station 

 apart. 



The sharpest breaks from one canyon to the next, concerning 

 numbers and kinds, are shown between Monterey and Hueneme can- 

 yon ; then again from Coronado to Santa Cruz canyon, and from 

 Catalina to San Clemente rift valley. In each of these cases there is a 

 discontinuity in geography, or in kind of sediment. 



The screenings (see ANALYSES) resulting after the removal of 

 fine silt from a sample, when washed through screens of varying 

 mesh-sizes aboard ship, give an approximate idea of some of the 

 physical aspects, especially the kinds of dead and non-living fractions 

 in a given volume of sea bottom. Some samples from all of the long- 

 shore canyons have yielded a variety of screened fractions excluding 



