132 Gr7ibb.—The Attachments of Porphyra nmbilicalis. 
under the name of P. linearis , Grev., but Harvey (6) proposed that it 
should be included as a var. linearis of P. vulgaris, as he considered it to be 
only a growth-form of that type ; this conclusion is fully upheld by the 
succession of forms under observation at Swanage. 
The majority of the very narrow fronds are fertile and remain so 
throughout the season. On the rocks near high-water level the fronds 
hardly increase at all in size during the winter and spring, but remain 
approximately 5 to 15 cm. in length and 0*3 to 1 cm. in breadth. 
Porphyra is by no means confined to the upper rocks, but the purple sheets 
stretch down over the ridges of the rocks to below mid-tide level, even 
extending to the exposed Himanthalia zone. In the lower reaches 
a gradual succession from the linearis to the typical vulgaris state is 
noted, the latter being fully attained about April, when var. laciniata is also 
found growing on these rocks in small quantities. Both varieties, however, 
disappear rapidly towards the end of May and the beginning of June, and 
by the end of June there is hardly a frond to be found. The time at 
which this disappearance takes place seems to be determined by the 
weather, for from all accounts Porphyra appears to be unable to withstand 
the heat of the south coast, and in districts where it does exist all the year 
round, i. e. Clare Island ( 3 ) and the Faroes ( 2 ), its survival is probably 
due to the cooler climate and the spray with which it is saturated even at 
low water. 
3. Material and Methods. 
% 
The long thin fronds of Porphyra hanging down the sides of the rocks 
are only attached by minute adhesive discs, and these were detached by 
scraping the rocks with a knife. The fronds and discs were fixed on the 
spot in weak Flemming’s solution in sea-water, allowing the fixative to act 
for thirty minutes only. In the laboratory the discs were examined and 
some were embedded in paraffin in the usual way ; microtome sections 
from 3 \x to 5 \x in thickness were cut in a longitudinal direction, and were 
stained in Heidenhain’s iron-alum-haematoxylin (twenty-four hours), 
counterstaining in safranin in 70 per cent, alcohol (thirty minutes). Hand 
sections were also examined and preparations made from discs, which in 
some cases were boiled in caustic potash to separate the tissues. This, 
however, was not found to be a satisfactory method, as although it fulfilled 
its obj,ect in removing the gelatinous ‘cuticle’ (1, p. 5) which surrounds 
the tissues, yet at the same time it causes the hypha-like filament 
composing the disc to swell up and assume new forms, so that it was 
impossible to tell whether one were examining an artificial or the natural 
state of these threads. Instead preparations were made by teasing and 
pressing the disc structures until the tissues had become separated one 
from another and then staining in aqueous safranin. 
