LATEX-HAIRS 345 



branches to the inner surface of the severed piece. In these circum- 

 stances the slightest mechanical injury to the leaf results in the 

 exudation of abundance of protective latex. In certain succulent 

 Euphorbias (E. ojjicinarum, E. canaricnsis), some of the distal branches 

 of the latex-tubes actually insinuate themselves between adjoining 

 epidermal cells, so that their tips come to lie immediately beneath 

 the cuticle. 



Finally, Stahl, Piccioli and Kny have shown, that in the involucral 

 bracts of species of Lactuca and other Cichoriaceae, the ultimate branches 

 of the laticiferous system develop into peculiar latex-hairs. These hairs 

 are from three to five times longer than their width, and may have 

 blunt or pointed ends ; as a rule their walls are heavily cutinised. 

 According to Zander, each hair arises, together with the pediment-cells 

 that half-surround its base, from a single epidermal mother-cell. 

 Morphologically, therefore, these organs represent genuine hairs, which 

 have become continuous with underlying latex-tubes, owing to the 

 absorption of the intervening basal wall, or else by perforation of the 

 pediment-cells. Usually each involucral bract bears from 60 to 100 

 latex-hairs. At the slightest touch these break and liberate drops of 

 latex. 



Among the various tubular structures which are more or less closely 

 related, both phylogenetically and physiologically, to genuine latex-tubes, 

 special mention may be made of the remarkable rows of vesicles that 

 occur in the bulb-scales of species of Allium. The cell-rows in question 

 traverse the scales in the longitudinal direction, lying close beneath 

 their outer surface ; towards the base of each scale the different rows 

 are connected by means of anastomoses. The individual sacs are some- 

 what elongated, and are separated from one another by densely pitted 

 walls. They are iilled with a turbid liquid resembling latex, which, 

 at the time when the bulb begins to sprout, appears to contain a large 

 amount of plastic material. In this connection we may also refer 

 to the similar latex-cells of the Convolvulaceae, which have been 

 investigated in detail by Czapek. The cells in question are arranged 

 in longitudinal series, and occur in stem, leaf and root, in close relation 

 to the vascular strands. Except in Dichondra, the transverse septa do 

 not undergo degeneration. The contents comprise a peripheral layer of 

 protoplasm, and "latex " of unknown composition. As each internode 

 completes its longitudinal growth, the latex-cells lose their contents and 

 collapse. 



