The larva of Asterias must find firm rock or at any rate sea weed to which to attach itself, but there is another British star fish, Astropecten, which habitually lives on sandy ground. When the bipinnaria of this starfish reaches the age at which it should attach itself, this is impossible; so it continues to swim until the locomotor organs of the future starfish are so far developed that they are capable of functioning. Then the great ciliated preoral lobe is suddenly amputated ; the hinder part of the body falls to the bottom and crawls away as the young starfish.
The comparison of the development of Asterias and Astropecten gives the clue to the meaning of metamorphosis ; it is always a period of rapid change of structure during which the animal does not feed, which bridges over the transition from one set of habits to another. Always this involves the casting off of organs required by the earlier set of habits—and this casting off is the principal element in the change of structure. For in most cases the struc tures required for the second set of habits have already begun to form whilst the first set of habits persists, but these new struc tures are, as it were, sketched out in embryonic tissue and packed away under a fold. Metamorphosis is therefore a secondary falsification of the ancestral record embedded in a life-history.
See INVERTEBRATE EMBRYOLOGY. (E. W. MAcB.) The crisis of metamorphosis, by which an organism wholly changes its structure and mode of life, is not only of great interest as a phenomenon of general biology, but offers the most inter esting opportunities of research in the field of developmental physiology. Gudernatsch in 1911 demonstrated that tadpoles could be caused to metamorphose precociously by feeding them with thyroid gland. He thus showed that amphibian metamor phosis was due to a hormone (q.v.) circulating in the blood, which would explain the synchronization of all the numerous changes which occur together at metamorphosis. Since then a number of important facts have been elicited. All Amphibia which normally metamorphose can be made to do so precociously by the thyroid of any vertebrate, whether fresh or dried, given as food or injected. The effect is a quantitative one up to a point. The greater the dose, the more rapid and abrupt the metamor phosis; after a certain threshold value has been reached, how ever, increase of dose has no further effect. The rapidity of thy
roid-induced as of normal metamorphosis is also affected by ex ternal factors. A protein-rich diet accelerates it, one rich in fat slows it down. Heat hastens, while cold may actually inhibit it.
Further experiment showed that larvae whose thyroids were removed never metamorphosed, but continued to grow as tad poles. It is now certain that the secretion of the animal's own thyroid is the main agency in producing normal metamorphosis.
As the thyroid secretion is exceptionally rich in iodine, the natural supply of this element must constitute a limiting factor, and in waters exceptionally poor in iodine, we should expect a retardation or suppression of metamorphosis. There are certain lakes where newts never metamorphose: it is probable that these will be found to lack iodine. Iodine will also induce precocious metamorphosis, but only in frog and toad tadpoles, not in those of tailed Amphibia. The metamorphosis due to iodine is much more gradual than that induced by thyroid since what is provided is not so much ready-made metamorphosis-producing substance, but raw material which enables the animal's own thyroid to grow more quickly. Iodine even causes thyroidless frog tadpoles to metamorphose, but much more slowly than unoperated animals. Some at least of the body-cells of frog tadpoles must thus have some power of synthesizing the metamorphosis-producing sub stance.
In tailless amphibia (Anura) the thyroid is passing secretion into the blood throughout larval life. In tailed amphibians (Urodela), the thyroid appears to be entirely devoted to storage during larval life; when a certain stage of development is reached, the gland suddenly begins secreting its stored substance into the blood, so inducing metamorphosis. It would be of great physiological interest to discover what is the "releasing factor" which brings about this change in the thyroid.
This difference between anuran and urodele is correlated with another. The growth of limbs is not affected by thyroid in uro dele tadpoles, but is in Anura. In frog tadpoles whose thyroids have been removed, limb-growth, though not absent, is very slow ; in thyroid-treated specimens it is more rapid than normal, the rapidity varying with the dose.