On the Rhizocephalan Genus Thompsonia, etc. 5 



forms of this genus which infest Alpheids, under the name of Thyla- 

 coplethus, has reverted to the theory of Giard, supposing that each of 

 the external sacs represents a Cypris larva which has fixed and meta- 

 morphosed in that identical position. The simple structure of the sac 

 is another primitive feature, and we are led to believe that the more 

 complex structure of Sacculina is due to progressive evolution from a 

 form like Thompsonia. This view has also been taken by the few 

 workers who have studied material of this Indo-Pacific form. The 

 conclusions which are arrived at in this paper combat decisively the 

 claim that the genus represents in any way the ancestor of the group 

 Rhizocephala. As is so often the case, a form which seems at first 

 sight to fill a gap between a group and its relations proves on closer 

 examination but to illustrate an extraordinary capacity for specialised 

 development in the group to which it belongs. But though the prob- 

 lem of Thompsonia thus changes with widening evidence, it is yet a 

 form of great importance to the student of comparative anatomy and 

 general biology. 



PREVIOUS WORK ON THOMPSONIA. 



In 1874 Kossmann (6) published a description of the Rhizocephalous 

 Crustacea collected by Semper in the East Indies. In this class he 

 placed a parasite, two examples of which were found on the walking 

 legs of Melia tessellata. He named this animal Thompsonia globosa, the 

 generic name being given in honour of that distinguished Irish natur- 

 alist who first recognised the Cirripede affinities of the Rhizocephala. 

 The new genus was characterized by its globular shape and small size, 

 long stalk with peculiarly thickened chitinous ring, absence of a mantle 

 opening, reduction of the visceral mass, and retention of the larvae 

 until the Cypris stage was reached. Only two examples of the parasite 

 occurred on this host, so the most striking feature of the genus, the 

 presence of large numbers of external sacs of equal development, is 

 not here recorded. 



In 1902 Coutiere (1) described a new type of Rhizocephalan parasitic 

 on various Alpheids from East Indian and Australian waters and re- 

 markable for the fact that each host carried under its abdomen, attached 

 to the appendages, a great number of parasites, which resembled the 

 eggs somewhat in volume and aspect. This parasite, for which he 

 proposed the name of Thylacoplethus (several species of the genus 

 were described) approaches Thompsonia in its form, but must, he says, 

 be separated generically, first and foremost because its host, the shrimp 

 Alpheus, is so different from the host of Thompsonia, which is the crab 

 Melia. Besides, Thylacoplethus is distinct as the first example of a 

 truly gregarious Rhizocephalan. I shall deal later with my own 

 reasons for setting aside these conclusions of Coutiere and treating 

 the name Thylacoplethus as a synonym of Thompsonia. 



