497 
with the latter family, from which it differs among other things by a four-celled 
carpogonial filament and by the first gonimoblast-cell being cut off on the outer 
side of the auxiliary cell. In these characters it agrees better with the Rhodomelacew, 
but as, on the other hand, it differs very much from this family in the structure 
of the frond, it seems impossible to refer it to it!. 
E. CHEMIN in 1927 made the surprising communication that he had obtained 
the germination of the tetraspores of Harveyella mirabilis on a glass-plate. The ripe 
spores were without the slightest trace of pigment; they surrounded themselves with 
a membrane and were divided in 48 hours by successive bipartitions into a paren- 
chymatous disc much resembling that of Chondrus crispus. 
The French author remarks that the germinating spore 
begins to produce phycoérythrine immediately after the 
fixation on the substratum, and the cellular disc taking its 
rise from it is coloured intensely red. This very interesting 
statement is so unexpected that it requires confirmation. 
If this can be obtained, a relationship with the Rhodome- 
laceæ is decidedly excluded’. 
Harveyella mirabilis has been collected in all seasons 
in nearly all the inner Danish waters, most frequently in 
spring, always growing on Rhodomela subfusca (incl. R. vir- 
gata) at 4—26 metres’ depth. Antheridia have been met 
with in July (Bm and Bb), September (Kn, Lb), October 
(Ke, Su) and once in January (Kn). Procarp-bearing speci- Fig. 457. 
Harveyella mirabilis. Tetraspo- 
rangia in various stages of divi- 
once in October (Su) and once in April in one specimen sion. 835 : 1. 
together with tetrasporiferous ones, probably a retarded 
abortive specimen. Cystocarps were met with in December and January. Tetra- 
sporiferous cushions were frequently found in April to June, in July once only in 
the Northern Kattegat, but several specimens at Bornholm, and finally once in Au- 
gust in the Great Belt and once in October in the Sound. The species is most 
frequently met with in spring (May) and then always with tetrasporangia. 
It would be of interest to compare this occurrence with that on other coasts 
mens may probably be sought in autumn; I met with them 
1 It is remarkable that Srurcn in 1924 described another species, H. pachydermus, parasitic 
on Gracilaria confervoides; it differs by having a two-celled carpogonial branch and by numerous 
fusions of the ooblastema with cells of the gametophyte. As the number of cells in the carpogonial 
branch as a rule is a good systematic character, it ought to be examined 1) whether this number is 
really constant and 2) whether this species is rightly referred to the genus Harveyella. 
? To verify the statement of CHEMIN, I searched for specimens with ripe tetrasporangia in 
May 1929 and in June 1930, but in vain. In June 1930 the season of fructification seemed to be 
closed for that year, for although I dredged in several places where I- might reasonably expect to find 
fructiferous specimens of Harveyella, only few specimens were met with and none with well devel- 
oped ripe tetrasporangia, and the examination of the germination of the spores could not, there- 
fore, be made. 
D.K. D. Vidensk. Selsk. Skr., 7. Række, naturvidensk. og mathem. Afd., VII, 4. 64 
