. This sp 爬 ading should somehow be accomodated . Moreover , the stress 丘eld should remain the same after aceomOdation in order for a new dike to intrude in the same orientation . The current spreading episode in 恥 eland ( BJoRNssoN et α 1. , 1979)between North Ameri − can and European plates revealed that the sequence of events in the spreading pr ess is similar to that observed for Hawaiian volcanic activities . This implies that the process of pla 七 e separation and aceretion is nothing but

more »

... he activity of rift zones ・ Constructional plate boundaries may be regarded as compQsed of a chain of rift zones and associated feed − ing POIygenetiC CenterS . Room necesgary for repeated dike intrusion is supplied in the case ef spreading centers , by the late瓰 1 motion ( separation ) of lithosphere over asthenosphere 。 In the case of Hawaii , sliding of the volcanic edifice over a deep sea sediment layer may be the analogous meehanisrn such as appears to have occurred during the 1975 Kalapana earthquake , as studied by ANDo ( 1979}and FuRuMoTo and KovAGH ( 1979) . Kalapana earthquake had been anticipated by SwANsoN et α t.( 1976)as one of the repeated steps as the east rift zone has eontinuously dilated . Thu6 , the pri 皿 ary cause for the long, well developed rift zones of Hawaiian volcanoes may be in the existence of thick enough oceanic sedimenbS serving as a potential sliding plane beneath the volcanic edifices . Lack of rift zones in Galapagos shields which grew over the young ocean floor with rough topography is consistent with this view .