ON FOSSIL CRUSTACEA. 55 



one would be led to presuppose the Trilobites possessed of organs of loco- 

 motion of a stronger texture than mere branchial frills. 



The objection raised by Drs. Dana and VerriU to the special case of ap- 

 pendages in the AsapJius assumed by Mr. Billings to possess ambulatory legs, 

 is that the said appendages were merely the semicalcified arches in the inte- 

 gument of the sternum to which the true appendages were attached. 



A comparison, which these gentlemen have themselves suggested, between 

 the abdomen of a Macruran Decapod and the TrUobite in question is the 

 best refutation of their own argument. 



The sternal arches in question are firmly united to each tergal piece at the 

 margin, not along the median ventral line. If, then, the supposed legs of the 

 Trilobite correspond to these semicalcified arches in the Macruran Decapod, 

 they might be expected to lie irregularly along the median line, but to unite 

 with the tergal pieces at the lateral border of each somite. In the fossil we 

 find just the contrary is the case ; for the organs in question occupy a definite 

 position on either side of a median line along the ventral surface, but diverge 

 widely from their corresponding tergal pieces at each lateral border, being 

 directed forward and outwards in a very similar position to that in which we 

 should expect legs {not sternal arches) to lie beneath the body-rings of a fos- 

 sil crustacean. The presence, however, of semicalcified sternal arches pre- 

 supposes the possession of stronger organs than mere foliaceous giU-feet ; 

 whilst the broad shield-shaped caudal lA&te suggests most strongly the posi- 

 tion of the branchiae. In the case of the Trenton Asaphus I shall be satis- 

 fied if it appears, from the arguments I have put forward, that they are most 

 prohahly legs — 'feeling assured that more evidence ought to be demanded be- 

 fore deciding on the systematic position of so large a group as the Trilobita 

 from only two specimens*. 



With regard to the embryology and development of the modern King- 

 Crab (Limulus polyphaimus), we must await the conclusions of Dr. Anton 

 Dohrn before deciding as to the affinities presented by its larval stages to 

 certain of the Trilobita, such relations being only in general external form. 

 Dr. Packard (Reports of the American Association for the Advancement of 

 Science, August 1870) remarks, " The whole embryo bears a very near resem- 

 blance to certain genera of Trilobites, as TrinucJeus, Asaphus, and others ;" 

 and he adds, " Previous to hatching it strikingly resembles Trinucleus and 

 other TrQobites, suggesting that the two groups, should, on embryonic and 

 structural grounds, be included in the same order, especially now that Mr. E. 

 Billings has demonstrated that Asaphus possessed eight pairs of 5-jointed 

 legs of uniform si^e." 



Such statements are apt to mislead unless we carefully compare the cha- 

 racters of each group. And first let me express a caiition against the too 

 hasty construction of a classification based upon larval characters alone. 



Larval characters are useful guide-posts in defining great groups, and also in 

 indicating affinities between great groups ; but the more we become acquainted 

 with larval forms the greater will be our tendency (if we attempt to base our 

 classification on their study) to merge groups together which we had before 

 held as distinct. 



have, as a matter of course, been considered as belonging to a much lower group than the 

 Isopoda, in which the normal number of somites is seven. Wliilst admitting the justice of 

 this conclusion, we do not think it affords any good ground for rejecting the proposition 

 that the Isopoda may be the direct lineal descendants of the Trilobita. 



* One in Canada and one in the British Museum, both of the same species. 



