DISTRIBUTION OP THE IGNEOUS ROCKS 317 



lying rock are extremely infrequent. The chief reliance must be placed 

 on the occasional residual boulders, on the presence or absence of quartz 

 grains in the clay, and on the occasional cut banks along the streams- 

 The large scale sections of the proposed darasites indicate the complexity 

 of the relations between the various volcanic formations and the hope- 

 lessness of attempting to work out these relations from surface indica- 

 tions alone without the aid of sections derived from drilling. For the 

 reasons given above it is practically impossible to map the surface out- 

 crops of these various rock varieties. Their distribution can only be in- 

 dicated in a general way. 



The rock forming the hill on which San Carlos is located consists of 

 augite-andesite. This extends eastward down the river to the Rio Mel- 

 chorita, forming the hills which rise above the level alluvial plains. At 

 Palo de Arco occurs olivine basalt, and this rock continues eastward a 

 short distance beyond Castillo. At the Sabalos it presents an amygda- 

 loidal phase, and in the hill near the mouth of the Santa Cruz it consists 

 of a very coarse breccia. The high hills at the junction of the San Carlos 

 with the San Juan consist of hypersthene basalt. This is a holocrys- 

 talline rock, and one which probably cooled at some distance below the 

 surface. It may possibly mark the center of eruption from which lavas 

 in the surrounding region, which have a similar composition but less 

 perfect crystalline structure, were derived. A similar but less crystalline 

 rock also occurs in the hills on the north side of the San Juan river. To 

 the eastward, at Ochoa, the hypersthene basalt occurs south of the river, 

 while the rocks at the river bank on both sides and extending to the 

 northward are olivine basalts. This olivine basalt extends eastward be- 

 yond the San Francisco hills. At the Tamborgrande it is replaced for a 

 short distance by dacite ; then in the Tamborcito hills by hypersthene 

 basalt, but again comes in in the Sarapiqui hills, and thence extends east, 

 forming all of the hills which border the lower portion of the San Juan 

 river, and also those about Silico lake. The' dacite, while it does not 

 reach the surface at Ochoa, was encountered there in boring. It was also 

 found at lower Ochoa, beneath a bed of volcanic tuff or breccia and some 

 unconsolidated sediments. It comes to the surface at Tamborgrande, 

 and it probably continues north in the high ridge connecting the Tam- 

 borgrande hills with the eastern divide. It forms the surface through 

 the higher portion of the eastern divide, overlying andesitic tuffs and 

 passing under basalt. 



Fragniental igneous rocks. — Since the closely related fragmental rocks, 

 both the bedded tuffs and the conglomerates, do not weather in such a 

 way as to furnish residual boulders, their presence is much more diffi- 



XLVI— Bull. Geol. Soc. Am., Vol. 10, 1898 



