JI1STOLOUY OF THE l'< )IS( )N-< iLA NDS OK lil'FO AfiUA. 



SUMMARY. 



To sum up the results of these investigations on the cutaneous glands of the tropical 

 load Bnfti mjua, we can say that they may be divided physiologically by the nature of their 

 secretion into (1) adrenalin producers, and (2) those whose secretion contains none of that 

 compound. The adrenalin-producing acini are limited to the glandular masses behind the 

 eve and surrounding the tympanum, known as the "parotid" glands, the chromamn reac- 

 ( ion being negative in all other cutaneous glands. The contents of the poison-glands having 

 been discharged, probably through the agency of the contraction of a smooth muscular 

 coat surrounding the gland acinus in response to a stimulus of sufficient strength, the emp- 

 tied sac does not refill, but is resorbed, and its place taken by an immature gland of the 

 same type. These young glands bud from the duct or neck of the older gland-sac, about 

 which a ring of them may be found. In the process of maturation they grow downward 

 through the various strata of the cutis vera, carrying with them in their descent a layer of 

 tissue from the outer loose layer of the derma, and are suirounded by it during their exis- 

 tence. The secretion of these poison-glands is produced by and during the destruction of 

 the cytoplasm of epithelial elements lining the acinus of the young gland, which destruction 

 leaves only a cell nucleus within the acinar wall. 



Two sorts of secretion occur in the lumen of the mature poison-gland: (1) a granular 

 secretion, which is the first to be formed, and which is found filling the entire lumen of the 

 young glands and is located in the central part of the mature acini; (2) a clear, homo- 

 geneous or finely punctate fluid filling the periphery of the lumen, which takes on a bright 

 yellow color after fixation in fluids containing chromates. This chromaphil secretion is 

 not found in the young glands, nor does it ever occur in cutaneous glands in other parts 

 of the body. As the gland grows older without being used, the chromaphil secretion 

 diffuses into the central part of the gland, so that the entire venom stains yellow when 

 treated with chromate-containing solutions. The homogeneous liquid is the first of the 

 gland secretion to reach the skin surface when the gland discharges in response to a stimulus, 

 probably because it is less viscous than the granular secretion, and its appearance is followed 

 by the ejection of the first or granular emulsion in the center of the gland. In this way the 

 adrenalin component of the gland venom is the first poison to reach the mucous membranes 

 of an attacking animal. 



The adrenalin content of the venom is not secreted as such by the glandular epithelium, 

 since at no time in their life history do the epithelial cells show any yellowing after treat- 

 ment with chromate solutions, nor does the poison-sac contain chromaphil material until 

 long after the disappearance of all epithelial elements. The adrenalin is probably the 

 result of a change produced in a mother substance, which is very likely an amino -acid, 

 as Guggenheim first suggested, through the action of the naked nuclei of the old epithelial 

 cells which lemain attached to the inner surface of the wall of the poison-sac. 



Analogous examples of like processes occurring elsewhere in nature lead us to believe 

 I hat the change in question may be a process of decarboxylization, which results in t In- 

 formation of this aromatic amin base from a corresponding acid. 



We have no clue to the origin of this hypothetical ammo-acid, but such an acid might 

 readily be a result of the breaking down of the cellular elements of the gland, since many 

 amino acids are derivatives of protein destruction. The granular secretion would appear 

 to be formed directly in the cytoplasm of the gland-cells during their most active period of 

 swelling and breaking down. 



The chromamn organs of the body (adrenal, medulla, etc.) do not appear to influence 

 or be influenced by the presence of these cutaneous depots of adrenalin. 



