400 MISS M. ROBERTSON ON A PROTOZOAN [Apr. 6 
is from these that this account of the parasite has to a large 
extent been drawn. 
The genus Jcehkthyosporidium belongs to the family of the 
Bertramide and to the order of the Haplosporidia. It was 
created by Caullery and Mesnil in 1905 for the two parasites 
Ichthyosporidium gasterophilum and TIchthyosporidium phymo- 
genes *. 
Ichthyosporidium gasterophilwm, which is closely allied to if not 
identical with the parasite in the sea-trout, is found in Motella 
mustela and Liparis vulgaris. 
These are both shore-living fishes and inhabit the Fucus-belt. 
Caullery and Mesnil note that J/otella mustela (the rockling) was 
most frequently infected and that the infected fish were all above 
a certain size. The parasite was found only in the pyloric ceca 
and the ducts of the glands of the stomach. The authors do not 
mention if the presence of the /ehthyosporidium seriously affected 
the health of the fish. 
The Jehthyosporidium of the sea-trout can easily be recognised 
in sections of the tissues as large spherical or elongated bodies 
(figs. 1-8). The material only shows the trophical stages. of the 
parasite and the fixation is not quite without reproach. 
The parasites are multinucleate organisms, most often spherical 
in shape and possessing a well-marked outer envelope. They 
range from large individuals with many hundred nuclei, measuring 
120 mm. in diameter, to small forms with a thin envelope and few 
nuclei. Individuals with only two nuclei are to be found and 
some apparently with a single nucleus, but it is difficult to be 
quite certain of these. They are very rare, and in section- 
material of this type an error in such a point is very easily 
made. 
The envelope is secreted by the parasite; the inner part is 
smooth and structureless except for occasional striations (figs. 6 
& 7). The outer part of the envelope is often slightly crinkled. 
The shape of the organism is open to very considerable variation ; 
oval, vermiform, and even irregularly branched forms may be 
found. These are for the most part individuals which have 
quitted their original envelope and grown out into the tissues. 
They seem to secrete a secondary envelope, which is generally 
rather thin; moreover, this may not extend over the whole 
animal. 
Sometimes these large irregular specimens seem to be breaking 
up into several daughter individuals by the very simple process 
of plasmotomy. The products of this process are of very varying 
sizes. It is a very casual method and does not involve nuclear 
division—it is, in fact, merely the breaking off of a mass of nu- 
cleated protoplasm. 
* Caullery and Mesnil, C. R. Soe. Biol. Paris, lviii. (1905) pp. 640-642. 
