Nothing would injure most of these “small fry” more than a change in the water making it a few degrees colder or warmer than they were accustomed to. Since the constant circulation of the currents keeps the ocean water in all its parts almost precisely of the same density, and food seems about as likely to abound in one district as another, naturalists have concluded that it is temperature which decides the extent of coast or of sea-area where any one kind of invertebrate animal will be found. It thus happens that the life of Cuban waters is different from that of our Carolina coast; and that, again, largely separate from what you will see off New York; while Cape Cod seems to run out as a partition between the shore life south of it and a very different set of shells, sand-worms, and so forth, characteristic of the colder waters to the northward. Out in the ocean, however, the warm current of the Gulf Stream forms a genial pathway along which southern swimming animals, like the wondrously beautiful Portuguese-man-o’-war (Physalia), may wander northward for hundreds of miles beyond where they are found near shore; yet if by chance they stray outside the limits of the warm Gulf Stream, they will at once be chilled to death, as happened once to millions of tile-fish. Ocean currents carry floating burdens long distances. They bring the icebergs to form those terrible fogs of Alaska and Newfoundland; and they often bear far away the logs that float out of tropical rivers. [Illustration: A YOUNG SHIP-RIGGER.] These drifting logs often have plants growing upon them or contain quantities of seeds which are not injured by their short voyages. When, therefore, the coral polyps build up one of their reef-islands until it appears above the waves, thither the currents bring roots and seeds from neighboring islands, and quickly plant them upon the new barren shores, so that in a few seasons the little islet becomes green and wooded and ready to hold its own against the winds and waves. Moreover, the same drifting stuff will carry land animals as passengers,—insects, snails of many kinds, reptiles, and even four-footed beasts,—and so not only give the island a vegetation, but populate it with various of the smaller animals. This seems to you, perhaps, a very accidental and haphazard way of fitting out a country so that presently it may support human beings, nor is it the only means by which barren islands become productive; but it is important as far as it goes, and when we study into the distribution of plants and animals in an archipelago, we are pretty sure to find those of the same sort upon islands that lie in the same current—even to the human inhabitants. CHAPTER III THE BUILDING AND RIGGING OF SHIPS