Vol. 68.] GEOLOGY OP COSTA RICA. 121 



of about 100 feet. The lower part of the series is almost entirely 

 conglomeratic, and a tunnel connecting bends of the river is driven 

 for 1000 yards entirely through this coarse conglomerate. The 

 boulders are subangular to rounded, and even in the coarsest parts 

 there are distinct traces of horizontal bedding. The boulders are 

 entirely volcanic rocks, chiefly dark, basic porphyritic andesites ; 

 the majority are compact, but a few are highly vesicular. In this 

 basal portion, which is about 50 feet thick, the boulders are 

 practically in contact, with just a little clayey sand filling the 

 interstices between them. Above the conglomerate comes a thick 

 bed of sand, which shows good horizontal bedding and rests upon 

 a very uneven surface of the conglomerate. In places the sand 

 immediately after deposition has evidently been scooped out by 

 strong current-action, and the resulting hollow has been filled up by 

 conglomerate. These finer deposits show slight, but only slight, 

 traces of false-bedding, and they consist of much purer quartz- 

 sand than the matrix of the conglomerates. Numerous fragments 

 of wood are embedded in these accumulations, especially when the 

 arenaceous facies is replaced by a more clayey one. Por example, 

 about half-way down the pipe-line the deposit has almost the con- 

 sistency of Gault, and from this quite large roots and other plant- 

 remains can be obtained; in this locality a large spur of the 

 older volcanic series rises into the conglomerate which rapidly 

 thins out against it. 



In the gorge of the Eio Alajuela, about a mile above its junction 

 with the Rio Grande, the same general structure may be seen, but 

 here the detrital deposits are of an entirely different character. 

 The thickness of the deposits is about the same as at Cebadilla, but 

 boulders are comparatively rare and so far as can be seen perfectly 

 angular, although owing to the steepness of the gorge a close 

 inspection is impossible : the great mass seemed to consist of a fine- 

 grained sandy cla}' which showed only the merest traces of bedding. 

 It certainly appeared to me that in this section the deposits are of 

 the nature of old mudslides rather than true river-deposits, and 

 correspond on a larger scale to similar patches in the ' boulder- 

 clay ' of the Eio Reventazon which are clearly due to landslides : 

 these are referred to in § YI, on the ' Eoulder Clays of Costa Rica ' 

 (see p. 134). 



(3) The lava-flow. — Cebadilla provides another section through 

 the great lava-flow, but petrographically the rocks here are very 

 diflPerent from the typical andesite. The lava forms a great escarp- 

 ment at the top of the Rio-Grande gorge, and the average thickness 

 exposed cannot be less than 150 feet. The upper surface, as usual, 

 forms the floor of the San Jose Valley, and on the surface are many 

 large boulders which probably represent the weathering-out in 

 place of part of the lava. The actual junction of the lava with 

 the underlying sands is exposed on the side of a small track, which 

 leads northwards from the works towards the intake-station. The 

 igneous rock here is very dark and quite compact ; a curious 



