Morphology of the Citrus. 185 



that such a fruit was formed by the agglutination of 

 carpels, the originals of which may have been some- 

 thing like emarginate leaves. Of these there are many 

 representatives in almost every citrus tree. This, 

 however, might have originated otherwise. 



We have still the interesting structure of the citrus 

 pulp to account for. 



On one occasion while examining the transverse 

 section of a citrus fruit, it flashed across my mind that 

 the juice vesicles of the pulp carpels were only modi- 

 fications of the oil cells of the rind. Arguing back 

 from juice vesicles to rind oil cells, and thence to the 

 oil cells of the leaf, I did not see what else the vesicles 

 could be but modifications of the oil cells of the leaves 

 cr carpels ; that is, oil cells transformed to another 

 purpose, viz., the secretion of acid, sugar, and mucilage, 

 or whatever the juice vesicles may contain. Often 

 the juice of a citrus has an aroma sui generis, which 

 may possibly have some relation to that of the oil cells 

 of the rind covering. The slight bitterness which some 

 juices exhibit may also be derived from the bitterness 

 of the original inner whorl of rind, from which, in my 

 opinion, the pulp was evolved. In the juice vesicles, 

 we can, I think, also trace a repetition of the large and 

 small cells, although as these are squeezed together, 

 they are often out of shape. Fig. c, pi. 154, shows 

 the variously shaped juice-vesicles I took out of one of 

 the carpels of a citrus. The pedicels of the spindle 

 shaped ones I take to be the lengthened necks of the 

 larger rind cells, as seen in pi. 66, fig. b. Although 

 the oil cells of the rind are usually balloon-shaped, I 

 have met with some which were pointed, as shown in 

 fig. b, pi. 6, like the spindle shaped juice-vesicles. 

 PL 1 06, fig. c, and pi. i 24, fig. g, show other forms of 

 juice-vesicles, as well as pi. 209, figs, g and h. PI. 



