OSCILLATORIACE^. 



[ 476 ] 



OSCILLATORIACEiE. 



posed into spiral fibrous structures (PI. 4. fig. 

 15; see Spiral structures). Some of the 

 remaining forms, included here for the pre- 

 sent, diff'er considerably from the above, and 

 are imperfectly understood. Vibrio (PI. 3. 

 figs. 18-21) consists of monihform filaments 

 without an apparent sheath. Spirulina (PL 

 3. fig. 15) has the (solid?) filaments curled 

 spirally, and in the strange plant Didymo- 

 helix (PI. 1. fig. 10) two spiral filaments oc- 

 curred twined together. These last minute 

 forms generally occur imbedded in a gelati- 

 nous stratum, but their relation to this is 

 not yet clearly ascertained. 



The structure of the Oscillatoriacese, 

 judging from Oscillatoria, Microcoleus, and 

 Lyngbya, differs importantly from that of 

 all other Confervoids. The filaments are 

 not composed of rows of cells, but, in the 

 earliest condition, of a cylindrical thread of 

 protoplasm, coloured greyish, green, brown, 

 or pm'ple in different cases. The ends of 

 growing filaments are narrower and devoid 

 of striae, and have no perceptible cellulose 

 sheath ; when a little older, cross strias ap- 

 pear, consisting of double rows of granules 

 or dots, and the tubular cellulose coat is 

 evident ; finally, the striae become distinct 

 lines (see PI. 4. figs. 8-22). In this stage, 

 external violence will cause the filament to 

 break across at the striae, and the fragments 

 then slide along inside the cellulose sheath, 

 the broken ends always assuming a rounded 

 form like that of the free extremities (PI. 4. 

 fig. 8 b). When these fragments slide quite 

 out of the sheaths, the latter appear as con- 

 tinuous tubes (PI. 4. fig. 8 a), seldom vtdth 

 any cross markings opposite the striae of the 

 internal mass. In Lyngbya the division seems 

 to take place in a peculiar manner, accom- 

 panied by an interstitial grow^th comparable 

 to that of Zygnema. In a well-developed 

 filament, every eighth stria is strongest, the 

 intermediate fourths rather lighter, every 

 second one between them paler still, and the 

 intermediates of these only just marked; 

 while in Oscillatoria the striae seem to be 

 gradually less definite towards the growing 

 apex of a filament. The filaments appear 

 solid as ordinarily viewed ; but the endos- 

 raose resulting from placing them in syrup 

 or gum-water causes them to contract be- 

 tween the striae, or to break up into lenticular 

 disks. The ultimate fate of all the filaments 

 seems to be a separation into disks or glo- 

 bular gonidia, by breaking across at the 

 striae. 



In Microcoleus (PI. 4. fig. 9) and many 



Rivularieae there w ould appear to be a trans- 

 verse multiplication hke that occurring oc- 

 casionally in NosToc, as the filaments are 

 found lying side by side in gelatinously de- 

 composed outer (parent) sheaths. The fila- 

 ments of the Rivularieae are seated on a 

 large basal cell (PL 4. figs. 13. 16. 18), the 

 nature of which is not understood. 



The remarkable spontaneous motion of 

 many Oscillatoriaceae presents a considerable 

 variety of conditions. In Oscillatoria and 

 Microcoleus the ends of the filaments emerge 

 from their sheaths, the young extremities 

 being apparently devoid of this coat ; their 

 ends wave backw^ards and forwards, some- ^ 

 what as the fore-part of the bodies of cer- 

 tain caterpillars are waved when they stand 

 on their pro-legs with the head reared up. 

 The filaments also emerge from the tubes 

 and break up, and the fragments then exhi- 

 bit an oscillating movement like that of a 

 balance, together with an advance in a lon- 

 gitudinal direction. Lyngbya (PL 4. fig. 10) 

 does not appear to oscillate, at all events 

 when in long filaments; but it presents a 

 curling snake-like or worm-like movement. 

 Vibrio, Spirulina, and other forms, exhibit 

 only a tremulous oscillation ; the same ap- 

 pears to be the case with Bacterium. These 

 last organisms were included by Ehrenberg 

 among the Infusoria, but there is every rea- 

 son to regard them as vegetables. Leptothrix 

 and the allied genera are very imperfectly 

 known, and are only included here from the 

 absence of indications of closer afiinities else- 

 where; veiy likely they are mycelial fila- 

 ments of Fungi. 



All these plants occur on damp ground, 

 rocks or stones, and among Mosses and 

 other Confervae on rocks, stones, &c., in 

 fresh and salt water, and are allied in some 

 respects to the Nostochace^, but the 

 articulations of the filaments of the latter 

 are all perfect cells with a complete cellulose 

 wall, multiplying by division in the same 

 way as the Confervaceae. 



Synopsis of British Genera. 



A. OscillatoriecB. Filaments transversely 

 striated or moniliform, sometimes spi- 

 rally curled ; sheathed, or in the minute 

 forms, without evident sheaths; exhi- 

 biting spontaneous oscillating, creeping, 

 or serpentine motion. Increased by 

 transverse division. 



I. Bacterium (PL 3. fig. 17). Filaments 

 extremely small, short, wand-shaped, or 



