540 W- "• POLLARD, 



vides dichotomously corresponds to that branch which in Clarias does 

 not open to the exterior but joins its fellow of the opposite side. 



It is evident that sense organ 4 in Auchenaspis corresponds 

 functionally to the two sense organs 4 and 5 in Clarias one pore 

 having been lost in the former animal. Auchenaspis is the more 

 normal but it seems more probable that Clarias shows the more pri- 

 mitive arrangements for pores do not arise above solid dermal bone 

 such as occurs below pore 5 of Clarias, whereas it is known that 

 they and corresponding sense organs may disappear (Allis). It may 

 be here noted that pores 3 and 6 in Clarias also open above dermal 

 bone and not near sutures. Sense organ 6 is in both forms situated 

 in the noteworthy canal directed towards the crown of the head or 

 where the Ductus endolymphatici of the ear would open if present. 



The main canals, that is the squamosal portions of the lateral 

 line system agree in the two animals save that the sharp curves of 

 Clarias are not present in Auchenaspis while pore 8 and sense organ 

 9 of the latter are not present in the former. 



The course of the nerves is essentially the same in both. The 

 Ramus oticus is slightly more independent in Auchenaspis. 



A mandibular canal is present in both with several pores. The 

 exact number in Clarias was not observed. 



Callichthys and Trichomycterus. Comparing these two 

 South American forms, my specimens of both of which were young, the 

 resemblance is seen to be close. Callichthys was somewhat older 

 than Trichomycterus and does not show so many larval characteristics. 

 Organ 1 infraorbital of Trichomycterus is not represented in Callichthys 

 while below the eye two more sense organs are seen in the latter. 

 Sense organ 3 in the latter corresponds to sense organ 2 in the former. 

 In both the pore at the junction of the infra- and supraorbital systems 

 is present. The snout in Trichomycterus is more elongated than in 

 Callichthys the two in this respect showing a relation analogous to 

 that which obtains between Auchenaspis and Clarias. It does not 

 atiect the number of sense organs or pores. 



In the supraorbital line the sense organs and pores correspond 

 each to each and in both the first sense organ is supplied by a nerve 

 from the median side. 



The pores and sense organs also correspond exactly in the main 

 canal but the expansion between pores 6 and 7, in the suprascapular 

 bone, which is so marked in Callichthys is much less prominent in 

 Trichomycterus. No opercular canal is present in the latter. The 



