MacMittan: OBSERVATIONS ON PTERYGOPHORA. 733 



gophora branch dichotomously arid build a strong holdfast re- 

 sembling that of Lessonia rather than that of Nercocystis. 

 Each young hapteric branch, in cross section, shows the char- 

 acteristic structure, a great central mass of parenchymatous 

 tissue surrounded by an ill-defined cortical area with an hypo- 

 dermal cambium. No pith is present and the growth in length 

 and thickness of the hapteres proceeds solely by cambial ac- 

 tivity. In older holdfasts distinct growth-rings appear some- 

 thing that was not seen in the holdfasts of Lessonia, and seems 

 not to occur in the hapteric branches of Nereocystis, the indi- 

 viduals of which are shorter-lived. The appearance of these 

 rings of growth in the secondary cortical tissues of the holdfast 

 seems to be due rather to rhythmic changes in the character of 

 the cell contents than to regular successions of larger and 

 smaller cells, concentrically arranged. As will appear, this 

 character serves to distinguish to some extent between the 

 growth-rings of the holdfast and of the stipe. In both organs 

 the elements of the secondary cortex are arranged in extremely 

 regular rows, as seen in cross section. This regularity of ar- 

 rangement does not extend to the primary parenchymatous 

 tissue of the hapteric branch, so that the appearance of a cross 

 section of the older hapteres may be described as follows : At 

 the center is a large more or less circular group of parenchy- 

 matous elements regularly hexagonal in outline, varying in size 

 between rather narrow limits. Towards the periphery this cen- 

 tral tissue imperceptibly merges with the secondary tissue, the 

 cells of which become more quadrate in outline and assume the 

 characteristic position in rows which can be followed without 

 break directly to the cambial zone which lies near the periphery 

 of the organ. In the secondary tissue there are numerous rings 

 of growth and the cross section of an old haptere, a centimeter 

 or more in diameter, looks not unlike a section of stipe, save 

 for the absence of the characteristic lenticular pith. This is 

 altogether wanting in the hapteres. The growth-rings do not, 

 however, appear to arise consistently through quite the same 

 anatomical conditions as those of the stipe. Well-marked rings 

 in the holdfast may exist without difference in the size of the 

 elements of which they are composed. The optical appear- 

 ance, therefore, is in all hapteres examined determined by dif- 

 ference in the cell contents. A zone of cells will be formed in 

 which the contents seem to be more dense. Outside a zone, in 



