286 THE BIOLOGY OF STENTOR 



(unpublished). Such nuclei have not yet been examined after 

 staining; all we can say is that either separated nodes fall into 

 perfect alignment or they are able to rejoin and become enclosed 

 again in the common membrane. The latter seems more probable. 



Coalescence begins at both terminals and progresses toward 

 the mid-nodes, as Weisz (1950b) and others have noted. How is 

 this accomplished? Stolte suggested fusion by swelling of the nodes, 

 but I think what he observed was simply the coalescence of the 

 individual nodes into larger ones; and there certainly does not 

 appear to be an increase in the total volume of the nucleus during 

 condensation as his suggestion would imply. More probably, 

 internodal connections swell while the nuclear membrane contracts 

 and decreases greatly in area. 



There remains to be conjectured why the chain nucleus should 

 clump at all. In both Loxodes (Balbiani, 1893) and in Dileptus 

 (Jones, 1 951) there is a distributed nucleus consisting of many 

 separate macronuclei and these do not fuse during division and 

 regeneration. Coalescence therefore seems not absolutely necessary 

 in the life of ciliates, though it may bring advantages. That no 

 nuclear changes occur until the oral primordium is nearly com- 

 pletely developed, as Balbiani (1893) first emphasized, does not 

 suggest that either coalescence or moving all parts of the nucleus 

 close to the primordium is necessary to its development. Hence 

 Balbiani did not share Gruber's idea that clumping is to give a 

 single, central guidance to morphogenetic events, because these 

 processes are nearly completed before the nucleus fuses; and 

 besides, the macronuclear chain is all one nucleus anyway. 

 Balbiani's suggestion was therefore that the nucleus concentrates 

 in order to have the greatest effect ; but the action of the nucleus is 

 more likely to be promoted by increasing rather than by decreasing 

 its surface. As already mentioned, Johnson held that coalescence 

 of the nuclear chain is an instance of Haeckel's law of recapitulation 

 and hence is performed for "historical" reasons. That the macro- 

 nucleus clumps in order to insure equal division through the simple 

 splitting of a compact mass has also been suggested (Sonneborn, 

 1947), but in Stentor at least, clumping does not insure this end 

 (see p. 71). Many hypotrichs like Eiiplotes produce reorganization 

 bands at every division and somehow transform or rework the 

 macronucleus ; yet in Stentor no one has found evidence of any 



