destroy the nervous connection between the arms, for instance, by making two incisions at a and b, Fig. 18, this coöperation of the arms ceases. The normal starfish requires but a few minutes to turn over, but the specimen represented in Fig. 18 remained on its

FIG. 18. THE SAME EXPERIMENT ON A STARFISH WHOSE NERVE-RING HAS BEEN SEVERED IN TWO PLACES (a AND 6).

The right and left arms are consequently no longer connected nervously. If such an animal is laid on its back, the tube-feet of four or even all the arms in most cases tug simultaneously. This prevents the animal from righting itself,

back the whole afternoon, although the arms were struggling constantly to right it. The experiments seem to indicate that in a normal starfish the stimulus produced by the pulling of two or three arms in the same direction has an inhibitory effect on the other arms. This inhibition ceases when the nervous connection between the single arms is broken. Romanes

found that a single arm containing only the peripheral arm-nerve rights itself when laid on its back. Hence the central nerve-ring acts only as a conductor and not as a "centre" for this reaction (1).

2. In analysing this righting reflex of the starfish, there are two possibilities to be considered. Either gravity forces the starfish to turn the ventral side toward the centre of the earth, or contact-irritability, i. e., stereotropism, forces the animal to bring the ventral side in contact with solid bodies. The fact that the animals leave the horizontal bottom of an aquarium and attach themselves to the vertical sides shows that gravity is not the cause. Preyer made an experiment from which he concluded that the righting of the starfish is due to their being forced to have the ventral side down. He suspended a starfish in the middle of the aquarium by fastening each of its arms by threads to a cork that floated on the surface of the aquarium. If suspended with its back down, Preyer noticed that the starfish turned over. This might suggest the idea that the righting of the starfish is a geotropic phenomenon. I have repeated Preyer's experiment and confirmed his observation (2). At the same time, however, I made a control experiment which Preyer omitted. omitted. In the beginning I fastened the starfish to the cork-plate in such a way that the ventral side was turned toward the bottom. But the starfish even then turned over. This shows that the suspension makes it restless and causes it to perform all sorts of turning movements. I believe

that the ventral side of the starfish is positively stereotropic, or, in other words, that the starfish becomes restless if its ambulacral feet are not in contact with solid bodies.

3. Preyer accredits the starfish with possessing "intelligence." He placed one arm of an Ophiuris in a piece of rubber tubing in order to see whether the animal would be clever enough to rid itself of this impediment to its movements. He found that after

a time the arm "freed itself " from the tube. I have repeated the experiment in these animals and found. that the Ophiuris pays no attention to the rubber tube. The animal of course loses it after a time unless it fits too closely, but it is always purely a matter of chance, and there is no more intelligence involved than the clothes-line displays when the clothes are blown from it by the wind. Romanes found that when one arm of a starfish is stimulated the animal moves in a direction opposite to the stimulated arm. This also looks like intelligence, for the animal seems to be able to avoid a danger. The late Professor Norman called my attention to the fact that when one arm of a starfish is stimulated the feet of this arm are drawn in and the arm becomes inactive. This is, however, only true of the stimulated arm; the others remain active. Therefore, according to the parallelogram of forces, a movement away from the point of stimulation will take place. Intelligence plays no part in this phenomenon.

4. The tendency to crawl upwards on vertical

surfaces is a pronounced reaction of Echinoderms and is quite common in other animals, for instance in the Actinia mesembryanthemum of the Mediterranean, and in the Coccinelli. This tendency is also present in plant-organisms -for example, Plasmodia,- and here Sachs has traced it back to negative geotropism. I will repeat here the description which I have already given in a former publication of the phenomenon as it appears among Echinoderms (3).

No one who observes the animals on rocks or posts near the surface of the ocean when the water is quiet can fail to notice the relatively large number of Echinoderms. Many of these-for example, the Cucumaria cucumis, which is very common in the Bay of Naples-always live near the surface, not beyond a depth of about 30 m. It can be shown that Cucumaria, like Plasmodia or Coccinelli, are forced, when on vertical surfaces, to crawl upward. Cucumaria has a slender pentagonal body, 10 cm. or more in length, with radial, branching tentacles on its oral end. There are five ridges on the body, and in these are situated longitudinal rows of tube-feet, by means of which the animal crawls upward, even on smooth glass walls. If placed in an aquarium, it crawls about on the bottom until it comes to a vertical side; it then crawls upward and remains on the highest point, if possible just below the surface of the water. This position then usually becomes permanent, and the animal is converted into a sessile organism.

If a Cucumaria is allowed to attach itself to a

B

vertical glass which can be revolved around a horizontal axis in the aquarium, it will crawl upward whenever the glass is turned. This is not a compensatory movement produced by the centrifugal force, for during the rotation of the glass the animal re

FIG. 19. GEOTROPIC REACTION OF CUCUMARIA CUCUMIS.

The animals are in a battery jar (a, b, c, d). It is filled with water and rests on the bridge B B in the aquarium A A. Running water is supplied through the tube g at o. They collect at the highest point (c, d) of the glass.

mains quiet, and not until a quarter or half an hour after the rotation does it begin to migrate upward. Neither is the upward migration caused by the light falling in from above. If the animals are placed in an aquarium in which light is allowed to enter only from the side or from below, they will still crawl upward on the vertical sides. In a dark room they behave just as they do in the light.

One might believe that the need of oxygen determined the upward migration of the Cucumariæ to the