DERIVED ORGANIZATION 91 



are formed by the fusion of chromomeres, and some evidence that 

 such structures divide longitudinally. These two conditions, which 

 are relatively rare, are", the only conditions whereby many of the 

 so-called chromosomes of Protozoa resemble those of Metazoa, and 

 if we use the term chromosome at all it should be in a definite, 

 limited, morphological sense and only for those nuclear structures 

 of Protozoa which conform in origin and in fate to chromosomes of 

 Metazoa. I shall use the term chromosome, therefore, only for 

 those compact intranuclear aggregates of chromomeres which divide 

 as unit structures and which are resolved into chromomeres after 

 such division. 



A brief review of some of the frequently recurring types of chro- 

 matin structure at the time of nuclear division will show how diffi- 

 cult it is to speak with assurance of chromosomes in Protozoa. The 

 series is not to be construed as an effort to establish a phylogenetic 

 chain of stages culminating in well-defined chromosomes, nor as a 

 means of pointing out that one is a "higher" type than another. 

 Certain vital functions are undoubtedly associated with the nucleus 

 and with the chromatin of the nucleus, and the fact that some types 

 of organisms with peculiar nuclei continue to live and reproduce is 

 evidence enough that such nuclei are adequate for their needs. 

 The variations in type arise through the association of chromatin 

 with other nuclear or cytoplasmic constituents, and this involves 

 more or less formality in preparation for its perpetuation by exact 

 bipartition to daughter cells. All traces of chromosome formality, 

 however, as well as reduction processes, appear to be absent in 

 gamete nuclei formed by rhizopod chromidia. 



One group of types is represented by massive nuclei as found in 

 the macronuclei of the Infusoria. Here the resting nuclei are made 

 up of closely packed granules or chromomeres and there is little 

 formality or mechanism associated with their division during repro- 

 duction. Each granule elongates and divides into two parts, thus 

 doubling the number of chromomeres. The mass thus formed is 

 passively distributed to the daughter cells by division of the nucleus 

 through the center. It is a quantitative distribution, for the 

 daughter nuclei do not contain representative halves of the indi- 

 vidual chromomeres and the inference is that all of the chromo- 

 meres are qualitatively identical. To this type also I would assign 

 the peculiar chromatin granules of Dileptus gigas which are distrib- 

 uted throughout the protoplasm unconfined by a nuclear membrane. 

 Each granule divides where it happens to be and with the majority 

 of granules both halves remain in one daughter cell after division 

 (Fig. 46). 



These macronuclei, however, particularly the band-form types of 

 the hypotrichous and peritrichous ciliates and the multinucleate 

 chain-form types of hypotrichs, may undergo characteristic pre- 



