534 HERSEY AND BACKUS [CHAP. 13 



The scattering cross-section of euphausids, and other such small animals 

 without air-bladders, is sufficiently low as to make it highly improbable that 

 they are the scattering agents in deep scattering layers (as we have argued 

 earlier). It is only in rare cases that they occur with the necessary density. The 

 larger air-bladderless animals mentioned (large crustaceans, squids, fishes and 

 others), as well as animals with air-bladders, all have scattering cross-sections 

 sufficiently large to make the population requirements consonant with observed 

 or, at least, probable populations. This is to say that observations of volume 

 back-scattering coefficient (m) are not helpful in deciding which of these larger 

 animals comprise deep scattering layers. 



Even though the data are imperfectly analyzed, spectrum analyses of broad 

 band scattering at a number of points over the western North Atlantic show 

 that scattering layers there do have resonant peaks, with one peak to each 

 1 yer in the frequency range examined (1-30 kc/s). This indicates, but does not 

 finally demonstrate, that the principal constituents of these layers are fishes 

 with gas-filled swim-bladders. Positive identification of the scatterers will have 

 to be accomplished by a combination of acoustical and other observational 

 techniques in which the animal or its image is captured during the act of sound- 

 scattering, 



7. Ideas and Miscellaneous Observations 



A. Submarine Illumination and the Vertical Migration 



Two studies have been made in which the movements of deep scattering 

 layers during their vertical migration have been compared with direct measure- 

 ments of submarine illumination. In one of these studies scattering layers over 

 the San Diego trough were observed with a 17.5 kc/s echo-sounder and with 

 explosive sound sources, while light measurements were made with a submarine 

 photometer. In this study the layers exactly followed the movements of 

 certain isolumes save for a lag at the beginning of the evening ascent (Kampa 

 and Boden, 1954). Somewhat different results were obtained in a similar study 

 using a submarine photometer and a 12 kc/s echo-sounder in deep water off 

 the coast of southern New England. A scattering layer, whose midday depth 

 was about 380 m, split into two elements near sunset, both of which moved 

 surface wards more rapidly than certain isolumes, and during their migrations 

 reached levels that were illuminated 100 times more brightly than their midday 

 level (Clarke and Backus, 1956). 



B. Water Temperature and Deep Scattering Layers 



That the midday level of deep scattering layers cannot be correlated with 

 any conspicuous feature of the temperature-depth profile has already been 

 noted. However, Moore (1950) observed during hours of daylight shoalings 

 and deepenings of a principal deep scattering layer, which accompanied 

 shoalings and deepenings of certain isotherms as a ship passed through cold 



