208 
PACIFIC SCIENCE, Vol. XVIII, April 1964 
MATERIALS AND METHODS 
The collections of Polysiphonia in Dr. M. S. 
Doty’s herbarium ( DUH ) ; Polysiphonia col- 
lections deposited at the Bernice P. Bishop 
Museum (BISH) by I. Abbott, E. Bailey, J. 
Rock, D. Rogers, and J. Tilden; and my own 
collections around Oahu (M) were the materials 
used in this study. 
A. Collection of the Biological Materials 
As much area as possible was covered at 
every station. The materials collected were first 
placed in plastic bags and later transferred to 
wide-mouthed bottles filled with 10 per cent 
formalin. 
B. Investigation of the Biological Materials in 
the Laboratory 
Little difficulty was met in the preparation 
of the materials for the detailed studies. There 
were three types of preserved materials: old 
mounted dry specimens, materials preserved in 
formalin for several years, and fresh collections 
in 10 per cent formalin. 
A drop of water was added to each of the old 
mounted specimens to make them soft enough 
to avoid damaging any structures when prepar- 
ing samples for microscopic study. It was usually 
necessary to boil such dry specimens for a few 
minutes to restore the normal cell shape. On 
the other hand, the old specimens preserved in 
formalin did not need to be boiled but had to 
be stained to distinguish the pericentral cells, 
using 1 per cent aniline blue or 1 per cent 
safranin O. The newly collected materials were 
softened in sea water-formalin solution for 24— 
48 hours before examination. 
A dozen or more slides of each collection 
were prepared. A few thallus fragments were 
taken from each collection and placed on one 
or more slides with a drop of 10 per cent 
glycerine solution. 
For counting the number of pericentral cells 
the materials were either crushed on the slide 
or cross-sections were cut. Important charac- 
teristics of the different species were illustrated 
with a camera lucida. Measurements were made 
with an ocular micrometer. 
THE GENUS Polysiphonia 
The genus Polysiphonia, established by 
Greville in 1824, belongs to the family Rho- 
domelaceae in the Rhodophyta. According to 
Falkenberg (1901), Polysiphonia should include 
all those radially symmetrical members of the 
Rhodomelaceae in which ( 1 ) at least the ulti- 
mate branches are evidently polysiphonous, ( 2 ) 
most of the branches arise exogenously by a 
more or less diagonal division of subapical cells 
before these have cut off pericentral cells, (3) 
all branches are essentially similar and inde- 
terminate, and (4) only one tetrasporangium is 
borne normally in each segment. 
IMPORTANT TAXONOMIC FEATURES 
A. Trichoblasts 
Rosenvinge (1903) referred to the structure 
that other investigators have called "leaves” as 
"trichoblasts.” Commonly, one trichoblast occurs 
on each segment. Hollenberg (1942) com- 
mented on the trichoblasts as being usually ar- 
ranged spirally in a counter-clockwise direction 
toward the tip of the branch, looking at the 
branch from the apex. In the present paper this 
would be referred to as a right-hand spiral. 
Hollenberg also stated that the divergence of 
the trichoblasts in the spiral is relatively con- 
stant and usually bears some relation to the 
number of pericentral cells. The trichoblasts 
may be simple or forked, consisting of one to 
several segments of cells. They are usually pres- 
ent at or near the apices. Generally, they are 
deciduous, leaving conspicuous scar cells after 
they fall off. 
B. Branches 
In the Rhodomelaceae, Falkenberg (1901) 
distinguished two types of branches: (1) de- 
terminate branches, which do not ordinarily 
give rise to further branches, and (2) inde- 
terminate branches, those of potentially un- 
limited growth. The branches are either of 
endogenous or exogenous origin. Branches aris- 
ing after the pericentral cells have been formed 
are designated as "endogenous,” whereas those 
formed before the formation of the pericentral 
cells are referred to as "exogenous.” The branches 
