28 Papers from the Department of Marine Biology. 



that Thompsonia is not like the other genera of Rhizocephala, the 

 occurrence of which is confined in each case within a very narrow 

 division of the Decapods, but enjoys a very wide diversity of hospi- 

 tality; for Kriiger reports that not only does T. japonica parasitise 

 Brachyura, but is even found upon a hermit crab (Pagurus striatus). 

 The two points, then, upon which Coutiere lays such great stress are 

 shown to be without importance by later work. 



Why, then, did Hafele not recognise the identity of Thompsonia and 

 Thylacoplethus? His negative decision rests upon points of difference 

 in structure, but he makes some reservation owing to the unsatisfactory 

 nature of Coutiere's unillustrated account. I will quote his concluding 

 sentence: 



"Gerade aber der Umstand, dass Coutiere von einem 'Manteau a double 

 paroi chitineux dans lequel est suspendue une masse viscerale, et qui porte 

 une ouverture cloacale' spricht, lasst mich mit ziemlichen Sicherheit vermuten, 

 class in Thylacoplethus eine ganz andere Gattung vorliegt." 



I am at a loss to understand why Hafele did not find an inner layer 

 of chitin (the internal boundary of the mantle), but I think it will be 

 found to exist in T. globosa when better preserved material is examined. 

 With regard to the existence or non-existence of a cloacal aperture, 

 this apparent difference is due to the fact that the organ appears only at a 

 very late stage in development. But in any case I hardly think that, 

 in view of the obvious agreement in structure and development, there 

 could be any reason for separating generically Thompsonia (as described 

 by Hafele) and the forms parasitic on Alpheids (as described by 

 Coutiere and myself). My own observations fail to reveal any differ- 

 ence between the parasites of Brachyura and Alpheids and an exam- 

 ination of the figures given by Hafele for Thompsonia globosa (for stages 

 which the parasite of Thalamita does not show) and comparison with 

 sections of corresponding stages in the Alpheid parasites show an 

 essential similarity. 



There is one point to which I refer further in the next section in 

 Coutiere's description of Thylacoplethus, but may be briefly mentioned 

 here. It is the occurrence of parasites on the sternum and not on the 

 appendages of the host. But this difference can hardly be conceived 

 to be of generic value. 



AMENDED DIAGNOSIS OF GENUS. 



Thompsonia Kossmann ( = Thylacoplethus Coutiere). 



A colonial Rhizocephalan infecting various Decapod Crustacea. Root 

 system diffuse, widely distributed in body of host, sending branches into 

 thoracic and abdominal appendages, which give off numerous sacs containing 

 reproductive cells and becoming external at a moult of the host. These 

 external sacs consist of a mantle and visceral mass without an intervening 

 mantle cavity : mantle thin and devoid of muscle fibres. Visceral mass con- 

 tains ovary only without generative ducts, testis, or nerve ganglion. Develop- 



