Tadpoles were treated with intraperitoneal injections of mammalian prolactin and growth hormone (NIH). The growth rate of hwophysectomied tad. poles (R. pipiens) was accelerated by both hormones, prolactin having the greater effect. Metamorphosis in the bullfrog ( R . catesbeiuna) was inhibited by prolactin but not by growth hormone. The results are interpreted to indicate a general antag. onism between prolactin and thyroid hormone in amphibian development with respect to both growth rate and metamorphosis.

Spontaneously metamorphosing bullfrog tadpoles and those induced to metamorphose by injections of thyroid stimulating hormone (TSH) were given a single intraperitoneal injection of thymidine-3H (10 muCi/g body weight). The brains were dissected out at various periods 3 hours to 14 days later, and processed for autoradiography. At the 3-hour interval after thymidine-3H injection, ependymal cells were labelled, but not the external granular layer (EGL) cells. Furthermore, in all the metamorphosing tadpoles intense labelling was restricted to the ependyma of the marginal region of the cerebellar plate. At 48 hours, labelled cells were seen in the EGL of the marginal region. At the 4-day interval, most of the EGL was labelled, and labelled cells were also seen migrating from the EGL into the internal granular layer (IGL). By 14 days, several labelled cells were seen in the IGL. Although the sequence of cerebellar histogenesis in the frog is similar to the general pattern described in other vertebrate groups, the results indicate that the EGL of the frog cerebellum does not serve the function of a secondary proliferating zone.

Calcium binding protein (CaBP) immunoreactivity in the cerebellum of bullfrogs was examined, concentrating on cells associated with the auricular lobe. While anti-calretinin and anti-parvalbumin also immunoreacted with the same cell populations, anti-calbindin exhibited the most robust and typical pattern of immunostaining. Calbindin immunoreactivity was observed in various populations of cells in the auricular lobe and interauricular granular band of the cerebellum, in the cerebellar peduncle, and in a bundle of interauricular commissural fibers which course through the dorsal, marginal, part of the molecular layer. Cells in the granular layer of the ventral part (i.e., corpus cerebelli) of the cerebellar plate were not CaBP-immunoreactive, nor were any fibers in the molecular layer of this cerebellar region. We believe that axons of CaBP-immunoreactive granule-like cells of the auricular lobes contribute to the formation of the interauricular fiber bundle, which corresponds to the lateral commissure of urodele amphibians. The pattern of calbindin immunoreactivity in the auricular lobes and marginal part of the cerebellar plate provides additional evidence that this cerebellar compartment, which is already present in tadpoles, has a distinct origin, biochemical characterization and connectivity and is separate from the compartment that forms the corpus cerebelli of frogs during metamorphosis.

The cerebellum of the frog tadpole remains in an immature state throughout the premetamorphic (growth) phase. An external granular layer is absent except for a few scattered cells. A well defined Purkinje cell zone is also lacking. Large cells, which presumably are the immature Purkinje cells, are found in the subventricular zone.During the prometamorphic period (accelerated hind leg growth) an external granular layer, several cells thick, becomes established by a migration of cells from the base of the cerebellum. A second migration involving the movement of the external granule cells toward the ventricular region begins at this time. However, the most active period of this latter migration is at the end of the prometamorphic period and during the days following fore leg emergence, viz., metamorphic climax. Concomitantly, a established.Amphibian metamorphosis, in general, entails sweeping morphological and biochemical changes and has, therefore, attracted a great deal of experimental investigation. However, except for a few studies such as those on the neurons of the mesencephalic V nucleus (Kollros and McMurray, '55) and on Mauthner's cells (Moulton, Jurand and Fox, '68), metamorphic changes in the brain have received little attention. This paucity of interest in the central nervous system of the metamorphosing tadpole is not difficult to explain. Once the embryonic development is complete, the tadpole has a brain which is essentially identical to that of the adult. Metamorphosis, therefore, does not bring about any dramatic changes which are readily noticeable in the brain.In the course of exploratory studies on the histological changes in the tadpole brain, our attention was drawn by the abrupt changes in the cerebellum during metamorphosis. Ramdn y Cajal ('11) had observed many years ago that the cerebellum of the frog tadpole had immature Purkinje cells with poorly developed dendritic processes. Larsell ('25) later reported that the tadpole cerebellum did not acquire a well defined Purkinje cell layer J. COMP. NEUR., 146: 133-142. well defined Purkinje cell stratum is until late in larval life. However, no further studies seem to have been pursued on this or other aspects of the morphological changes in the cerebellum during metamorphosis. It is the purpose of this report to describe light microscopic observations on the cerebellar changes during normal metamorphosis. MATERIALS AND METHODSTadpoles and postmetamorphic froglets of the bullfrog, Rana cntesbeiana, and of the leopard frog, R. pipiens, were used.Those of the bullfrog were collected from a pond in Pcekskill, New York and those of R. pipiens were raised from eggs obtained by induced ovulation (Rugh, '34). About 80 animals of each species were sorted into groups of different stages ranging from young tadpoles (Taylor and Krollros ('46) stage VII in the case of R. pipiens) to postmetamorphic froglets of up to ten weeks after the emergence of fore legs.In some cases the upper jaw was cut out and fixed in Bouin's solution for three days. The b...