NORMAL MODE OF SECRETION IN THYROID GLAND 43 



sidered two possible explanations of this phenomenon, namely, 

 that it was due to lymphatic obstruction, and that it was due to 

 accelerated activity of the gland, and decided in favor of the latter 

 alternative because hgation of the thoracic duct alone produced no 

 such changes in the gland. 



Langendoilf (loc. cit.) on the other hand, while admitting the 

 occasional occurrence of colloid droplets in the cells, did not con- 

 sider them of much significance from the secretory standpoint 

 because of their extreme rarity. Anderson also saw them, not 

 in the normal cell, but as a result of prolonged pilocarpinisation 

 of the anunal. Schmidt could find no effect on the structure 

 of the epithelial cells as a result of pilocarpine injections, and at- 

 tached more importance to the colloid cells as an indication of 

 secretory activity. Bensley ('14) on the contrary, in studying 

 the involution of the hyperplastic gland of the opossum pro- 

 duced by the administration of iodides, found that the cells of 

 the gland practically all contained globules of colloid, and that 

 they could be seen discharging it into the lumen, while colloid 

 cells were almost completely lacking. In this case the restoration 

 of the intrafollicular colloid was wholly by the formation of in- 

 tracellular globules which discharged into the lumen. The proc- 

 ess however was an extremely slow one; after seventeen days, 

 though practically every cell contained a globule of colloid, as 

 large as, or larger than the nucleus, there was little intrafollicular 

 colloid, and at the end of twenty-four days of daily administra- 

 tion of iodides the condition was but slightly advanced; intra- 

 cellular colloid remained about the same as in the preceding case 

 but the follicular colloid was somewhat increased. Recent ex- 

 periments on the hyperplastic glands of the opposum have amply 

 confirmed these results; iodine administered daily to the animal 

 with a hyperplastic gland produces a gradual involution marked 

 by the slow accimaulation in the cells of colloid droplets, usually 

 a single drop to a cell, and the ultimate discharge of these into 

 the newly formed lumen. We may consider it proven therefore 

 that the production of colloid under certain special conditions is 

 by this method. 



