The motion of the blood is the more per ceptible as it is full of minute white globules, which pass through the minute vessels in a sin gle row, and are easily seen through the trans parent parietes. These observations may be still more assisted by holding the animal vertically, with the nucleus downwards ; when, as the blood, driven into the ventral sinus, is forced to ascend against its own weight, its current is less rapid, and the movements of the blood-discs can be more easily followed. "As these blood globules are of a certain consistence, a resist ance," says Van Hasselt, "is necessarily cre ated in the whole mass of the blood, which ultimately overcomes the projectile power of the heart. After a short interval, during which the opposite forces are balanced, the heart assumes a spiral movement, contrary to the preceding. Hence it follows," he ob serves, " that, since the blood is driven as much backward as forward in the vessels of the body, and since it is only by their anas tomoses that the circulation can be said to be carried on, all the system of the pulmonary vessels can consist only in accessory ramifica tions, which have no direct influence on the principal circulation ; and that two separate systems, arterial and venous, do not exist, the two being united, or rather never having been separate."* The nerve.ganglion is more or less de veloped in all the Salpians. It is situated on the superior surface of the thoracic chamber, near the posterior extremity, and just above the insertion of the branchia. In its neigh bourhood is often observed a pigmentary spot, or rudimentary organ of vision.
In S. cordifornzis the ganglion appears as a compound body, formed of two central glo bular portions and four irregular lateral ap pendages, two on a side. Numerous filaments radiate off in every direction. Four of these filaments are sent towards the posterior ori fice, two of them passing on either side of the space described by Prof. Eschricht as the " oval organ." This latter is a long slit with an inner smooth rim, and an outer oval, trans versely striated border. In S. zonaria the gan glion is a somewhat triangular, globose body, with a single appendix on each side, sending off four anterior filaments, two lateral pairs and a posterior pair. The " oval organ," situ ated between the last pair, is close up to the ganglion, and less distinct than in S.cordifornzis.
Many of the Salpe are highly phosphores cent 1- ; but sometimes this condition is due to the presence of phosphorescent Crusta ceans inhabiting their internal cavity. The long chains of phosphorescent Salyer swim ming near the surface, have been described as occasionally producing the effect of long ribands of fire drawn along by the currents.
Generation of Salpce. —" One of the most important discoveries," says M. Krohnt,"with which physiology has been enriched in our days, is without contradiction that of the remarkable phenomena observed in many inferior animals, and termed, propagation by alternate generations.'" To M. Steenstrup# the merit is incontestably due, of having been the first to bring together, in an ingenious man ner, all the facts belonging to the subject. In
treating of the phenomena observed by Cha misso in one species of Salpa, and analogous phenomena studied in other groups of animals, and in his generalisations of the facts, M. Steen strup has placed them in a strong light, and given them their full value. M. Steenstrup's observations have also tended to strengthen the views of Chamisso 1- so often contested ; and the researches of M. Krohn, on the natural history and zootomy of eight species of Salpw, fully confirm them.t Every Salpian is viviparous ; and each species, as shown by Chamisso, is propagated by an alternate suc cession of dissimilar generations. One of these generations is represented by solitary or isolated individuals ; the other by aggregated or associated individuals, united into groups known as " chains." Each isolated individual engenders a group of aggregated individuals, and each of these produces in its turn a solitary individual. The isolated individuals are therefore multiparous, and the associated individuals uniparous. This is not the only difference existing between the two alternating generations ; for, if we compare the associated individuals, representing the aggregate gener ation, with the solitary individual forming the isolated generation, we shall find that they differ amongst themselves not only with respect to external conformation, but also in many particulars of organisation.
The definition, then, of species should in this case include the characteristics of the two dissimilar generations, isolated and aggregate, which alternately succeed each other. It is desirable therefore, for the determination of each species, to preserve but one name. This must comprehend the two dissimilar states which have hitherto been falsely considered as two distinct species, and to which have been given Specific names, that can now serve only to designate one or the other of the hetero morphous conditions.* Amongst the characters that distinguish the different generations of each species of Salpa, one of the most remarkable is offered by the disposition of the muscular bands. These are variable in the two generations, but constant for each of them. There is, however, a still more essential difference, and that is, the mode of propagation peculiar to each generation. The aggregate individuals proceeding from the isolated individuals grow by gemmation within the mother animals, on a cylindrical prolongation, which may justly be termed a proliferoust stolon, but which differs from stolons observed in many other animals, in not ever appearing externally. The stolons of the " social " Ascidians spring up bare, the animal not being connected by a common gelatinous envelope, as is the case with the " compound" Ascidians ; in which latter group the stolons necessarily remain hidden in the common envelope, somewhat approaching in this respect the stolons of the Salpians.