On the Rhizocephalan Genus Thompsonia, etc. 27 
ancestor with many external sacs of normal Rhizocephalan type. A 
likely scheme of evolution is the following: 
The typical Rhizocephalan with single external sac and single root system. 
V 
Rhizocephalan with single root system but several external sacs of 
normal structure. 
V V 
Peltogaster socialis. Thompsonia. 
Separate external sacs and root A single root system and many externa 
systems. sacs of simple structure. 
In this remarkable life-history we find many phenomena which could 
hardly have been expected in so high a phylum as the Arthropoda. 
The analogy with the Fungi is strikingly shown in the mycelium-like 
root-system, producing its singular asexual reproductive organs, which 
are not dissimilar to the sporangia of a form like Mucor. A more con- 
vineing parallel may be drawn with a lower animal phylum, the Ceelen- 
terata, where typical Hydromedusan forms, like Obelia and Podocoryne, 
produce medusa buds with a complex structure and germ cells differ- 
entiated in situ; while there are other genera, like Eudendrium and 
Aglaophenia, in which the medusoid individuals are entirely degen- 
erate, with the simplest structure, and the germ cells are formed in 
the ccenosare and migrate thence into the medusoids. If Sacculina 
and Peltogaster correspond roughly to the former class, Thompsonia 
occupies a place similar to that of the latter. 
THOMPSONIA AND THYLACOPLETHUS. 
When Coutiére established his genus Thylacoplethus he remarked 
that in form it approaches Thompsonia globosa Kossmann: 
“Mais il doit en étre séparé génériquement, méme avant toute comparaison 
de la structure interne; Thompsonia a été trouvée fixée sur les pattes d’un 
Crabe (Melia tesselata Latr. provenant des Philippines), c’est-a-dire sur un 
Crustacé tres éloigné des Alphées dans la systématique, et d’autre part Thyla- 
coplethus offre le premier exemple d’un Rhizocéphale aussi completement 
gregaire.”’ 
In spite of the very incomplete description of Kossmann’s parasite, 
Hafele is almost certainly correct in identifying the Japanese form 
described by him as Thompsonia. There is only one serious discrep- 
ancy to be explained. In the type 7. globosa only two external sacs 
were found upon the original host, while 7’. japonica is as “ gregarious” 
as Coutiére’s Thylacoplethus. This is not a point, however, which 
weighs with me at all, for I know how easily these sacs are detached. 
There were probably a larger number present at the time of capture. 
The hosts were in both cases brachyurous Decapods and in both cases 
the parasites were attached to the thoracic legs. It seems, however, 
