38 SIB. J. A. DOUG-LAS OX GEOLOGICAL SECTIONS [vol. lxxvi r 



stoutly-built forms. This species has also been described from the Argentine 

 by Prof. E. Kayser, 1 and by Ivor Thomas. 2 



Tropidoleptus caejnatus Conrad. 



This sj>ecies has now been shown by Henry S. Williams 3 to recur in three 

 distinct horizons in the Portage and Chemung formations of New York : it 

 can, therefore, no longer be regarded as a distinctive fossil of the Hamilton 

 Shales. 4 



Beyond Cabanillas the Devonian strata dip steadily beneath the 

 broad alluvial plains of Juliaca, which have been left by the gradual 

 recession of the waters of Lake Titicaca, to reappear in the district 

 north of the lake. 



With the view, therefore, of obtaining further insight into the 

 structure of the surrounding country, we decided to take the more 

 direct route to Puno, and proceeded in a south-easterly direction 

 from Las Huertas. Between this point and the hacienda of Yana- 

 rico, the transgressive Cretaceous beds are again seen resting on 

 steeply-dipping Devonian quartzites. The country then becomes 

 covered by large areas of swamp and tracts of alluvium, in such 

 wise that its geological structure is almost entirely obscured. 



Beyond Vilque, however, a new feature appears in the landscape,, 

 in the form of prominent flat-topped hills or 'buttes.' These were 

 found to be due to the dissection of a thick horizontal sheet of 

 lava, an olivine-basalt, which was traced nearly as far as Puno. 

 Such an occurrence is of considerable interest, as being almost the 

 only sign of volcanic activity in this inter-Andean region, com- 

 parable with the great outbursts which have built up the lofty 

 peaks of the Western Cordillera. , 



The vesicular nature of the lava and its wide lateral extent 

 suggest that it was extruded in the form of a fissure-eruption, which 

 spread far over the surface of the surrounding country. Its undis- 

 turbed horizontality, when we consider the post- Cretaceous folding 

 that has affected the district, shows that its formation must be of 

 comparatively recent date. 



No trace of feeders was detected in the immediate neighbour- 

 hood ; but this is not surprising, as the slopes of the hills are 

 everywhere covered by talus and vegetation. 



(A 120) Olivine-basalt. Tiquillaca. 



A highly- vesicular greyish-brown lava showing occasional flakes of biotite. 



Microscopic characters : — In mineral composition it exhibits a marked 

 similarity to the basaltic dykes which have been described above as 

 cutting the Jurassic rocks of Lagunillas, and the microscopic structural 

 differences are only such as can be explained by the different conditions 



1 Zeitschr. Deutsch. Geol. Gesellsch. vol. xlix (1897) p. 288 & pi. xi, figs. 1 & 2. 



2 Ibid. vol. Mi (1905) p. 254 & pi. xii, fig. 19 



3 ' Recurrent Tropidoleptus Zones of the Upper Devonian in New York ' 

 U.S. Geol. Surv. Prof. Paper 79 (1913). 



4 For a comparison with Tropidoleptus fascifer Kayser, see ' El Devoniano 

 de la Argentina Occidental ' by Dr. J. M. Clarke, pp. 6-7 : Anales del Minis- 

 terio de Agricultura, Eepublica Argentina, Seccion Geologia, etc. vol. viii 

 (1912) No. 2. 



