Mount Etna is currently characterised by intense effusive and explosive act
ivity of its summit vents, whereas 319 non-active Holocene flank cones are
spread across its flanks at altitudes of between 2990 and 475 m. In volcani
c areas the relationships between fracture occurrence and cone growth/locat
ion are well established. With this in mind, the spatial distribution of th
e Mount Etna flank cones was analysed in order to make some inferences abou
t the fracture systems that feed the cones. The positions of the flank cone
s were acquired by the use of a digital elevation model of the volcano with
a geometric resolution of 10x10 m. Spatial distribution of the cones was a
nalysed through counting-box and sand-box methods, checking for fractal or
multifractal behaviours. The four data sets analysed consist of the whole n
umber of parasitic cones (319), and cones located on the NE (50), south (14
3) and west (61) rifts, respectively. The cones have a non-trivial power-la
w distribution. The sand-box method gave the best results with a fractal ex
ponent D-f for all cones of 1.41+/-0.02 over the length range 0.2-10 km. Th
e same analysis was performed on the other data sets: South Rift (1.42+/-0.
02); West Rift (1.39+/-0.02); and NE Rift (1.43+/-0.02), The cones do not h
ave a multifractal distribution, as suggested also by the strong similarity
between fractal exponents of the different data sets. Data suggest a stron
g control over flank cone distribution by fracture length and density. Thes
e two characteristics are, in turn, expressions of highly connected fractur
es activated as magma feeders by the volcano's present stress field. We int
erpret the rifts as inherited structures that represent preferential sites
of fracture connectivity.