analogous reflex in a frog or turtle are of a more complicated character. Nature works with very simple tools. The tool employed in the reflex of localisation is the curvature produced by stimulation,-contact, for instance. We meet with this in its simplest form in plants, in which the side that comes in contact with a solid body becomes concave. Plants certainly possess no central nervous system containing mysterious reflex structures. In their case, irritability and conductibility suffice as an explanation. In Medusa the method appears more complicated only in so far as in them the contractile tissue is real muscle-fibre. In the frog, the only further complication is the fact that the conduction takes place through a special kind of tissue-namely, nerve-tissue. In its first anlage, this central nervous system is of a very simple segmental character. I believe that the central nervous system preserves this simple character better than any other tissue. The muscles undergo considerable displacement during the development, but the changes occurring in the central nervous system by no means equal those occurring in the muscular system.

It seems thus possible to explain the above-mentioned phenomena of coördination in Medusa by means of the simple facts of irritability and conductivity without attributing any other functions to the ganglion-cell except those which occur in all conducting protoplasm.

BIBLIOGRAPHY

1. O. u. R. HERTWIG. Das Nervensystem und die Sinnesorgane der Medusen.

2. ROMANES, G. J. Fellyfish, Starfish and Sea Urchins. The International Science Series, 1893.

3. LOEB, J. On the Different Effects of Ions upon Myogenic and Neurogenic Rhythmical Contractions, etc., American Fournal of Physiology, vol. iii., 1900.

4. LOEB, J. Ueber Ionen, welche rhythmische Zuckungen der Skelettmuskeln hervorrufen. Festschrift für Fick. Braunschweig, 1899.

5. QUINCKE.

Ueber periodische Ausbreitung an Flüssigkeitsoberflächen, etc. Sitzungsberichte der Berliner Akademie der Wissensch., 1888, ii., S. 791.

6. PORTER, W. T. The Coordination of the Ventricle. American Journal of Physiology, vol. ii., 1899.

The

CHAPTER III

THE CENTRAL NERVOUS SYSTEM OF ASCIDIANS AND ITS SIGNIFICANCE IN THE MECHANISM OF REFLEXES

1. If we wished to observe the order of the natural system in this book, we should not let the Ascidians follow the Medusa. We consider it more profitable, however, to discuss simple cases before taking up the more complicated ones. Having reached the conclusion, at the end of the preceding chapter, that the spontaneous coördinated activities in Medusæ are not due to specific morphological structures of the ganglion-cells, we will now attempt to find out whether the reflex actions of animals depend upon the structure of the central nervous system or of the peripheral parts. In Ascidians the central nervous system consists of a single ganglion (d, Fig. 9). This ganglion is situated between the oral and aboral tubes (a and b, Fig. 9).

Ciona intestinalis (Fig. 9), a large, transparent Ascidian, possesses a very characteristic reflex. If either the oral or aboral opening be touched, both openings close, and the whole animal contracts so

that it becomes small and round. This reflex is de termined by two groups of muscles, first by ringmuscles in the oral and aboral openings, second by

FIG. 9. CIONA INTESTINALIS.

a oral, o aboral opening; c, foot, &, location of

ganglion.

longitudinal muscles, which run lengthwise through the animal. By the contraction of these muscles the animal is protected from the entrance of foreign bodies into the body cavity. This reaction is a typical reflex act, and is eminently purposeful. According to the prevailing ideas concerning the decisive rôle that the ganglion plays in reflexes, the procedure is as follows:

If the oral or aboral

opening be touched, the stimulation is conducted through the peripheral nerves to the ganglion, where a mysterious reflex mechanism is brought into play, which gives the muscles the command to contract in a manner corresponding to the nature of the stimulus. Ferrier, for instance, in his text-book, mentions the one ganglion of the Ascidians as illustrative of the significance of the ganglion in reflexes.

I removed the ganglion from a number of Cionæ.

For some time after the operation, in most cases for about twenty-four hours, the animals remained contracted. At the end of this period they began to relax again. To my great surprise, I found that the typical reflex continued. If we let a drop of water fall on such an animal, the typical reflex act is produced just as in the normal animal. Hence the reflex cannot be determined by specific structures of the ganglion. But what does determine the reflexes, and what is the function of the ganglion?

The answer to the first question must be that the reflex is determined by the structure and arrangement of the peripheral parts, especially the muscles. The mechanical stimulus throws the muscles directly into activity, and the stimulation is transmitted from muscle-element to muscle-element directly, as in the heart or the ureter. But is the central nervous system superfluous in this animal? We get the answer to this question if we determine the threshold of stimulation. The threshold of stimulation for producing this reflex is higher in animals which have been operated upon than in normal animals. As the source of the stimulus, I used the kinetic energy of drops of water, which fell from a pipette upon the animal. Since the weight of the falling drop in the pipette is always the same, the minimum of the height from which a falling drop can produce a contraction is a convenient measure of the irritability; the latter is of course equal to the reciprocal value of the threshold of stimulation. In one case there were in an