230 Miscellaneous. 



persistence of this septum, the anther continues quadricellular. In 

 this case, usually {Lycopersicon, Tradescantia, &c.), two half-valves 

 rest by their commissure upon the septum, which, after their dehis- 

 cence, becomes contracted or destroyed; and at this moment .it 

 might be supposed that the separation or destruction of the septum 

 preceded, instead of following, the dehiscence. A second type is 

 furnished by Mchmea, in which each of the four loculi splits in its 

 median line. A third type of quadricellular anthers is presented in 

 Passiflora, Scabiosa, Schaueria, &c. : in these the subdivision of 

 each lobe is maintained until the dehiscence ; but this is less by the 

 septa, which are too short, than by the contiguous valves reflected 

 and applied against each other. 



On the question whether all the subepidermic utricles become 

 changed into filamentous cells (as would appear to be the case from 

 the statements of Mirbel, Meyen, and Purkinje), the author says 

 that, by tracing the development of the tissues of the anther in 

 Tradescantia, it is distinctly seen that of two layers of utricles situ- 

 ated beneath the epidermis, only one (the outer) is converted into 

 filamentous cells, whilst the inner layer is destroyed. This tissue 

 within the endotheciutn of Purkinje was perceived in the young 

 anther by Mirbel and Meyen, who paid no further attention to it. 

 It is more distinct still in Passiflora, where its utricles, which alone 

 are tinted yellow, papilliform, and radiate, are already distinct at the 

 appearance of the pollinic utricles, and are developed paralleliy to 

 these and the pollen, but disappear a little after the production of 

 the filaments in the utricles of the middle zone. 



In Tradescantia and Passiflora, as in most plants observed by the 

 author, the internal tissue is not tranformed ; it is destroyed after a 

 transitory existence. In other plants also it is not transformed, but 

 persists until the dehiscence of the anther {Canna, Colchicum, Pedi- 

 cularis, &c.); and this persistence is general (?) in the anthers which 

 are destitute of filamentous cells {Pyrola, Melastoma, &c.). 



The tissue indicated as being more interior than the membrane 

 called endothecium, and as lining the cavity of the cells, is not 

 only characterized by its position and evolution, but also frequently 

 by the form, consistence, and coloration of its utricles. It is 

 as distinct from the endothecium of Purkinje as the latter from the 

 exothecium, and is, in fact, a third membrane, which must hence- 

 forward be included in the general structure of the anther. This 

 third membrane, from its position, will be the true endothecium, 

 the membrane so named by Purkinje becoming the mesothecium ; 

 and thus the anther, at least when young, consists, not of two, but 

 of three membranes. Nevertheless, at the approach of dehiscence, 

 these membranes may be reduced to one in the anthers of some spe- 

 cies (^Calendula, &c.) ; and in some plants even the young anthers 

 only contain two layers of utricles. 



With regard to the conversion of the simple utricles into filamen- 

 tous cells, the author states that, although rapid, it may be followed, 

 and usually commences in the anther at its point of attachment and 

 at its line of dehiscence. 



