because ths considerable significance of even the s.nallest varia- 

 tions (in the study of circulation) and the- unexpectedly complex 

 regional inequalities that have actually been found to exist, ual:e 

 it less safe to deduce the salinity of intervening sectors of 'later 

 from T/idely separated observing stations. 



Ilodern standards of research require siciultaneous determination 

 both of temperature and of salinity fror, the surface down to the 

 bottom, if an oceanographic observing station is to be classed as 

 coiTOlete, the tirae having long passed when a surface and a bottor: 

 reading were thouglit sufficient. And in the ocean basins, (with few 

 exceptions) these requirements have been met only by the major 

 deep sea e:cpeditions, whose tracks, reasonably closed-meshed in 

 the Atlantics, have covered the Pacific and Indian basins (and 

 especially the Southern Ocean) with only a very sparse web indeed. 

 Even in the Atlantic, north of 20° N. latitude, only about twenty 

 such coi:ipl3te serial determinations of salinity coiroined with terv- 

 perature had been published for depths greater than 1500 fathoms up 

 to February 1, 1923, a depth-limit including the greater part of the 

 basin in question. The case is far worse for the Pacific, where 

 only thirty-one complete observations deeper than 500 fathoms had 

 been published up to that date for all the vast area from the 

 American coastline westward to longitude 180°; only 35 so deep for 

 the entire Pacific basin; and only seven deeper than 1500 fathoais. 

 And while a large number of serials deeper than 1000 meters_ ( since 

 published) have been taken in the eastern margin of the Pacific 

 within the last few years by the U.S. Bureau of Fisheries, and by 

 the U.S. Coast and Geodetic Survey, these have all been locat3d close 

 in to the A;:ierican coast, or around the Hawaiian Archipelago, Up to 

 1928, a dozen stations, extending out from the coast of Chile, g-ave 

 the only accurate data as to the salinity of the bottom of th3 

 Southern Pacific on the American side. 



The vmole southeastern part of that ocean was, therefore, nearly 

 virgin ground with respect to its abyssal cali.-ities until crossed 

 by the "Carnegie" in 192&-1929; its northeastern abyss hardly lees so. 

 On the western side we find another vast blank, so far as the .salin- 

 ity of the deep v;ater (for that matter, all except the surfaca) is 

 concerned, extending westward from the longitude of the Ha.waiio/n 

 Archipelago nearly to the Tuscarora deep off Japan, and to th3 basin 

 between thj Japan and the Philippines, whore a considerable nuiber 

 of deep records of salinity, as well as of temperature, have boc"u 

 obtained. This expanse, thousands of miles in extent, is crossed 

 (cast and wrist) by only one line of five observing stations at about 

 the latitude of Sa.n Francisco, In fact, along only one meridional 

 profile, falling between longitudes 140° Vk\ and 160° W. arc data 

 available for reconstructing the sub-surface salinity of the central 

 basin of the Pacific; even for this we must turn back to the ob- 

 servations taken by the "Challenger" half a century ago.^ Not a 



1, Existing data have also allowed the construction of such a 

 profile around the eastern margin of the Pacific. 



single really deep salinity-determination has yet been obtained in 

 the Antarctic extension of the Pacific. 



