1918] Gardner: New Pacific Coast Marine Algae II 437 



the genus. In liis notes on the species {loc. cit., p. 75) he states 

 distinctl}' that vegetative cell divisions occur perpendicular to the 

 substratum and in that direction only, and hence colonies consisting 

 of only one layer of cells can be produced in this genus. Examination 

 of a part of the type material, Coleanema arenifera, reveals this con- 

 dition to be correct. It is principally upon that character tliat the 

 genus may be distinguished from Plcurocapsa, which has vegetative 

 cell divisions in three planes, although Achille Forti says in his diag- 

 nosis of Xenococcus that it also divides vegetatively in three planes. 

 (Forti. in De Toni, Syll. Alg., vol. 5, p. 133). Both Xenococcus and 

 Pleurocapsa reproduce by gonidia, Batters having discovered them 

 in X. Schoushoci Thur. X. chaetomorphae conforms to the Xeiw- 

 cocus method of vegetative growth rather than to that of Pleurocapsa, 

 and hence is here placed in that genus. 



This species was found associated with Derniocarpa pacifica, and 

 it is somewhat difficult to distinguish the two species. Both species 

 produce gonidia in great abundance, and plants of both species may 

 be found in all stages of growth at the same time. They are very 

 abundant, intermingled, and as seen from above are very angular on 

 account of lateral pressure resulting from growth. There appears 

 certainly to be vegetative growth in the Xenococcus, as the continuous 

 areas covered by it, comprising many hundreds of cells, seem far too 

 great to have arisen from a single group of gonidia, and should the 

 gonidia escape singly they could never by chance become so uniformlj- 

 and closely associated as they often are to be found. The cells of 

 Xenococcus in all stages of growth are extremely angular and very 

 variable in shape and size ; sometimes being much crowded at the 

 cross walls of the host plant, the tendency is to elongate vertically. 

 Frequently groups of cells seem to have started to grow on tlie cell 

 walls of young cells of the host between the cross walls, and as the 

 host cell elongates the cells of the epiphyte seem to elongate abnorm- 

 ally in the direction of the long diameter of the host (pi. 36, fig. 3). 

 There are no particular shapes or sizes of gonidangia, since any of 

 the long narrow cells, small angular cells, or large spherical cells maj' 

 be transformed into gonidangia. The gonidia are formed by succes- 

 sive divisions of the whole of the protoplast (pi. 36, fig. 4). On the 

 whole it seems almost certain that we have two plants of ditferent 

 genera closely and intimately associated here, and it certainly is not 

 possible to distinguish the individuals of the two in all stages of their 

 development. I have previously referred to this difficulty in the 

 introduction (p. 429). 



