164 TRANSLOCATION IN PLANTS 



These many weaknesses in the hypothesis of Mlinch, and 

 the strong opposing evidence lead me to the conclusion 

 that the hypothesis as it now stands is not tenable. 



31. The Hypothesis of Crafts.— Crafts (1931, 1932, 1933) 

 has proposed a mechanism that, in many respects, is 

 similar to that of Miinch. Its main difference lies in his 

 assumption that the entire cross section of the phloem, 

 including especially the walls, is effective as a channel for 

 transport. He was led to this suggestion for several 

 reasons. He found that a solution containing about 8 to 

 10 per cent dry matter exuded from the cut phloem of 

 cucurbits. Kraus (1886) found approximately the same 

 concentration of material, while Miinch (1930) found from 

 12.3 to 23.7 per cent dry matter (14 to 31.2 per cent if 

 expressed as per cent of water as he gives it) in material 

 flowing from cut phloem of several species of trees. Crafts 

 estimated that this material flowed through the phloem 

 of Cucurhita at an average linear rate equal to a linear 

 displacement of 0.3 cm. per minute through the total 

 phloem. On first cutting, a rate nearer 1 cm. per minute 

 was obtained. The sieve-tube lumen occupies less than 

 20 per cent of the total phloem area and the pores only 

 8 per cent of this, or not over 1.6 per cent of the total 

 phloem area. As a result of more detailed studies of the 

 sieve plates (1932) he concluded that the pores were not 

 actual openings but were completely filled with proto- 

 plasmic strands with a much smaller cross section than 

 this estimated area. Applying Poiseuille's formula and 

 assuming the sieve pores to have a diameter of 2/x, which 

 was larger than what he actually found, he calculated that, 

 in order to bring about a flow of solution through the sieve 

 pores at the rate at which material was actually collected, a 

 pressure gradient of 0.2 atmosphere per centimeter would 

 be required. No such pressures, however, seem available. 

 Measurements before and after dehydration of the 

 phloem demonstrated that the fresh phloem walls contain 

 over 50 per cent water and also occupy over 30 per cent 

 of the total cross-sectional area of the phloem. He 



