INTRODUCTION. XXXllI 



towards the darker edge of the vessel, sometimes they pro- 

 long their wandering com'ses, sometimes they remain in the 

 same place, causing their beak to vibrate in rapid circles. 

 Finally they collect in dense masses, containing innumera- 

 ble grains, and attach themselv^es to some extraneous body 

 at the bottom or on the surface of the water, where they 

 hasten to develope filaments like those of the mother plant. 

 The spherical sporules elongate at first into egg-shaped 

 bags, attached to the strange body by the narrowest end. 

 Their development only consists in a continual expansion, 

 without emitting any root. The green internal matter di- 

 vides in the middle by a partition, which appears at first 

 sight as a hyaline mucilage, but which gradually changes 

 into a complete diaphragm. It is thus, by successive divi- 

 sions ofthejoint first formed, that the young plant increases. 

 The position of the mamilla in each joint is uncertain, at 

 least I have seen it very different in neighbouring joints. 

 The exit of the sporules does not take place at the same 

 time in the different joints. One often sees those of one of 

 the articulations already escaped, while in the neighbouring 

 one they are not yet completely formed. Commonly the 

 uppermost joints empty themselves first, so that it is not 

 rare to see all the upper part of a filament entirely trans- 

 parent, whilst the lower part continues still to develope. 

 In this manner the formation and dissemination of the spo- 

 rules continues during the whole summer, and thus a single 

 filament suffices for the formation of an infinite quantity of 

 sporules. If one remembers that each joint contains per- 

 haps many hundreds of spores, it is not astonishing that 

 the water becomes perfectly coloured with them ; so that 

 we might readily take for a Protococcus, or other simple 

 Alga, what are only the spores of a Conferva. I suspect 

 that from such a mistake have arisen the theories of meta- 

 morphosis proposed by many modern algologists." 



c 



