BOTAVICAL SUBJECTS. 357 



identical in general structure. They both consist of an outer 

 layer of cells Avith inner layers which, by degrees, break up and 

 are replaced by globules of essential oil which occupies the centre 

 of the gland or juice vesicle, as seen in Fig. 127 (a). 



This is what Penzig says, p. 53 : " The oil-glands commence 

 by groups of cells which increase in number. As the gland 

 enlarges, the central cells become completely broken up, and the 

 central space becomes filled with essential oil ; so that the fully 

 formed gland has a central cavity filled with oil, an inner ^\i\\\ 

 formed of the brokcn-up cells, and an outer cell wall enveloping 

 the whole." 



At p. 92 he says : The juice vesicles (eniergeuze) commence 

 as nipples covered with epidermic cells containing groups of other 

 cells. The fully formed ones consist of epidermis which encloses 

 large cells, the outer ones of which contain small drops of oil 

 suspended in their juice. At maturity, however (p. 9-i), in those 

 varieties with abundant juice, the central cells get broken up and 

 the juice fills the central cavity of the pulp vesicles. 



This might appear not qta'fe like the structure of the oil-glands. 

 But in describing the juice vesicles of the (Egle sepiaria D.C. 

 {Citrus trifoliata, Lin.), p. 145, Penzig does not leave us in doubt 

 that their .structure corresponds almost perfectly with those of 

 Citrus ; their essential oil at maturity gathers in large drops in 

 the middle of the juice vesicles, destroyiiuj their central cells, as 

 in the oil-glands. In addition to oil they contain sugar, citric 

 a<'id, &c. 



So that, barring form and addition of acid-sweet juice, the 

 structure of the oil-glands of the leaves and peel of the Citrus 

 is essentially/ identical with that of the juice vesicles or hairs of 

 the pulp. 



Therefore, what I said in *' Cultiv. Or. and Lem. of India," &C., 

 viz., that in the Citrus pulp the oil-glands of the peel are replaced 

 by the juice vesicles or ]u\ce-hairs, as some botanist.^ call them, 

 is supix)rted by Penzig's analysis of their structure. 



Then both the oil-glands of the peel and the juice vesicles of 

 the pnlp have the same structure as the glandular hairs or dermal 

 warts of Dictamnus fraxinella, shown in Fig. \G0. 



In another part I have endeavoured to show that the hairs on 

 the bark and leaves, and especially glandular hairs, could only 

 have had their genesis in reprodvctirc hairs, sueh as we see in the 



