Analysis of Gravity Field— STRANGE, WOOLLARD, and ROSE 
387 
Fig. 4. Mass distribution model and a comparison of the observed and computed free-air gravity anom- 
aly profiles along line A~A! of Figure 3. 
of direct drilling and sampling to remove these 
remaining ambiguities. In any case, such changes 
would not change the basic interpretation of 
the data. 
CONSTRUCTION OF MODELS FOR 
VOLCANIC CENTERS AND RIFT ZONES 
In constructing density models to represent 
the volcanic centers and rift zones, the primary 
constraints are: (a) the magnitude of the 
anomaly, (b) the shape of the anomaly, (c) 
the densities already chosen for the bulk of the 
Ridge with which the densities of the volcanic 
pipes and rift zones are to be contrasted, (d) 
the range of geologically permissible ' densities 
for the intrusive material, and (e) the seismic 
evidence as to dimensions of the intrusive 
bodies. Despite this rather impressive array of 
restrictions, much latitude still remains as to 
choice of density model. As the Koolau vol- 
canic center and rift zone have been more 
extensively studied gravimetrically and seismi- 
cally than most of the other intrusive com- 
plexes, these features were chosen for the 
analysis. In addition, the gravity anomaly asso- 
ciated with the Koolau volcanic center is one 
of the larger anomalies and the center lies 
essentially at sea level. There are, therefore, 
fewer uncertainties in the gravity anomaly pic- 
ture and more restraints on the mass distribu- 
tion model constructed to represent this in- 
trusive complex than on any of the other 
intrusive complexes. Thus, several models which 
would satisfy all of the known parameters as- 
sociated with the Koolau volcanic center and 
