220 E. V. COWDRY 



gastric an^ salivary glands) ; when it merely seeps through into the tissue 

 spaces, veins or lymphatics, as the case may be, the position of the reticular 

 apparatus is highly variable (islet cells and anterior lobe cells of hy- 

 pophysis) ; and, finally, when its position becomes reversed, as it often 

 does in the thyroid, we naturally look for a reversal (probably temporary) 

 in the direction of secretion. 



At present we have no conclusive anatomical evidence bearing upon the 

 question of whether the secretion is continuous or intermittent. We may 

 say, however, that in none of the endocrin organs do we see indications 

 of well marked periodicity, as is the case in certain of the cells of the 

 stomach and other glands pouring their secretion into the alimentary 

 tract. 



Reaction to Experimental Conditions. The structure of the thyroid 

 epithelium is easily modified experimentally, but the changes are difficult 

 to control and interpret. We have good reason to believe that it is 

 affected (1) by variations in diet, (2) by the activity of other endocrin 

 organs, (3) by inanition, (4) by variations in temperature, and (5) by 

 unhygienic conditions, as well as by medication ; and we strongly suspect 

 that many other factors are involved. 



Diet. In connection with diet, Bensley (c) has found the opossum 

 very well suited to experimentation, because, when this animal is kept in 

 captivity, the thyroid undergoes a very marked hyperplasia, which he has 

 found to be characterized "by disappearance of the stored colloid and of 

 the intracellular crystals, and by the appearance in the free ends of the 

 epithelial cells of granules," which he interprets as a new secretion ante- 

 cedent (see p. 217). He experimented with various diets and with iodin 

 administration. With an unrestricted diet the hyperplasia was very great 

 and was not inhibited by dosage with iodin. When, however, the animals 

 were allowed just sufficient food to maintain body weight, no traces of 

 hyperplasia could be seen. Apparently, therefore, the hyperplasia "can 

 be controlled by diet alone," even when refractory to iodin. He then 

 attempted to ascertain the particular food constituent in the unrestricted 

 diet responsible for the production of hyperplasia, and obtained results 

 suggestive of the protein fraction, which have since been confirmed by 

 Burget. 



(Hand Feeding. Kojima finds that feeding thyroid, in the rat, has 

 much the same effect as iodin administration, causing a flattening of the 

 thyroid and an increase in size of the parathyroid cells, while administra- 

 tion of the extract of the posterior lobe of the hypophysis, or of adrenalin, 

 brings about an enlargement of both. According to Wilson and Kendall, 

 structural changes are always accompanied by variations in iodin content. 



Temperature. Mills believes that variations in external temperature 

 exercise a marked effect upon the structure of the gland. He experi- 

 mented on dogs, cats, guinea pigs, and rabbits, and found that "animals 



