150 D. H. SCOTT. 



lateral protvusions, which penetrate between the parenchyma 

 cells. In a great many cases they end blind. The points 

 at which they are formed appear to be determined by the 

 pressure due to the turgesceuce of the surrounding cells, 

 that is to say, they are formed where the resistance to their 

 growth is least. 



When the branches formed in this way happen to 

 meet the branches or trunks of other latex-vessels their 

 walls are absorbed at the point of contact, and their contents 

 become continuous. The appearance presented recalls the 

 phenomena of conjugation among the algae — a comparison 

 which was already made by Schacht. An important differ- 

 ence between the two processes is, however, to be found in 

 the fact that a laticiferous vessel produces these outgrov/ths 

 quite independently of the presence of a neighbouring 

 vessel. 



The development of the laticiferous vessels in the em- 

 bryonal tissue of older plants agrees, so far as my observa- 

 tions extend, with that in the seedling. The vessels very 

 generally have a crooked course, which is partly explained 

 by the fact that the cells which take part in the formation 

 of one and the same vessel do not necessarily lie in the same 

 longitudinal row. It is quite usual for cells situated in con- 

 tiguous rows, and at somewhat different levels, to undergo 

 fusion by absorption of that part of the side-wall which is 

 common to both. In this case parts of the latex-vessel 

 will lie in the same straight line with cells which belong to 

 the parenchyma, the vessel itself having a broken course. 



With reference to the subject of this paper, I further 

 examined the secondary cortex of Scorzonera, Taraxacum, 

 and Chelidonium. In the root of Scorzonera laticiferous 

 vessels are formed from the cambium in great abundance. 

 In cross-section they appear ranged in radial rows, which 

 are not usually separated by more than six layers of paren- 

 chyma cells. As a rule each radial row is double. The 

 latex-vessels are accompanied by sieve-tubes, Avhich are 

 fairly numerous. In a tangential section it can be seen 

 that the portions of tissue containing the latex-vessels and 

 sieve-tubes form a coarse network, the meshes being occupied 

 by the cortical rays. Frequent anastomoses between the 

 vessels occur, both in the radial and tangential direction, 

 some being direct, while others are effected by means of out- 

 growths. The cross-walls are absorbed very early. It is 

 only quite near to the true cambial layer that one finds 

 them still present. The early stages of development are 



