1916. No. 10. THE LOWER CAMBRIAN HOLMIA FAUNA. 89 



in Hohiiia and Kjerulfia. However this is just as little constant as in 

 Mcsonacidae, it being loosely attached in certain cases, e. g. Pamdoxidcs 

 ola}idicus Sjögren, the allied genus Ccntropleura^, and possibly Paradoxides 

 Hayivardi Raymond-, The above condition in these forms is also what 

 may be assumed for Callavia. 



The hy.postome attachment or rostrum is not found in Faradoxides 

 as a special portion of the shell; there is strong evidence, however, that 

 it has originally been developed as in the Mcsonacidae mentioned above, 

 but that the sutures have completely grown together and disappeared 

 (page 82)3, 



It must be admitted that the peculiarities here mentioned in connection 

 with the hypostome and the hypostome attachment might be interpreted as 

 indicating a direct genetic descent; they ma}^ however, also undoubtedly be 

 explained as indications of descent from common ancestral forms and converg- 

 ent evolution. 



We now pass over to the thoracic segments and their structure. 

 It is not our intention to devote much space to a consideration of the lack 

 of medial spines in Paradoxidae, although they are more or less markedly 

 developed in all genera of Mcsonacidae, which latter Walcott regards as 

 the stock form. On the other hand we must strongly emphasize the fact 

 that the pleural furrows in these two families are quite differently developed. 

 In Mesonacidac they are broad, the posterior limitations are indistinct, and 

 they run entirel}' parallel to the anterior margin of the thoracic segments. 

 In Paradoxides, on the contrary, the}' are more sharply defined, are nar- 

 rower, and have a marked oblique and receding course. 



This difference is so marked that with only one segment it is possible 

 to determine the family. The development of the pleural spines is some- 

 what uniform in the two families; it must be observed however, that in 

 Paradoxidae the 2nd segment is now and then more strongl}^ developed 

 than the others, whilst in Mcsonacidae as is known the 3rd segment shows 

 a similar stronger development in certain genera [Olenellns line). 



It may be doubtful, however, how great importance should be attached 

 to this difteience in the development of the pleural furrows. If we investi- 

 gate this phenomenon in other Cambrian trilobites it appears certain that 



^ K. A. Grönwall, Bornholms Paradoxideslag, 1902, pag. 123. See also his excellent 

 illustrations of the hypostome of Parad. Davidis Salter, pi. 2, fig. 5. 



2 According to Raymond the hypostome depicted as Par. Harlani Green, belongs to his 

 new species Haywardi. (Notes on the ontogeny of Paradoxides, Bull. Mus. Comp. 

 Zoolog}', Vol. LVIII, 1914I. 



3 See especially G. Holm, Om Olcnellns Kjendfi, 1888, page 14. 



