202 
Fritsch.. StiicLies on Cyanopbyceae. 
be due to sucli clianges. YTben a filament lias a row of bete- 
rocysts one beliind tbe otlier (cf. Fritsch. 04. p. S9< the distal 
one bas no mucilage envelope wbatsoever: tbe next beterocyst 
is surrounded by a layer of mucilage. formed before its Trans¬ 
formation. tbe tbird bas two such layers, and so on — Tbe 
cell-sbeatb. as in tbe case of tbe blaments treated witb lodine. 
becomes better defined, wlien stained witb Vesuvin: tbe inner 
investment bowever remains practically nncolonred. In young 
blaments. in wbicb a cell-sbeatli is not yet recognisable. staining 
witb Vesuvin makes tbe inner investment (tben tbe only one ? 
particnlarly prominent: for tbe protoplast is separated from tbe 
brownlv stained external mucilage by a narrow colourless area. 
representing tbe inner investment. 
(b) Nostoc. It is scarcely necessary to give rnany details 
bere. as tbe structure practically agrees witb tbat of Anabaena . 
I was not able to obtaiu sporogenous blaments and tberefore 
tbe recognition of many of tbe points was of considerable clif- 
bculty. Witb tbe belp of lodine or Vesuvin bowever tbe cell- 
sbeatb was brougbt out prominently and especially tbe lateral 
parts are tben web debned. Tliere can be no doubt in sucb 
stained specimens, tbat tbe lateral portions of tbe outer envelope 
(cell-sbeatb') are only proper to tbe individual cells (cf. bg. 3b. 
In manv cases too tbe tbick dark line, whicb marks tbe lateral 
portion of tbe cell-sbeatb extends round on to tbe terminal 
portion of tbe ceb for a bttle way. wbicb is probably due to a 
sliglit tbickening of tbe margin of tbe open end of tbe cybn- 
drical sbeatb: tlie cell-sbeatb tben appears [ ] in optic-al 
section. 
(c) Gloeocapsa and Gloeothece. An interesting case is fur- 
nisbed by Gloeocapsa : in most cases (especially in tbe large- 
celled forms) two envelopes are quite readily distinguishable 
around tbe cell-contents viz. tbe colourless inner investment 
(bg. 9, i. i.) and surrounding tbat a well-marked cell-sbeatb 
(bg. 9, c, s), wbicli bere extends rigbt round tbe cell. Division 
takes place in a manner quite similar to tbat described above 
for Anabaena : tbe cell becomes constricted at its middle, wliilst 
tbe separating colourless mass (transverse wall) gradually appears 
(bg. 9). TJltimately bowever it develops to a far greater extent 
tban in tbe previous cases, so tbat tlie daugbter-cells become 
more or less widely separated and tbe open ceU-sbeatb gradu- 
ally closes in rigbt round eacb daugbter-cell. V e tbus see tbat 
tbe normal vegetative condition in a Gloeocapsa or Gloeothece 
presents tbe same structure as do tbe spores of an Anabaena. 
i. e. in its reproductive cells tbis latter genus reverts to tbe 
primitive tvpe of structure. wbicb probably appertained to its 
ancestors. Tbe fac-t tbat tbe spores of an Anabaena divide so 
as to form a blament is due in part to a condensation of tbe 
intercellular septum. in part to tbe loose difbuent cbaracter of 
tbe external sbeatb (cf. p. 45. 46). 
