AZOLLA. 



515 



yellow netted sac, to the upper end of which a grumous obsoletely 

 lobed mass is attached, in this confervoid mass, loose filaments are 

 seen* The yellow sac in the drawing, is under pressure and 

 shrivelled, it is loose, and may be dissected away when it separates 

 with the grumous mass, each lobe of which becomes a frond, looking 

 centrally to the punctate body. 



The original sac in both seems erect, that is, its lower margin is 

 not so definable as the upper. 



a. Cellular body. b. Grumous lobed body. c. Space empty. 



Fig. III. — In this the sphacelation of the apex is commencing and 

 spreading irregularly, the structure of the apex evidently of 4 teeth 

 as before closely connivent, the line of communication formed with the 

 funnel and granular mass, which is also very distinct, very plain. The 

 only difference, and it is a minor one, is in the left hand, or larger 

 body, the sac of which is larger and presents a diffused granular 

 grumous mass in the centre. To this, one filament adheres, but it 

 is not seen to penetrate. In the other the nucleus of the sac is 

 more defined and smaller, in the cells of the apices of both, a few 

 red nuclei are visible, confervoid filaments among male cells. 



Fig. IV. — Represents two bodies of one involucrum under pres- 

 sure, (the left hand and larger one entire, contains no green granules) 

 forming what I have elsewhere called the granular communication, for 

 these which filled the space between the foramen and apex of the nu- 

 cleus at first, have passed out, as is partly represented, the other is 

 split, and shews the internal structure well. I think in this case both 

 would have been yellow sacular ; the larger certainly would have been 

 so ; I have also represented a part of a confervoid filament (17) 

 found about the base of the bodies to be compared with those 

 squeezed out (18 and 19) of these, some are twice as long as others : 

 some look as though intending to coalescence into a mass. It is to 

 be remarked that in the filaments themselves the joints vary in 

 length, though perhaps none are so long as the largest of those 

 squeezed out. 



Great analogies with Cycas. ! ! 



Fig. V. — This represents a rare case, in which both pistilla appear 

 to produce the pedicellate bodies, in which alone the ordinary ternary 



