NUCLEI AND KINETIC ELEMENTS 79 



periodic or cyclical in nature. Such variations have to do with 

 the concentration of the chromatin substance about the endobasal 

 body or centriole, being massive and dense in certain phases and 

 distributed in others. In Endamoeha dysenteriae the centriole in 

 the latter phase is distinct and definite but in the former phase it is 

 hidden by the dense chromatin (Fig. 35). From such conditions 

 Hartmann infers that all massive types contain hiflden centrioles, 

 a conception applied by Naegler to all of the smaller amrebae and 

 endamoebee, but is limited to comparatively few types according to 

 Glaser. 



Typical endobasal bodies in the form of centrioles are contained 

 in the first maturation nuclei of Uroleptvs mohUis. Here each 

 massive micronucleus fragments into chromatin granules which 

 remain in a dense reticulum at one pole of the enlarging nucleus until 

 the chromosomes are formed. A centriole, hidden in this mass, 

 divides and one-half traverses the nucleus to form the first pole of 

 the maturation spindle but remains connected by a centroflesmose 

 with the other centriole which, in turn, forms the other pole of the 

 spindle (Fig. 36, b, g). Similar centrioles are found in widely 

 separated groups of Protozoa. In Coccidium. schubergi, according 

 to Schaudinn (1900), the endobasal body divides with a long con- 

 necting centrodesmose. Here, however, part of the material of 

 the centrodesmose collects into two granules with a more densely 

 stained connecting thread, thus producing a structure which Doflein 

 interprets as analogous to the mid-body (Zwischenkorper) of 

 INIetazoa and plant cells. In PoJyiomeUa agilis as described by 

 Aragao, in some trypanosomes, and in many minute amoebse, cen- 

 trioles showing a similar history are of frequent occurrence (see 

 Chatton, Xagler, Glaser, et al.). In Amoeba diplomitotica, iYragao 

 (1904) has also described two types of endobasal bodies in the same 

 species. One resembles the homogeneous endosome of Euglena 

 fir id is in having no centriole (Keuten), while the other type consists 

 of a substance similar to that of the first type within which a 

 centriole is embedded the latter forming a typical centrodesmose 

 during division. It is probable, however, that this supposed dift'er- 

 ence is only a matter of technic. The fate of the centrioles after 

 division differs in different cases. In some e. g., Budo lacerfce, Belar, 

 1921, Figs. 37, 38), they come from the nucleus and reenter the 

 daughter nuclei;* in others they arise from basal bodies and become 

 basal bodies of the flagella after division (e. g., Chilomastix aulostomi, 

 Belar, 1921; Paraixihipioiua, Jameson, Spongomonas, Hartmann, 

 etc.). 



While the embedding matrix in most of the above cases is similar 

 to chromatin in its reaction, and forms an important part of the 



* See, however, the earlier contradictory accounts of Prowazek (1904), Alexeieff 

 (1914), and Kuczynski (1918). 



