150 J. J. LISTER. 



nucleus was present in all but one (which had two nuclei, and 

 was regarded as pathological), and the phases presented by the 

 nuclei fall into a continuous series. 



It is, of course, possible, though not probable, that Rhum- 

 BLER did not happen to meet with the microspheric form, which 

 is always the rarer, or that this form is in Saccammina so 

 different from the megalospheric, that the two forms are not 

 regarded as belonging to the same species. As the evidence 

 stands however, it appears that we have in Saccammina a 

 genus in which the differentiation into microspheric and 

 megalospheric forms has not occurred. Its nucleus would not, 

 therefore, be strictly comparable with the nucleus of the megalo- 

 spheric form of a dimorphic species, such as Polystomella crispa 

 (Linn.).] 



In the second, or reproductive phase, minute nuclei (1 to 2 /jl) 

 are at first found scattered irregularly through the protoplasm 

 in varying numbers (fig. 30), and broad channels of communica- 

 tion become opened up between the inner and outer chambers. 

 The little nuclei are sometimes found as compact bodies, some- 

 times excavated by vacuoles, and the changes of which these 

 conditions are the expression appear to be simultaneous 

 throughout the animal. 



It appears that, while the nuclei are sometimes readily 

 stained by picro-carmine, they may pass through a phase in 

 which they do not react to this stain, for specimens are not 

 unfrequently met with in which the large nucleus has dis- 

 appeared, and the direct communications between the chambers 

 have become established, and which, therefore, appear to be in 

 the reproductive phase, but in which no stained bodies of any 

 kind are recognisable. 



Ultimately the nuclei become evenly dispersed throughout 

 the chambers and divide by karyokinesis, the protoplasm 

 becoming aggregated about them in spherical masses (3'5 fj, in 

 diameter), each of which contains a dividing nucleus (figs. 27 

 and 28). At a later stage each nucleus, presumably the 

 daughter nucleus of this division, becomes the centre of a 

 flagellated zoospore (fig. 32, a and 6). These are of approxi- 



