64 



destitute of roots, which grow entirely in water, and only so long 

 as they are entirely surrounded by this element, but they send 

 forth roots so soon as they come in contact with the soil. The 

 roots of the Riccia are simple, very rarely branched, and their 

 number varies, as in other plants, in different individuals. 

 These roots of the Riccice consist in mere simple radical fibrils, 

 as is the case in all mosses ; yet in Riccia natans, where several 

 deviations in form are already evident, these fibrils which are 

 developed in the earth are both articulated and unarticulated. 

 Such small variations occur even in more perfect plants ; now 

 and then even ramified fibrils are met with. But besides these 

 delicate rootlets, M. Lindenberg has also observed in some 

 species, in R. purpurascens and natans, stronger and more rigid 

 radicles, which at their extremity are incrassated in an oval, 

 clavate, or globular form, and then again emit from all sides 

 small slender fibres. He terms these roots sprouting roots, 

 and the thick tuberosities at the extremities are developed ac- 

 cording to his observations into new plants ; but they are in- 

 correctly designated as radical buds ; they are evidently gems, 

 the occurrence of which at the extremities of the roots of such 

 imperfect plants is very remarkable, as in Lemna alone has 

 anything similar been observed. In R. natans they are formed 

 even at the extremities of the simple radical fibrils, but the de- 

 tails have not been completely demonstrated. 



M. Lindenberg regards the Riccice as those inferior plants, 

 in which perfect cellular tissue first occurs, which can only be 

 admitted in a certain sense ; for not only in the Fungi, but also 

 in Algae, we find not merely very regularly formed cells, but 

 also a regular deposition of these cells on one another, ex- 

 actly as in the more perfect plants ; these cells differ from the 

 perfect cellular tissue merely in their contents and in the firm- 

 ness of their walls. M. Lindenberg thinks he has found in 

 the structure of the Riccice a confirmation of M. Kieser's 

 theory, " that the ideal primitive form of vegetable cells, forced 

 to deviate from their original form of a sphere or ellipsoid 

 in consequence of their being united into an uninterrupted 

 tissue, is the rhomboidal dodecahedron." All the cells, he ob- 

 serves, in the parenchyma of the Riccia are of this form, or 

 may be reduced to it, although at the same time the constant 



