FRESHWATER ALG^ OP THE SOUTH ORKNEYS. 107 



subsequent description). There is, further, an absence of connecting-stages between the 

 normal Sco&eWa-individual and the Chlorosphaera-cells (saving such a stage as is shown 

 in fig. 16). And, lastly, the Chlorosphsera-ceHs, as above mentioned, are of very diverse 

 dimensions, whereas the range of size of the Scotiellci-imMvidnals is small ; and I have 

 never met with any small enough to give rise to resting-cells of the dimensions found 

 in the smaller Chlorosph3Bra-cel]&. The cells of the latter (even the largest) are, 

 moreover, occasionally found in considerable masses (twenty to fifty or more), separated 

 from one another only by their mucilage-envelopes, which then often become polygonal 

 as a result of mutual pressure ; this is much more like what we should expect in a 

 Chlorosphsera, than in resting-cells of a form like the Scotiella, occurring as it does 

 as isolated individuals. My object in discussing this point in detail has been to 

 avoid the immediate adoption of an assumption which at first sight seems very plausible. 

 At the same time, I do not wish to deny that, with only the present material available 

 for investigation, we cannot completely dismiss the possibility of certain of the large 

 CV;/o;u^>//,3?ra-cells ultimately proving to belong to the life-cycle of S. antarctica ; l 

 it does not, however, seem probable. 



In the course of the previous paragraph reference was made to resting-cells of 

 Scotiella, which appear to arise in the following way (PI. I., figs. 12-15). Kather 

 rarely one finds individuals of this alga in which the contents are markedly contracted 

 or more or less rounded off (figs. 12, 13, also 16) ; in such cases some or all of the 

 wings appear irregularly folded and stratified, so that they no longer show up so 

 markedly (cf. figs. 12 and 16). Various stages of this kind have been found which 

 seem to form a complete series connecting the normal ScofiW/a-individual with large, 

 thick-walled resting-cells (fig. 15), the general shape of which is rounded, while the 

 membrane is double, with a more or less undulated outer layer. The formation of 

 these resting-cells (akinetes) appears to be initiated by a swelling up of the wall of 

 the individual, and the rounding off of its contents ; as a result of the swelling of the 

 wall the outline of the wings is obscured, while the wall becomes more or less rounded 

 off around the contracted contents (cf. figs. 13, 14); at the same time a stratification 

 becomes apparent in the swollen wall, which subsequently becomes more strongly and 

 markedly thickened (figs. 14, 15). The contents appear to undergo a gradual change 

 during the formation of the resting-cells. At first (fig. 13), as in the ordinary 

 individuals, we have a central mass of granular protoplasm with a cap of fat at either 

 end ; as the contents round off, however, this fat disappears (used in forming the thick 

 walH), and the mature resting-cell contains only granular protoplasm (fig. 15) possibly 

 with a certain amount of fat diffused through it. Resting-cells of the type just 

 described are fairly common in parts of the yellow snow material, but I have not met 

 with any structures which could be interpreted as germination-stages of these cells ; 

 their further fate must therefore at present remain an open question. 



1 If so, they might prove to be germination-stages of the resting-cells described in the next paragraph, but there 

 is no evidence for this. 



