to plant them here and there in good soil at considerable distances from where any have heretofore been grown. "For a time enemies do not find them." Pear trees planted scatteringly are more likely to remain healthy than in orchards. "Perhaps one reason why plants have become extinct or nearly so is their lack of means of migration. As animals starve out in certain seasons when food is scarce, or more likely migrate to regions which can afford food, so plants desert wornout land and seek fresh fields. As animals retreat to secluded and isolated spots to escape their enemies, so many plants accomplish the same thing by finding the best places with some of their seeds sown in many regions. Frequent rotations seem to be the rule for many plants when left to themselves in a state of nature. Confining to a permanent spot invites parasites and other enemies and a depleted soil, while health and vigor are secured by frequent migrations."
Commensals.—Curious associations are formed among animals for mutual aid in the struggle for existence. Some of them are societies of the same species, like those of ants and bees; colonies in which many individuals—as ascidians and bryozoa—join into a single mass and act as one; and associations of animals of different species constituting commensalism where both are benefited, or parasitism, when the advantage accrues to only one of the parties. The hermit crab and certain ascidians furnish very fine examples of commensalism. The hermit crab is known as an inhabitant of shells bereft of their proper owners. Some sea anemones also fasten themselves on shells, and seem to prefer those which have been adopted by hermit crabs. The association is shown by M. Henri Coupan, in La Nature, to be one of mutual benefit. The actinia defends the crab and its home against all intruders by means of its tentacles—veritable batteries of prickly stings; while the crab, with its long claws reaching out to catch whatever is good to eat, brings food within reach of the ascidian. Mr. Percival Wright, having taken the crab from a shell to which an ascidian had attached itself, found that the latter abandoned the shell in a short time. M. L. Faunt reversed the experiment, taking the ascidian away, when the crab deserted its quarters, found a shell with the ascidian on it, and occupied it very quickly. He further observed the maneuvers executed by the crab to secure the attachment of an ascidian to its shell. Sometimes a large ascidian will wholly cover a shell; or several smaller ones will spread themselves over the same shell so as to form a continuous envelope over it. The ascidians become so attached to their commensals as to seem unable to live without them, and even to die soon after being separated from them.
Drift of Ocean Currents.—Of sixteen hundred and seventy-five floats bearing requests to the finder to return them which Prince Albert of Monaco dropped into the Atlantic during three research cruises, with a view to learning something of the movements of surface currents, two hundred and twenty-six were returned to him up to the year 1892. By working the course which each of them had probably been following, the prince undertook to draw a definite map of the currents. As the elements employed were always numerous for each region, he thinks his results were near the truth in its general lines. The floats landed on almost all the shores of the North Atlantic, from the North Cape to the south of Morocco, along Central America, and on the islands of Canaries, Madeira, Azores, Antilles, Bermudas, Shetland, Hebrides, Orkneys, and Iceland. None appeared as far south as the Cape Verd Islands. The drifts seem to indicate an immense vortex, beginning toward the Antilles and Central America with the Gulf Stream and the equatorial current; passing the Banks of Newfoundland at a tangent, it turns to the east, approaches the European coasts, and runs southward from the English Channel to Gibraltar, after having sent a branch running along the coast of Ireland and the coast of Norway as far as the North Cape. It then returns to the west, encircling the Canaries. Its center oscillates somewhere to the southwest of the Azores. The author's observations enabled him also to establish a very good average for the speed at which these floats traveled in the different sections of the vortex, and for every twenty-four hours: Between the Azores, France, Portugal, and the Canaries, it was 5.18 miles; from the Canaries to the Antilles, the Baha-
