24 MADEIRA. 
quently tumble off and are caught in small glazed pots, that 
are purposely placed underneath to receive and prevent them 
from doing further mischief This contrivance, however, must 
be too expensive to be general. 
We had scarcely descended the heights and reached the 
village of Santa Cruz, which is situated on the eastern point 
of the island, before it became dark ; and we had still ten 
miles to travel, along the edges of rocky precipices overhang- 
ing the sea-shore, where we had little to trust to beyond the 
cautious steps of the mules. In this part of the journey, one 
of the most brilliant and beautiful meteors I ever beheld 
passed o'^ er our heads, from the body of the island, with a 
rustling noise like that of a sky-rocket, appearing to fall in a 
curvilinear direction into the sea. . The duration of its light 
we conjectured to be from eight to ten seconds. The first 
impression, made by a light so exceedingly brilliant, sug- 
gested the idea of a fresh eruption from the volcanic moun> 
tains we had Just descended, and all e3^es were turned towards 
that quarter. Our troublesome journey, as I observe by my 
notes, was much shortened in philosophizing on this pheno- 
menon. On the principles of the old-fashioned doctrine which 
then (now thirteen years ago) prevailed, it was concluded 
that the electric fluid, passing through contigu:)us or blended 
volumes of oxygen and hydrogen gasses floating in the upper 
regions of the atmosphere, might effect that chemical com- 
bination which is knov/n to produce water; and that, not- 
withstanding the clear starry hemisphere, we might be caught 
in rain before we reached Funchal. This did not exactly 
happen, but several smart showers fell in the course of the 
