268 THE JOURNAL OF BOTANY 
colour,” and shining. 
I take no account of the land and water states of these 
plants, since they are merely temporary conditions directly in- 
duced by the environment, and not varieties. 
T have not seen the original description of var. chondrosperma, 
Fenzl, and for the meantime have relied on that given by Rouy 
and Foucaud—“ Graines opaques, fortement tuberculeuse.”’ 
He describes the seeds as larger, “more inclining to claret- 
hinj 
TARAXACUM SPECTABILE, Dahlst., ssp. GEIRHILDa, Beeby, ined. 
Differs from the type in the leaves being always undivided, 
strongly with the dark, brown-red midri - The flowers are one- 
fifth to one-fourth more in diameter than those of the type when 
grown side by side; of a darker (browner) yellow; and copiously 
urnished with pollen, while the type is epolliniferous. The 
phyllaries and fruits are similar to those of the type. 
The name has indirect reference to one of the localities, the 
Loch of Girlsta, formerly Geirhildarstadr; here Geirhild, Hrafna 
Floke’s daughter, was drowned when accompanying her father 
on his voyage to Iceland about the year 870, as we are told in 
Landndmabok. 
The sub-species occurs plentifully on rock ledges, among 
heather, and in grassy places by the east side of Lang Klédi Loch, 
Northmaven ; also among rocks and heather by the Loch of 
Girlsta, and elsewhere in Nesting and Weisdale 
I 
when I foun pollen-grains on the stigmas; but as the ot was 
A NEW SIPHONEOUS ALGA. 
By A. & E. S. Gepp, 
SoME seven years ago we prepared a monograph of the genus 
Udotea in connection with the Reports of the “ Siboga” Expedi- 
tion to the Dutch East Indies, but decided to postpone its pub- 
7 
text of this paper is practically ready for publication, but as other 
algologists are working at the same group of plants from the West 
Indies, we think it opportune to publish a preliminary diagnosis 
