No humoral substance has excited in the vertebrate melanophore such constancy of respoiise as adrenalin. All reactive integumentary melanophores in vertebrates concentrate their pigment to adrenalin and all reactive retiiial melanophores in these animals disperse their pigment to this hum0r.l Notwithstanding this remarkable constancy, the part of the melanophore system through which adrenalin acts, at least in the integumentary elements, is wholly in doubt. It is the object of this paper to indicate what this part is, at least in the fish Fundulus het eroclitus Linn.
Lieben, who was unaware of the earlier discoveries of Corona and Moroni (1898) on the effects of adrenalin on melanophores, published in 1906 an account of the concentration of this pigment in these cells in the frog by adrenalin and since this action took place in the leg of this animal after the sciatic nerve had been cut, he ascribed it to the direct action of the humor on the melanophores and not to a possible action through nerves. This conclusion was criticized by Fuchs ( '14, p. 1530), who pointed out that the melanophores were under the control of the autonomic nerves and that the fibers of these nerves might be expected to follow the blood-vessels rather than the sciatic nerve. From this time forward 'The concentrating action of adrenalin on the pigment in the vertebrate integumentary melanophores is so fully substantiated by a host of workers as to require no special comment. The dispersing action of this humor on the vertebrate retinal pigment was misinterpreted by Klett ( 'OS), but was subsequently correctly observed and reported by Fujita ( 'll), by Biglley ( '19), and by Gibson 311
( '22)-JOURNAL OF CELLULAR AND COMPARATIVE PHYSIOLOOY, T ' O L 5, NO. 3