SUPPLEMENT 17 



of similar form to the thickened regions, and alternate regularly with them, 

 but in reticulate types they are round or elliptical. 

 1. 24, for deliminated read delimited 



69, 1. i, for are on an average about I mm. long, 0-02 mm. broad read may 

 be 4 mm. long, 0-03 mm. broad, 



1. 7, for a millimetre apart read four millimetres 



I. 9, for fifty read one hundred and thirty-three 



II. 12-13, f or l fi t us turn . . . which may read STRASBURGER'S experiments 

 (1891) with the fresh wood of the silver fir may be readily understood ; they 

 may 



11. 25-45, for Similar results . . . uniform thickness ? read These experi- 

 ments do not indeed suffice to give us a numerical basis for estimating the 

 resistance offered by a thickened cell-wall and a bordered pit to the passage 

 of water, still they demonstrate how much the transit of water is facilitated 

 by the pits. Under these circumstances the question we have to ask ourselves 

 is, why is the whole wall not uniformly thin ? 



70, 11. 8, 9, for We need not . . . thickening read In spiral vessels also, which 

 frequently go on increasing greatly in length, the opening out of the spiral 

 band is a special adaptation, without which a vigorous extension would not be 

 possible. 



71, 1. 35 P. 72, 1. 22, The alternating . . . accepted as proved read The 

 alternating air-bubbles and water columns (the so-called JAMIN'S chain) show 

 considerable differences in size ; SCHWENDENER found the average length of 

 the air-bubble to be 0-3 mm. and of the water column 0-2 mm., and EWART 

 (1905, p. 76) records similar results. In such a system, movement of water 

 must obviously take place quite otherwise than in tubes filled with water from 

 end to end. 



One very important question, which doubtless can be solved experi- 

 mentally, is whether only those vessels which are full of water act as water- 

 conducting elements, or whether those which contain chains of air-bubbles and 

 water columns can carry out this function also. According to STRASBURGER'S 

 observations the amount of air in the vessels in the periphery of the stem is 

 less than that in the centre, and the youngest annual ring contains no air at 

 all. Could it be proved that this state of affairs was at all general (according 

 to EWART, 1905, the young vessels contain air, but his observations undoubtedly 

 require confirmation) it would be a strong argument in support of the cohesion 

 theory, and one could hold the view that the ascent of sap was rendered possible 

 by the cohesion of water particles ; but such a view can certainly not be 

 substantiated. 



72, 1. 34, for cavities of the vessels read cavities in a cell-wall 



11. 34-5, for in a microscopic sense read too small for the microscope 

 1. 42 P. 74, 1. 47, for Taking into account . . . superincumbent water 

 columns read Taking into account the sum total of our knowledge of the 

 subject, we must at least reckon with the possibility of the ascent of sap being 

 either exclusively or in part carried out by vessels containing JAMIN'S chains. 

 The extensive literature on the subject, into which we cannot enter, discusses, 

 sometimes critically (SCHWENDENER, 1983 ; STEINBRINCK, 1894), sometimes 

 in a more fanciful way, the problem how water can reach a height of 150 m. 

 in such a conduit purely by the suctional force of transpiration, some assuming 

 that the air-bubbles remain stationary, some that they move along with the 

 water columns. No final decision has been arrived at on the question ; it has 

 not been decided whether transpiration alone is capable of producing this 

 effect or whether other forces must be regarded as co-operating agents. In one 

 JOST B 



