72 



One. Since the publication of the remarks on R. sativus, Linn, in 

 the Phytographia Canadensis (pag. 83, tab. 8 A, fig. 18-24). Pro- 

 fessors Fischer and Meyer ( Index 4 sent, pag. 44 ) have formed the 

 Raphanistrum Gayanum from the variety therein named of R. sativus 

 which is distinguished by the articulation at the base of the pod, for 

 1here is no other difference. I have since carefully cultivated several 

 garden sorts, and they have yielded me these two varieties indiffe- 

 rently at different times, but when left to sow themselves they have 

 assumed entirely articulated pods, frequently even strangulated be- 

 tween the seeds as in Raphanistrum. I am inclined therefore to con- 

 clude I hat this second variety is the normal form of the species though 

 to distinguish it, it must continue to bear the name of the learned 

 and accurate observer who first noticed it. My experience as yet has 

 been as follows. Of six varieties bought at a seed shop the yellow (1) 

 and violet alone produced all their pods articulated, the large va- 

 riety called raifort produced some plants with articulated, some with 

 entire pods. On the following year the first two varieties, bought at a 

 different shop, produced very few plants with articulated pods, 

 whilst the grey variety had all its pods articulated, The disappear- 

 ance of the lower articulation is effected, not by its abortion, but as 

 may be easily seen in several varieties, by its coalition and fusion 

 with the upper, a process which takes place likewise in the beak of 

 the fruit of Sinapis arvetisis, Linn. The dissepiment confounded by me 

 formerly with the enclosing membranes (Phyt. Can. p. 84) , exists, 

 and the pod is decidedly bilocular. 



Erucaria Aleppica. Gaert. 



Hab. Inmontosis circa Onilet alibi in Baetica. 

 HlRSCHFELDIA ADPRESSA. Moencll. 



Sinapis incana. Linn. 



Hab. Ubique in Hispania et Lusitania meridionali . 



Obs. Hirschfeldia differs only from Erucaria by the conformation of 

 its embryo. The radicule is incumbent in both, but in Hirschfeldia 

 the cotyledons are conduplicated, that is their sides are folded in- 

 wards as in Sinapis, whereas in Erucaria their upper extremity is fol- 

 ded downwards, so as to form whith the radicule a spiral convolu- 

 tion. This is the only essential difference between the twogenera, and 

 I know not whether it is sufficient to keep them asunder. This will 

 depend on the importance which shall be found to belong to the pli- 

 cations of the embryo in the formation of natural groups in this fa- 

 mily, being the divers means which nature has imagined to pack it in 

 a convenient compass. It remains to be seen whether she has 

 accomplished this indifferently in various ways in all the most na- 

 tural groups. 



Brassica Cheiranthos. Vill. 

 (3 montana. DC. 

 Hab. a In Lusitaniae montibus Lunae. fJ In alpibus Eliberitanis. 



(1) The names used by gardeners allude to the colour of the root. 



