IN ORCHIDE.E AND ASCLEPIADE^. 513 



impregnation ; and in the latter state the utriculi are 

 marked with from one to three areolae of similar appear- 

 ance. 



The nucleus may even be supposed to exist in the pollen 

 of this family. In the early stages of its formation at 

 least a minute areola is ofteu visible in the simple grain, 

 and in each of the constituent parts or cells of the com- 

 pound grain. But these areolae may perhaps rather be 

 considered as merely the points of production of the tubes. 



This nucleus of the cell is not confined to Orchidese, Pis 

 but is equally manifest in many other Monocotyledonous 

 families ; and I have even found it, hitherto however in 

 very few cases, in the epidermis of Dicotyledonous plants ; 

 though in this primary division it may perhaps be said to 

 exist in the early stages of development of the pollen. 

 Among Monocotyledones the orders in which it is most 

 remarkable are Liliacese, Hemerocallidese, Asphodelese, 

 Iridese, and Commehnese. 



In some plants belonging to this last-mentioned family, 

 especially in Tradescautia vir(/inica and several nearly 

 related species, it is uncommonly distinct, not only in the 

 epidermis and in the jointed hairs of the filaments,^ but in 



' Tlie jointed liair of the filament in this genus forms one of tiie most 

 interesting microscopic objects with wliich I am acquainted, and that in three 

 different ways : 



1st. Its surface is marked with extremely fine longitudinal parallel equi- 

 distant lines or strise, whose intervals are equal from about l-15,000th to 

 l-20,000Lh of an incli. It might therefore in some cases be conveniently em- 

 ployed as a micrometer. 



2ndly. Tiie nucleus of tlifijointorjifill is very distinct as well as regular in 

 form, and by pressure is easily separated entire from the joint. It then 

 appears to be exactly round, nearly lenticular, and its granular matter is either 

 held together by a coagulated pulp not visibly granular, — or, which may be 

 considered equally probable, by an enveloping membrane. The analogy of this 

 nucleus to that existing in the various stages of development of tile cells in 

 which the grains of pollen are formed in the same species, is sufficiently 

 obvious. 



3rdly. In the joint when immersed in water, being at the same time freed 

 from air, and consequently made more transparent, a circulation of very minute 

 granular matter is visible to a lens magnifying from 300 to 400 times. This 

 motion of the granular fluid is seldom in one uniform circle, but frequently in 

 several apparently independent threads or currents : and these currents, 

 though often exactly longitudinal and consequently in the direction of the slrias 

 of the membrane, are not unfrequenlly observed forming various angles with 

 these striae. The smallest of the threads or streamlets appear to consist of a 



33 



