Sidney Klein ; 327 
scribed and figured (Fig. 3). As early as four hours after feeding, a 
change in the number and size of the granules may be noticed. They are 
obviously reduced in number and are somewhat smaller. At six hours the 
condition represented in Fig. 5 is presented. The cells are longer and 
narrower and the granules are small, few in number, and occupy the 
distal segment of the cells only. Indeed, at this stage the cells have 
assumed very much the same appearance as they present in the animal 
which is feeding irregularly, although the amount of prozymogen is 
greater in the six hour stage. At no stage of secretion have I found the 
fusion of granules into large masses described by Nicolas, if a fixative 
was employed which was effective in fixing the granules in all parts 
of the tissue. With some of the fixatives mentioned at the beginning of 
this paper great differences were found in granules in different cells both 
in staining power and in form. Very common in these imperfect fixa- 
tions were the granules composed of a faintly stained central mass with 
a deeply staining crescent at one border corresponding to the safranino- 
philous body of Nicolas. I have no doubt that these appearances are 
entirely due to imperfect fixation, and that the crescent-shaped granules 
have no existence in the living cell. On the other hand, the different be- 
havior of granules towards the same fixative even in the same cell indi- 
cates clearly that some of the granules differ from others either chemi- 
eally or physically, and it is probable that this difference is due, as 
Nicolas supposed, to a change in the granule preparatory to its solution 
and extrusion from the cell as a part of the secretion. 
As regards the prozymogen the experiments did not result in a great 
change in its amount. It is probable that an equilibrium is established 
between the rate of production and use of this substance which results in 
the amount in the cell being kept fairly constant. 
SUMMARY OF RESULTs. 
The results recorded show clearly that the cells of Paneth correspond 
in their structure and microchemical reactions to the enzyme-producing 
cells of other granular organs, such, for example, as the cells of the paro- 
tid gland, the chief of the fundus glands of the stomach, and the cells 
of the pancreas. 
In common with these cells the cells of Paneth contain granules which 
do not stain in the specific stains for mucin, such as mucicarmine and 
muchamatein, and which react like zymogen granules to such stains as 
iron hematoxylin, neutral gentian and acid rubin. In addition they 
contain in their basal segment a substance which is distinguished by its 
