44 R. R. BENSLEY 



This conclusion raises the question whether the production of 

 colloid under normal conditions of functioning is by the same 

 method, and, if so, what are the implications of this fact from the 

 standpoint of secretory rate? 



Hiirthle claimed that the formation of colloid droplets in the 

 epithelial cells of the thyroid gland was one of the ways of for- 

 mation of colloid and that their presence was an indication of 

 accelerated thyroid activity. Langendorff pointed out that they 

 were extremely rare, and therefore could not have the secretory 

 importance claimed by Hiirthle. That Langendorff's contention 

 in this respect is correct will readily be admitted. Indeed, one 

 may search complete series of sections of small thyroid glands, 

 and thousands of sections of larger ones without finding in them 

 a single droplet of intracellular colloid. In six thyroid glands of 

 man obtained at autopsies on executed criminals, and examined 

 by the writer, only one contained epithelial cells with colloid 

 droplets in them. In pathological glands from cases of exoph- 

 thalmic goiter, simple colloid goiter, and colloid adenoma, on the 

 other hand, they occurred with variable frequency. In the one 

 normal gland that contained them the colloid drops occurred with 

 great frequency. For the most part they were placed not at the 

 free margin of the cell, but deep in the protoplasm, often along- 

 side of the nucleus, and in many cases several drops formed a row 

 extending from this deeper location to the free border. In many 

 follicles, however, the colloid droplets occupied the tips of the 

 epithelial cells, and in others the colloid masses inside the follicle 

 could be seen to be made up of a cluster of small droplets, ap- 

 parently derived from different cells, which had failed to fuse 

 with one another inside of the follicle. This gland also con- 

 tained an unusual number of colloid cells of Langendorff. 



The obvious participation of these intracellular colloid drop- 

 lets in the replenishment of the intrafollicular colloid, on the 

 one hand, and the slowness of this process demonstrated by expe- 

 riment, and the rarity of the occurrence of such droplets under 

 normal conditions, on the other hand, suggest the following pos- 

 sibilities, which, however, are not, as will appear more clearly 

 later, mutually exclusive: (1) the formation of colloid is an inter- 



