THE METHOD OF MOVEMENT 135 



tube is cut. It is even possible that the walls are suffi- 

 ciently elastic and the internal pressures so great that 

 ordinary mounts of phloem tissues give a decidedly wrong 

 impression as to the normal size of the sieve-tube lumen, 

 the thickness of its walls, and the size of the sieve pores. 

 Crafts (1932) has recently observed that the supposedly 

 large pores that have been described in sieve plates of 

 Cucurhita are not large pores but extremely minute pores 

 less than 2/x in diameter and seemingly completely filled 

 with protoplasm. Schmidt (1917) also failed to find pores 

 through the sieve plates. What has appeared as large 

 pores, Crafts finds in his fresh specimens to be callose 

 masses which may have been dissolved out by fixing 

 reagents or which may have appeared as pores because 

 of lack of staining or improper focus. Although Crafts 

 has failed to find real pores of the size that have been 

 assumed to be present in sieve plates, the fact remains 

 that in making mounts for observation the phloem system 

 has always been cut into and high internal pressures have 

 been released so that perhaps no one has yet observed 

 sieve tubes and plates in their normal, distended condition. 

 Although externally applied pressure or suction has 

 failed to cause appreciable flow through phloem tissues, 

 there is fairly clear evidence that internally applied pres- 

 sure may, and at times does, cause considerable flow. For 

 example, many plants will show an exudation from the 

 phloem when that tissue is cut (Hartig, 1860, 1861 ; Miinch, 

 1930; and many others). The volume of exudate is so 

 great that much of it must have passed longitudinally 

 through the phloem for a considerable distance (Crafts, 

 1931; Dixon, 1933; and others). Therefore internally 

 applied pressures are capable of forcing materials through 

 the seemingly impervious phloem. Furthermore this offers 

 rather conclusive proof that either the pores of the sieve 

 plates are actually open and allow for mass flow of solution 

 through them or that Crafts (1931) is correct in his sug- 

 gestion that such flow occurs within the cell walls them- 

 selves. The former alternative seems to me much more 

 likely. 



