RHODOSPERME.E. 71 



mule or vegetating bud. I do not pretend that analogical 

 inferences are here to be received as proof; I merely wish 

 to claim for them plausibility — and force, quite equal to that 

 of the evidence brought forward by the advocates of te- 

 traspores. 



Arguments in favour of the conceptacular fruit may be de- 

 rived from watching its developments and that of the frond, 

 and reasoning on the morphological relations of the parts. 

 Polysiphonia offers a peculiarly favourable field of observa- 

 tion. If we examine a young growing specimen of a plant 

 of this genus, we find the tips of all its branches to terminate 

 in a tuft of dichotomous fibres. These are peculiarly ob- 

 vious in P. Jihrata and P. Jihrillosa ; but will be found in 

 every species, if the specimen examined be in a sufficiently 

 young state. In P. hyssoides they are persistent, and found 

 at all ages on every part of the frond, constituting the single- 

 tubed ramelli of that species ; and in Dasya they form the 

 ramelli which clothe the branches. The branch which bears 

 these fibres or ramelli consists of a number of elongated 

 cells (siphons) placed, in a radiant manner, like the spokes 

 of a wheel, round a central cavity. Round the tips of the 

 branches these radiating cells are gradually shorter, and 

 each cell of the uppermost whorl or wheel ends in one of the 

 dichotomous fibres {ramelli). The rameUus never changes 

 its shape or character till it falls away, but the cells of 

 the branch below it lengthen, and grow wide till they assume 

 their proper size and shape. Growth, in this case therefore, 

 takes place below the insertion of the ramellus. Such is the 

 case in the primary branches. When a new lateral branch 

 is about to be given off from a primary one, a ramellus, simi- 

 lar to those at the apex of the old branch, makes its appear- 

 ance opposite to one of the dissepiments of the old branch. 

 At the base of this ramellus a cellular nucleus begins to be 

 formed, which increases in size and gradually assumes the 

 appearance of one of the ordinary branches, new ramelli 

 being developed at its apex as it acquires complexity. As 

 such ramelli are constantly found on all the growing apices, 

 it is natural to suppose that they are actively concerned in 

 causing the growth which takes place exactly at the point 

 of their attachment ; for, if they were unnecessary, we should 

 scarcely find them so universally present on growing points. 

 Besides, similar fibres occur on the young parts of other 

 Algae, especially of the Sporochnoidese and Dictyotacea^, in 

 the former of which orders they are evidently very essential 

 organs. Everything connected with these fibres — their pro- 



