502 RET. GEORGFE HEW SLOW ON A THEORETICAL 



one only completely investing the axis, terminating below and 

 rounding itself off, over the radicular extremity of the embryo. 

 Malpighi, I believe, first drew attention to it in Grasses, and 

 Hichard generalized upon it and thought all monocotyledons 

 were " endorhizal." Mirbel, however, disputed this conclusion, 

 and showed that many, as Palms, were exorhizal, and remarks 

 concerning the coleorhiza, " Elle appartient visiblement au coty- 

 ledon " *. That it is a different thing from the ruptured cortex 

 over all other roots is clear from the fact that the inner surface 

 of the tube of the coleorhiza is perfectly regular and provided 

 with what may be called an epithelium. Consequently it is not 

 at all covered with that irregular condition of the cells which 

 surrounds a secondary root w r hich has dissolved and absorbed 

 the cortex of the mother-root, through which it makes its way 

 out. Mirbel, however, elsewhere seemed to regard the coleorhiza 

 in this light, for he says : — " A bien considerer la coleorhiza, ce 

 n'est autre chose qu'une ecorce plus ou moins epaisse, qui se 

 detache d'elle-meme de chaque ma melon radiculaire " f. 



Fseudo-coleorliiza of Tropseolum. — Some light may be perhaps 

 thrown on the origin of the coleorhiza in endogens by the study 

 of the embryo of Tropaolum. In this plant the cotyledons are 

 very thick, a cross section of one being a semicircle ; each has 

 a groove down the middle. This indicates a conduplicate con- 

 dition, which is seen to characterize the other leaves in the 

 plumule. By making transverse sections from below upwards 

 to the insertion of the cotyledons, the apex of the radicle with 

 a pileorhiza of loose cells is observed quite free in the centre ; 

 while the thick sheath is made up of four "quadrants," united 

 in pairs : that is to say, they represent downward and pointed 

 prolongations of the two cotyledons, which are thus proved to be 

 sagittate in form J. The four " quadrants " curve inwards below, 

 nearly meeting over the tip of the radicle. These two pairs of 

 downward prolongations form the sheath or H pseudo-coleorhiza." 



Now, if we suppose them to be completely welded together, 

 they would form a perftct coleorhiza over the end of the radicle. 

 As a matter of fact they are so closely adpressed together that 

 they are in places near the circumference practically coherent ; 



* " Examen de la Division des Vegetaux en Endorhizes et Exorhizes/'p. 427. 

 t 'Elements de Botanique/ i. p. 81. 



J Cf. Irmisch's figure — "Trcypaolttm brathyceros, Hook., und tricolorum. 

 Sweet, nach ihrer Knollenbildung," Beitr. Vergl. Morph. Pfl., Taf. v. fig. 5. 



