no CHEADLE 



nucleoli may escape and retain their identity within the cell. The cytoplasm 

 becomes so thin and unstainable (with ordinary histological stains) that the 

 cells in a sense appear to be clear and often can be tentatively identified on 

 this basis. Plastids, or structures comparable to them, generally appear in the 

 sieve elements, but grains of typical starch do not form in these plastids. 



A fourth feature of sieve elements — the sieve areas — is associated chiefly 

 with the wall. Young sieve elements, and young parenchyma cells generally, 

 characteristically have thin areas in their walls through which narrow cyto- 

 plasmic strands (plasmodesmata) presumably interconnect with those of 

 similar neighboring cells. In sieve elements, the thin areas (primary pit 

 fields) become more clearly differentiated by an increase in the size of the 

 plasmodesmata and by the deposition around each of an elongate collar of 

 callose — the callose cylinder (fig. 19, 20, CAC). At this stage the plasmodes- 

 mata are best called connecting strands (fig. 19, 20, CS), according to Esau, 

 because they may consist of something more than typical cytoplasm (e.g., 

 slime). 



Sieve areas are presumably of great importance in vertical conduction. 

 One indication of this is that the sieve areas on the end walls of sieve ele- 

 ments in the angiosperms, particularly, are more highly differentiated (larger 

 connecting strands and callose cylinders) than they are elsewhere. So obvious 

 is this in most observed angiosperms that the sieve elements can be said to 

 be arranged in longitudinal series known as sieve tubes, each cell of which is 

 called a sieve-tube member (fig. 16, 18). (The length of sieve tubes is un- 

 known.) Sieve tubes are thus to the phloem what vessels are to the xylem. 

 Another indication of the importance of sieve areas in translocation is that 

 during dormancy of sieve elements, or during senescence, excessive amounts 

 of callose occur around the connecting strands and over the entire sieve area, 

 thus effectively reducing the size of, or even breaking, the interconnection 

 of the strands between two cells. In those plants in which sieve elements 

 become reactivated following dormancy, the callose is reduced in amount 

 and the connecting strands enlarge and apparently re-establish continuity 

 between adjacent sieve elements. On the other hand, at the cessation of con- 

 ducting activity of sieve elements, both the connecting strands and the callose 

 disappear, leaving empty pores in the sieve areas — at this stage they really 

 appear as sieves. The sieve elements at this time also become devoid of 

 cellular contents and may even be obviously filled with air; they may or may 

 not become crushed at this nonconducting stage, depending upon the species 

 and cellular make-up of the phloem. It is hardly conceivable, in view of the 

 changes just reviewed, that sieve areas are not intimately associated with 

 translocation. 



The four features (absence of nuclei, relatively thin, poorly staining cyto- 

 plasm, absence of typical starch, sieve areas) seem generally common to all 

 sieve elements, except for the inevitable minor exceptions (e.g., somewhat 



