44 Puget Sound Marine Sta. Pub. Vol. 1, No. 8 



chambers. Type regions of the midrib were cut in cross-section also. 

 Safranin proved to be the best general stain, but for sieve tubes and 

 plates methylene blue was better. The best stain for sporophylls was 

 iron alum haemotoxylin. Camera lucida drawings of typical pith regions 

 throughout the transitional region were made. The mucilaginous matrix 

 stained so heavily that occasionally it slightly obscured the cell arrange- 

 ment. 



The structure of the stipe of A. fistulosa so closely corresponds 

 to that of the other Laminariaceae as reviewed by Sykes (4) that only 

 the briefest statement regarding it is necessary. The central or medullar 

 region includes the original pith area, and also the layers of cells added 

 to it by the inner cortical cells. Much of the later growth results from 

 the ingrowing of many cortical cells, forming filaments. These wander 

 among the original pith cells, forming a complex web. 



The three structural facts of prime importance from the stand- 

 point of this paper are: (a) early in the life of the plant the cells of 

 the medulla cease to divide transversely to the long axis of the plant, 

 as is reported in Macrocystis pyrifera (4), Laminaria saccharina (4), 

 and Sacchoriza dermatodea (2), while the cortical cells continue to divide 

 transversely; (&) this growth of outer tissues after the inner cells cease 

 dividing stretches the inner cells, producing a considerable degree of 

 elongation; (c) the cortical cells also continue to divide radially and 

 tangentially after the inner cells have ceased dividing. 



Sykes (4) states that only in Nereocystis and Macrocystis do true 

 sieve-tubes arise from the inner cortical cells. Methylene blue was the 

 only special stain used to bring out the callus regions, i. e., the parts 

 of the sieve-tubes in which cellulose undergoes hydration, preparatory 

 to the obliteration of the sieve-tubes. Hence the investigation was not 

 sufficient to determine with certainty whether the tubes in A. fistulosa were 

 true sieve-tubes. However, these tubes so fully correspond to the sieve- 

 tubes figured and described by Sykes (4), and so closely resemble those 

 found in the higher plants, that there seems to be no doubt that they 

 are true sieve-tubes. 



The elongated cells of the primary pith tissue are usually peculiarly 

 enlarged, hence they are often termed "trumpet-hyphae," although in 

 structure they are much like the sieve-tubes formed by the inner cortical 

 cells. Their only distinction of any consequence seems to be their dif- 

 ference in origin. Sykes (4) has shown in other forms of the Laminar- 

 iaceae that these "trumpet-hyphae" are rightly termed sieve-tubes, hence 

 the cells bearing sieve-plates, of both outer medulla and pith, will be re- 

 ferred to in this paper as sieve-tubes. 



The pith area in the stipe of A. fistulosa runs almost wholly across 



