STRUCTURE OF NUCLEUS 57 



nuclei of Protozoa. It is supposed to be embedded in the karyosome, 

 and only to become recognizable as centrosomic in nature during nuclear 

 division, when it divides into two parts which separate from one another, 

 though remaining connected for some time by a fibril, the centrodesmose. 

 As the daughter centrioles move apart they take up positions at the ends 

 of the elongating nucleus, while spindle fibres surrounding the centro- 

 desmose may form between them (Fig. 59). The chromatin of the nucleus 

 may form definite chromosomes, which arrange themselves as an equatorial 

 plate at the equator of the spindle. Two daughter plates are formed, 

 and these travel towards the centrioles at opposite poles of the nucleus. 

 When the nuclear membrane divides, the spindle fibres and centrodesmose 

 disappear, a new karyosome is formed around the centriole, and the 

 chromosomes break up into granules, which are distributed on the 

 nuclear membranes or linin network. Observers are, however, by no 

 means convinced that such a granule is a true centrosome, for in many 

 Protozoa undoubted centrosomes exist in the cytoplasm outside the 

 nuclei. 



The size of the karyosome in proportion to that of the entire nucleus 

 varies considerably. In many nuclei, especially those of small size, it 

 has been the rule to regard the bulk of the chromatin as being aggregated 

 in the relatively large karyosome, and to suppose that little, if any, is 

 distributed upon the nuclear membrane or linin network. It is becoming 

 increasingly evident, however, that all nuclei contain some chromatin 

 on the linin network or membrane (peripheral chromatin). In other cases 

 the karyosome is relatively small, while definite chromatin granules occur 

 upon the nuclear membrane or the linin network. The nuclei of the first 

 type are often spoken of as of the karyosome type, but every transition 

 between the two types of nuclei occurs. This question as to whether a 

 centriole is always present or not is a very difficult one to decide, for the 

 statements regarding it are most conflicting. Some observers are able 

 to find centrioles in nearly every nucleus they examine, while other 

 equally competent observers fail to detect them. The difference of opinion 

 is to be accounted for by the fact that the centriole is merely a minute 

 granule, the nature of which can only be determined by its behaviour 

 during actual division of the nucleus. While division is taking place, there 

 are always numbers of granules in the nucleus. Many of these are chro- 

 matin granules, and as spindle fibres are often present while the nucleus 

 is dividing, it is easy to interpret any two granules and a connecting fibre 

 as centrioles and centrodesmose. It thus happens that it is more than 

 doubtful if most of the structures which have been described as centrioles 

 are actually of this nature. The mere presence of a central granule in 

 a karyosome of a resting nucleus appears to some observers to be sufficient 



