IIIST<)l,<)<;-\ OF 'I UK I'OISOiN'-CLANDS <>!' HI F< I AGUA. SI 



striated muscle, hut sonic, misled by their appearance in cross-sect ion, have described them 

 as the mother cells of the gland epithelium (Seeck, ISIII; Weiss, ISllSi. They have been 

 correctly described by Seeck, but he distinctly stales that they are non-muscular. 



Resting upon the matrix which supports the muscular coat, one finds the remains of 

 the cells which once formed the secreting epithelium of the gland which exists in the adult 

 stages only as nuclei (fig. (i) that are apparently uncovered by any cell cytoplasm. Then- 

 is by no means a continuous layer of these nuclei; instead, they are assembled in groups, 

 often with wide intervening spaces between them. These nuclei, which are probably held 

 in place by the remains of a cell membrane, are large round or flattened masses which in 

 the very oldest glands, as in the younger acini, show a definite irregular chromatin network 

 with one or two large chromatic masses. 



The collar of the gland is a cone-shaped mass of cells which rests with its base over the 

 opening of the gland acinus. It has a lumen continuous \\ it h the lumen of the duct , which 

 is in most cases not patent except during the expulsion of the venom. The collar is the 

 final point of attachment of the muscular layer of the gland-wall. From it the fibers run 

 downward, in a spiral manner, over the upper part of the gland acinus. Since the collar 

 is solid it forms an ideal fixed point for the muscular layer when it contracts to expel the 

 secretion. Muhse has described for this collar a radiating structure and in the toads which 

 she examined has found both collar and duct to be built of an outer and an inner zone of 

 cells (fig. 0), continuous respectively with the transitional and the molt strata, the dead cells 

 of the latter reaching a depth approximately level with the lower surface of the epidermis. 

 In Biifo aijuii this layer of dead cells goes much lower and may reach down the duct for 

 a depth of 200 or 300 micra. Each zone in these toads is composed of several layers of cells. 



The ultimate attachment of the muscle coat surrounding the gland acinus is, as Muhse 

 describes it, a point arbitrarily selected as the dividing line between collar and duct, at 

 which the cells assume a "cog-wheel" structure, cells of the outer zone being ai ranged in 

 groups radiating from the ring of inner zone cells. The muscle fibers in the gland matrix 

 are intimately mixed with these radiating groups of cells, from which attachment they run 

 downward and slant about the gland acinus in such a manner that their contraction opens 

 the duct lumen, which is closed at the lower part of the neck. P^urther contraction, of 

 course, compresses the body of the gland and squeezes it empty of secretion. 



The gland duct or neck is cylindrical in form (figs. 4 and 7) and continuous with the 

 cells of the gland collar, retaining the two radial zones up to its junction with the epidermis. 

 The narrow lumen, which is not uniformly patent under normal conditions, but whose 

 walls may touch one another, opens into a deep and wide pit or foveola, plainly visible to 

 the naked eye on the skin surface. The lumen becomes narrower as we go down the duct 

 until the junction of the duct with the collar is reached, where it is entirely closed. It may 

 be patent throughout the collar, but is usually closed partly or entirely in this part of the 

 gland. If it is patent in this region, the lumen is very tiny and it is always found closed at 

 the opening of the acinus. 



Careful inspection of our sections has led us to disagree with Mrs. Muhse's opinion 

 that the cells of the epidermal stratum germinativum have no connection with the duct 

 epithelium after the formation of the gland-bud. We have been able to trace the germinal 

 layer in an unbroken line down the length of the duct into the collar and are forced to the 

 conclusion thai all three layers are represented in the gland-duct, at least in this particular 

 variety of toads. We agree with Seeck, who slates that the stratum germinativum is con- 

 tinuous throughout the gland-duct. 



