THE ERGASTOPLASM IN THE MAMMARY GLAND AND ITS TUMOURS 1 75 



provide. It must be pointed out that even in the case of the pancreas, the 

 morphological demonstration of a relation between ergastoplasm and 

 protein granules is restricted to the one example of guinea-pig pancreas [7]. 



Likewise in the mammary gland, aspects showing this relationship are 

 exceptional. It is in one pathological case only that an example could be 

 found where protein granules had accumulated in the lamellar system of 

 the ergastoplasm (Fig. 5). When such images are found however they lead 

 to the belief that protein elaboration may take place in the RNP-granules 

 bordering the membrane of the endoplasmic reticulum and that secreted 

 products then proceed towards the Golgi apparatus where they will take 

 on their final form. Siekevitz [8, 9] has elegantly developed the matter. 



When the mammary gland enters the involution phase which follows 

 its active secretory phase, the reduction of the alveolar trunk is spectacular 

 and similar to that of the virgin gland. Its ultrastructural counterpart, 

 likewise is striking. The rare persisting ducts or buds are lined with mono- 

 tonous-looking cells. The nuclei are shrunken, the cytoplasm clear, hardly 

 containing under-developed organelles. The ergastoplasm in particular is 

 now only represented by rare dispersed RNP-granules or isolated dotted 

 membranes. Not only is this decrease quantitative but from the special 

 ultrastructural point of view with which we are concerned, the ergasto- 

 plasm has not retained the lamellar organization which was characteristic 

 of active secretion and under this form it has disappeared together with 

 the secretion granules themselves (Fig. 6). 



Though this remarkable arrangement of the ergastoplasm has always 

 been noted and described, its absolute relation to actual secretory activity 

 has not been specially emphasized. Yet it is striking. If one creates con- 

 ditions in which the physiological secretion is modified as under the 

 influence of hormones in the case of the mammary gland this organization 

 typical at first (Fig. 2) does not persist in the absence of sucking and is 

 disrupted after prolonged stimulation (Fig. 8). 



2. The ergastoplasm in cancer of the mammary gland 



The bond between normal secretion and presence of parallel-organized 

 ergastoplasm is further demonstrated if tumours are studied. 



It is most striking that in over 100 cases of liiiman breast cancer studied 

 in the electron microscope [10, 11] none has been found where ergasto- 

 plasm organized in parallel arrav was present in noticeable amount. 

 There has been one exception to the rule onlv (Fig. 9), but in this case so 

 many other features were aberrant [12] that it cannot be taken int(^ account 

 in the present discussion. The rarity of ergastoplasm under this form is 

 here again linked to absence of the normal physiological secretion. 

 Other secretion products may be found in these cancerous cells such as 



