210 CAREY P. McCORD AND FLOYD P. ELLEN 



types. The simpler and less conspicuous form is limited to 

 the epidermis. It consists of a cell-body with two or more 

 simple processes {Ep. M., figs. 5 and 6). They may lie singly 

 in the epidermis or, in such abundance as to form a definite 

 reticulum. These melanophores are not contractile (3). 



The second type of melanophore is found in greatest abund- 

 ance in the sub-epidermal connective tissue. During late 

 metamorphosis these cells migrate to the corium. In the ex- 

 panded condition the sub-epidermal melanophores present a 

 very typical 'mossy' appearance. The cells lie so closely ap- 

 proximated that they form a nearly continuous sheet (Sub. M., 

 fig. 6). There is a lighter, central space in each cell probably 

 representing the nucleus. In the contracted condition (Sub. M., 

 fig. 7) the melanophores appear as irregular dots in which no 

 structure is visible. 



It has been fairly well established that the contractile melano- 

 phores are innervated by motor fibres proceeding along both 

 sympathetic and spinal nerve paths (4). Hooker has shown 

 physiologically that the reactions of the melanophores of the 

 frog are synchronized by the action of the central nervous sys- 

 tem. By histological methods, Ballowitz (5) has demonstrated 

 motor nerve endings in melanophores of bony fishes. Laurens 

 (6) working with Amblystoma larvae has shown that the melano- 

 phores may contract as the result of direct stimulation. Often, 

 however, this primary reaction is overcome by an opposite, 

 secondary reaction initiated by the central nervous system. 



There are three principal theories to account for the mechanics 

 of melanophore contraction. Ballowitz (1) claims that the 

 contraction is an intra-cellular migration of the pigment gran- 

 ules within fixed cells. The protoplasm of a chromatophore 

 is filled with numerous, extremely fine, radially arranged, anasto- 

 mosing canals within which the pigment is forced back and forth 

 by the alternate contraction and relaxation of the protoplasmic 

 canal- walls. 



Spaeth (7) believes that the chromatophores of fishes are 

 fixed stellate cells, within which the pigment granules, carried 

 in a rather fluid cytoplasm, stream into and out of the processes 



