I use the term " analogous," and not " homologous," because the relation called special homology cannot be demonstrated, in any instance, between animals belonging to different sub-kingdoms. The head, the legs, the brain of an articulate, are only function ally the head, legs, or brain. They perform the same function as, but they cannot be shown to be homologous with, the head, legs, or brain of a vertebrate. Indeed I have long held that the sub-kingdoms should be limited by refer ence to special-homology: all animals among which homologies can be pointed out—which all conform to the same type — should be grouped together to form a sub-kingdom.
The view of the analogy of a radiate animal to a vertebrate or articulate animal just given, is considerably strengthened by the manner of development of the common medusa. The larva of the medusa is a polypiform living thing, anchored by one end and tentaculated at the other. This, after a while, becomes marked with numerous constrictions, a segmented animal, which constrictions become more and more deep until they completely divide the quondam .polypiform being into a number of pieces, each of which becomes a perfect medusa. Here is a segmented animal, each of whose joints becomes an independent radiate individual ; ergo, each individual is analogous to one segment of a vertebrate or articulate.
Towards a different view tends the fact, that though five, or a multiple of five, is usually the number of the rays of the star-fish, yet there are some members of the same group the number of whose arms are neither the one nor the other; for instance, there are eleven and twelve-armed asterias. Now there is such perfect constancy in the number of the parts of vertebrate and articulate segments, that that constancy seems an integral element of the idea of an archetype. There is, however, no constancy in the number of the vertebrate or articulate segments that go to constitute an animal, and therefore some may regard each arm of an asteria as analogous to a vertebra ;— the common starfish, as composed of five analogues of vertebral archetypes arranged in a circle, — as a segmented animal bent upon itself, with its anterior and posterior extremities adherent to one another. Which of these two
views is the correct one—whether either is correct can be decided then only when the true import of serial homology and of sym metry is definitely ascertained ; when it is de termined what is due to division, what to radiation, or whether division and radiation, one, or both, or neither, are concerned in their causation.
The skeleton of the Velella presents an in stance of departurelfrom symmetry, mainly in the oblique set of its vertical plate. The common medusa has a circular outline, and exhibits four quarters, which repeat one another exactly, so that it seems to be marked with a right-angled cross. The horizontal plate of the skeleton of Velella is distorted to a sub-rhomboid form, and marked with two diagonal seams that cross one another obliquely. The longer one of these diagonals is produced upwards so as to form the oblique sail-supporting vertical plate, which also pre sents a seam continued up from the point of crossing of the diagonals, indicating that it is composed of two parts uniting at this central point. This de facto unsymmetrical radiate is therefore easily reduced in idea to its pri mordial symmetry.
Infusoria. — Out of the vast number of various forms met with amongst these animal cules the greater number are symmetrical. This symmetry is usually not bilateral, but that of the star-fish. There are, however, some forms which are quite irreconcilable with sym• metry, which no line can divide into two similar halves, which are one, and present no repetitions of parts ; except, indeed, when they are undergoing fissiparous generation.