118 THE NAUTILUS. 



However, it is present in Tetraplodon ambiguus (Sow.). T have ex- 

 amined a young specimen (male) from Santarem, collected by J. D. 

 Haseman, together with Hyria corrugata. As far as could be made 

 out it agrees also in the rest of the structure. 



Possibly also the shape of the palpi are important for the distinc- 

 tion of these two genera. 



Nevertheless, Tetraplodon agrees with Hyria in the essential char- 

 acters of the soft parts, and it has the same affinity with the African 

 Spatha, Hyria has, and it differs from Spalha in the same features 

 as Hyria^ namely, in the structure of the gills and the marsupium. 



I may add, that I have also examined, superficially, a species of 

 the genus Castalina, and several species of Diplodon from South 

 America. In all these, the palpi, the attachment of the inner gill, 

 the diaphragm, the anal opening and the location of the marsupium, 

 are of the same type, as in Hyria^ Tetraplodon, and Spatha, and the 

 structure of the marsupium is like that of Hyria and Tetraplodon. 

 The mantle edges are not connected in front of the branchial opening.* 



Thus it is clear that a number of South American Unionida, 

 subfam. Hyriince., group with the African Spatha, which stands, in 

 Simpson's system, in a different family, Mutelida, distinguished, 

 according to Simpson, by the taxodont hinge teeth, and by the 

 embryo, which is a lasidium and not a glochidium, as in the Unionidae. 



So far I have seen fully developed embryos only in a species of 

 Diplodon.^ Here they are glochidia, of peculiar shape, not almost 

 equilateral as in the North American forms, but distinctly inequi- 

 lateral, with a low anterior and a high posterior extremity, which 

 ends in a point at the postbasal angle.' 



The embryo of Spatha is apparently unknown, in fact, the larval 

 form, called lasidium, is known only in two species of the genus 

 Glaharis,*^ and it is only by inference, on account of the supposed 



1 According to v, Ihering (Zool. Anz. 14, '91, p. 477 ff.), this connection is 

 variable in Casialina, present or absent. 



2 Belonging to the delodontus group, and allied to lethiops (Lea) and wagntr- 

 ianus Simps.; it is from Rio Iguassu, Parana, Brazil. 



' Much more oblique than the glochidium of Diplodon pcculiuris (Lea), fig- 

 ured by Lea (Observ. XII, pi. 34, fig. 80). 



*See V. Ihering, Zool. Anz. 14, '91, pp. 480-482. 



