90 



difficulty of trsicins, across th^ oceans, thj sm^ll thirmel vnri?tions 

 that hsve been recorded in th>- tropics, from the usu«l records 

 supplied by passing ships, Wc may instance the Caribbean Sea v/here 

 data tabulated Dy the United States Weather Bureau for the nine years 

 1920-1389 showea a maximum monthly departure of 1.2° F from the mean; 

 with only 39 months of the 108 showing deviations greater than 0.5'^ F. 



The crux of the matter is, however, to establish whether or in 

 what parts of the ocean, temperature-abnormalities or other changes 

 in the water do actually anti'date alterations in the weather of the 

 over-lying air. Nor can any general rule be assumed to apply in this 

 respect whether regionally or seasonally, the whole question being an 

 extriraely complex one. In the Gulf of Maine, to note a simple example, 

 it is sufficiently d -monstrattd that the tem.pcrature and direction of 

 the wind largely control the temperature of the wat^r in winter. 

 However, the subsequent effects on New England weathrr of these 

 weather-produced wat . r temperatures are unknown. 



Off Southern California again, the wind affects the temperature 

 of the surface both by producing upswelling from below, and by 

 sweeping cold watrT down from the North. How these Temperatures 

 react on the T-mp.-rature of the air, and so on the weather, is now 

 the subject of active investigation at the Scripp's Institution. 

 In most cases, in short, the sequenc.^ is not clear, even for regions 

 where sea and air temperatures have been under observation for many 

 years. In Scandinavia, for example, it has often oeen stated that 

 various atmospheric and terrestrial phenomena follow the cycle of s-^a 

 temperature. But recent students have found the sequence to be the 

 reverse, for while a close correlation exists between air and water 

 temperatures along the coast of Norway, it now s-ems that the varia- 

 tions in air temperature precede those in the water. Nevertheless, 

 this does not necessarily indicate that the atmospheric changes are 

 the primary on-;s> for the Kore mobile air may bring departures in 

 Temperature to a given coast more rapidly than the possibly activating 

 warmer or colder watir can come. 



This uncertainty as to the true sequence applies not only to the 

 states regularly prevailing over one part of the sea or another, but 

 even to sporadic events that have often bjen invoked as evidence of 

 the climatic effects of marine abnormalities; to the torrential 

 rains, for instance, that accompanied the abnormal development of the 

 warm "El Nirlo" current along the coasts of Ecuador and of northern 

 Peru early in 1925. Although most, if not all, stud-^nts who have 

 published accounts of this uVcnt, have looked to the high temperature 

 of the seas as the cause of the exceptional rainfall that attended, 

 it has been pointed to us that no definite proof of this has yet been 

 brought out, but that while the alteration of ocean currents in the 

 regions were probably a contriouting factor, it is also likely that 

 both events were coincident results of a marked reduction in the 

 strength of the trade winds. 



Uncertainty of another sort as to which is cause, which effect, 

 is illustrated in the North Atlantic where recent and very searching 

 investigations point to the direction of the wind as the cause of 

 variations in the winter temperature of the surface of the sea, but 

 where the winds in turn reflect the locations and intensities of the 

 pGrm.anent or semipermanent centers of high and low atmospheric 



