If it be asked why the blocks of each train are not more scattered, especially when far from their source, it may be observed that after passing through sounds separating islands, they issued again from a new and narrow starting-point; moreover, we must not exaggerate the regularity of the trains, as their width is sometimes twice as great in one place in as another; and Number 6 sends off a branch at p, which joins Number 5. There are also stragglers, or large blocks here and there in the spaces between the two trains. As to the distance to which any given block would be carried, that must have depended on a variety of circumstances; such as the strength of the current, the direction of the wind, the weight of the block or the quantity and draught of the ice attached to it. The smaller fragments would, on the whole, have the best chance of going farthest; because, in the first place, they were more numerous, and then, being lighter, they required less ice to float them, and would not ground so readily on shoals, or if stranded, would be more easily started again on their travels. Many of the blocks, which at first sight seem to consist of single masses, are found when examined to be made up of two, three, or more pieces divided by natural joints. In the case of a second removal by ice, one or more portions would become detached and be drifted to different points further on. Whenever this happened, the original size would be lessened, and the angularity of the block previously worn by the breakers would be restored, and this tendency to split may explain why some of the far-transported fragments remain very angular. These various considerations may also account for the fact that the average size of the blocks of all the seven trains laid down on the plan, Figure 50, lessens sensibly in proportion as we recede from the principal points of departure of particular kinds of erratics, yet not with any regularity, a huge block now and then recurring when the rest of the train consists of smaller ones.