Ch. XXIX.] LAVAS OF MADEIRA. (353 



of lavas the older rocks may have been more or less disturbed and 

 tilted, without destroying the general form of the old dome-shaped 

 mountain supposed by us to have been the result of repeated erup- 

 tions from the ceutral vents. 



The locality just referred to of Porto da Cruz exemplifies not only 

 the long intervals of time which separated the outflowing of distinct 

 sets of lavas, but also the precedence of the basaltic to the trachytic 

 outpourings. So also on the southern slope of Madeira, we observed 

 between the Jardim and Pico Bodes, situated in a direct line about 6 

 miles northwest of Fimchal, a well-marked series of trachytic rocks 

 of considerable thickness occupying the highest geological position. 

 They consist of white and gray trachytes, occurring at points varying 

 from 2500 to 3500 feet above the sea. Their position may be under- 

 stood by supposing them to constitute the uppermost beds repre- 

 sented at h in the section (fig. 706, p. 648), and on the slope above 

 h. The doctrine, therefore, that in each series of volcanic eruptions 

 the trachytic lavas flow out first, and after them the basaltic (see p. 

 657), is by no means borue out in Madeira, although some of the 

 newest currents, like those at the foot of the cones m, n, o (fig. 706), 

 are basaltic. 



Several of the latest and most powerful streams of lava which 

 have issued from the central axis of Madeira are composed of a 

 felspathic rock of a mixed character, on the whole more trachytic 

 than basaltic. It divides into spheroidal masses, often several feet in 

 diameter, which are very conspicuous when the contained iron has 

 become more highly oxidated. M. Delesse, who had the kindness to 

 analyze for me several. of our specimens, found certain varieties of 

 this rock to be without augite, and simply a mixture of blackish- 

 green felspar with olivine. These would, according to him, be 

 classed by most of the French geologists under the general designa- 

 tion of basalt. Whatever name we assign to this product it indi- 

 cates a change in the mineral nature of the materials last emitted 

 from the central axis. Where the island is narrow this spheroidal 

 trap often reaches the sea, but in the broadest and loftiest part of 

 Madeira it forms a superficial envelope, which extends for a certain 

 distance only from the central heights, as, for example, to near o (fig. 

 706, p. 648). Hence, near Funchal, we must ascend to a height of 

 1100 or 1200 feet before we meet with this felspathic formation, the 

 lower grounds along the coast being occupied by true basalts, which 

 never exhibit a spheroidal structure. 



Among other contrasts of character in the superficial volcanic for- 

 mations of Madeira, I may remark that many of the central peaks, 

 such as a, fig. 706, seem to be the mere skeletons of cones of erup- 

 tion, whereas other cones of like origin, such 'as m, x, o, met with at 

 lower levels and nearer the sea, are more regular, and have no pro- 

 truding dikes on their summits or flanks. This difference in form 

 may imply that the more degraded hills are of higher antiquity ; but 



