It is now firmly established that the coordination of chromatic response in Amphibia is predominantly, if not exclusively, due to the liberation of hormones by reflexes involving visual and skin receptors, and in Reptiles to direct innervation of the pigmentary effector organs (Hogben and Mirvish, 1928; Zoond and Eyre,1934). Among aquatic vertebrates examples of both types of coordination occur. The bulk of available evidence points to the conclusion that the chromatophores of Teleostean fishes are directly innervated and that the comparatively rapid responses which are exhibited by several species are brought about by simple reflex action. That this is not true of cyclostomes has recently been shown by J. Z. Young (1935) whose experiments demonstrate the archaic phylogenetic character of the control exercised by the Amphibian pituitary gland. From an evolutionary standpoint it would not be surprising to find among physiological mechanisms in Teleostean fishes examples of specialization comparable to the strikingly aberrant features which their anatomical organization displays. On the other hand it would be remarkable if the cartilaginous fishes proved an exception to a rule which applies both to Cyclostomes and to Amphibia. Recent work on the colour changes of Elasmobranchs supports the conclusion that the coordination of colour change in Teleostean fishes is highly specialized. Lundstrom and Bard (1932) have shown that total removal of the pituitary gland in
Mustelis canis
results in a state of pallor which ensues within a few hours after operation, reaching its limit about the twelfth post-operative hour. In their experiments the animals usually succumbed after three or four days with loss of righting reactions. Only a few survived as long as a week. The effect was not produced by removal of the anterior lobe alone, nor by severe traumatization of the hypothalamus. Complete darkening of the pale operated animals followed injections or extracts of ox pituitary and of the pituitary of the fish itself, the quantity present in the fish gland being greatly in excess of the amount requisite to induce full expansion of the dermal melanophores. The present investigation, undertaken to throw further light on the evolution of the chromatic function in Vertebrates, is based on several species of Elasmobranch fishes, namely the skates,
Raia Brachiura, R. clavata, R. maculata, R. microcelatus
, the speckled dogfish
Scyllium canicula
, the banded dogfish or nursehound
S. catulus (Scylliorhinusstellaris)
, and the monkfish
Rhina squatina
. The writer is indebted to Mr. G. A. Steven for invaluable assistance in identifying the various species used. In all these species the pigmentary effector system of the integument, like that of the American dogfish
Mustelis canis
, closely resembles that of Amphibia, and consists of three types of chromatophores which are more or less evenly distributed. These are the epidermal melanophores, larger more richly branched dermal melanophores, and xanthophores containing an orange yellow pigment. The same agencies, in the fishes to be described, evoked or maintained pigment diffusion (“expansion”) of all three types, and the concentration of pigment in the centre of the cell (“contraction”) in all three types. That is to say, the xanthophores of a skate or dogfish which was maximally pale were always fully contracted like the melanophores of both kinds, and the xanthophores of a dark animal were fully expanded. This is true of some—but not all—Amphibia. In general appearance the chromatophores of the species studied are more like those of a Urodele than those of a Teleost.
Spiders did not repeatedly gain, but repeatedly lost, foraging webs