(308 



be definitely determined whether mitosis was limited to young un- 

 ciliated elements, and amitosis to the larger older ciliated cells. No 

 mitoses were seen in the latter, but it remains uncertain whether the 

 mitotically dividing cells actually lack cilia. 



In the cuboidal cells lining the efferent ducts of lobster testis 

 (apparently non-ciliated in my preparations), amitotic nuclear divi- 

 sions are very abundant, and very frequently multiple. 



Discussion. From the foregoing observations it appears that 

 the predominant, probably the exclusive, method of division of 

 ciliated epithelial cells is by amitosis. The sole apparently real ex- 

 ception appearing in my material is the ciliated cell of the epididymis 

 of the horse. This exception will be considered below. The imme- 

 diate discussion is confined to the data derived from the mouse. The 

 cuboidal cells are of the efferent ducts (vasa efferentia) of the epi- 

 didymis, the tallest columnar of the main duct (vas epididymis). 

 The former have motile cilia, this latter non-motile; and both types 

 appear attached to basal bodies presumably derived from the cen- 

 trosome. These cells are neither embryonic nor degenerate nor patho- 

 logic nor essentially secretory. Nor does there seem to be any structural 

 arrangement entailing a stringency of food or oxygen supply. They are 

 specialized in that they are ciliated. The specialization underlying ami- 

 tosis in ciliated cells is probably not simply the presence of cilia; the 

 cause of amitosis here inheres perhaps in the fact that in the formation 

 of cilia the dynamic center (the centrosome) of the cells was used up 

 in the origin of the basal bodies from which the cilia developed (von 

 Lenhossek). Thus the fundamental condition determining amitosis 

 would seem to be one affecting the presence, normal behavior, inte- 

 grity, or virility of the centrosome. Amitosis in rapidly growing em- 

 brj^onic tissues may then be due to a lack of sufficient or of appro- 

 priate (e. g. oxj^gen, Wieman) nutrition; in degenerating tissue the 

 same factors are conceivably present, working their initial effect upon 

 the most delicate and sensitive organ of the cell, the centrosome; in 

 pathological growths similar conditions may prevail or the centrosome 

 may be more directly influenced by the morbid conditions, in less 

 extreme degrees giving origin to the atypical multiples spindles cha- 

 racteristic of carcinomata; in conditions of extreme intensity of 

 specialization or function, the element of inadequate food would 

 again appear and might affect the centrosome; and where the cen- 

 trosome is directly used up in the process of specialization as in ciliated 



I 



