When two volcanic cones are thrown up on the same line of fissure, their full development is interfered with, and irregularities in their form and characters are the consequence. In the plan (fig. 82) and the section (fig. 83) an example is given of the results of such a shifting of the centre of eruption along a line of fissure. By the second outburst, one-half of the first-formed cone has been removed, and the second-formed overlaps the first. [Illustration: Fig. 83.--Ideal section of the Puy de Pariou.] Sometimes a number of scoria- or tuff-cones are thrown up in such close proximity to one another along a line of fissure, that they merge into a long irregular heap on the summit of which a number of distinct craters can be traced. An example of this kind was furnished by the line of scoria-cones formed above the fissure which opened on the flanks of Etna in 1865 (see fig. 84). [Illustration: Fig. 84.--Fissure formed on the flanks of Etna during the eruption of 1865. _a._ Monte Frumento, an old parasitic cone. _b._ Line of fissure. _c, c, c._ New scoria-cones thrown up on line of fissure. _d._ Lava from same.] [Illustration: Fig. 85.--Plan of the Island of Vulcano, based on the map of the Italian Government.] [Sidenote: SHIFTING OF ERUPTIONS ALONG FISSURES.] Even in the case of great composite cones, however, we sometimes find proofs of the centre of eruption having shifted its place along the line of fissure. No better example of this kind could possibly be adduced than that of the Island of Vulcano, with the peninsula of Vulcanello, which is joined to it by a narrow isthmus (see the map, fig. 81, p 192). In fig. 85 we have given an enlarged plan of this island which will make its peculiar structure more intelligible (see also the section given in fig. 77, No. 6, facing p. 178).