94 SIEVE-TUBE ELEMENTS 



cells ill the gymnosperms lack companion cells in the above sense, 

 but specialized "albuminous cells" occur and have been regarded 

 as equivalent to the companion cells in angiosperms. However, 

 Abbe and Crafts (1939, p. 710) have questioned this analogy since 

 the albuminous cells "do not arise from a common fusiform 

 initial with the sieve-tube but from separate ray initials." The 

 primary phloem is usually relatively simple in structure, con- 

 sisting of sieve-tubes, companion cells and phloem parenchyma 

 as in Cucurbit a or only the first two types of cells may be present 

 (e.g., Zca Mays). But secondary phloem may be very complex 

 because of the presence of fibers, stone cells, vertical phloem 

 parenchyma and rays in addition to sieve-tubes and companion 

 cells. (Cf. Eames and MacDaniels, Ch. VIII, and Esau, 1939, 

 pp. 411-413.) 



Aside from the secondary functions of food storage and sup- 

 port, the chief physiological role of phloem is the conduction of 

 various organic solutes. Experimental studies on translocation 

 (cf. Crafts, 1939a and 1939b) seem to indicate clearly that the 

 main channels of transport for organic materials, and also of 

 certain viruses are the sieve-tube elements. As to the mechanism 

 of this movement in sieve-tubes there is, however, considerable 

 disagreement. Crafts (1939a) on the basis of a wide series of 

 recent studies on this problem has developed a "pressure flow" 

 theory which he summarizes as follows: "In the pressure flow 

 mechanism, solvent and solute are assumed to flow together as 

 a solution thi-ough elements of specialized structure or per- 

 meability, the i)rotop]asm of which plays an entirely passive role 

 in the process. The sources of energy in this mechanism lie in 

 the osmotic activity of the products of assimilation in green por- 

 tions of the plant and in the accumulative ability of cells in grow- 

 ing and storing tissues." Other explanations of the mechanism 

 of transport in sieve-tube elements assume that protoplasmic 

 streaming or highly-active protoplasm is concerned in some way 

 with the process. One result of the lively interest in the function 

 of phloem has been a series of exploratory studies on the structure 

 and deveh)pment of sieve-tube elements (cf. Esau, 1938). The 

 recent literature in this field, as well as tlic historical background 

 of the various problems, have been discussed recently by Esau 



